CHAPTER I
Defense Budgeting and Resource Allocation
DOES THE UNITED STATES spend too much on defense? Why does its current military budget, even excluding costs for Iraq and Afghanistan, exceed the Cold War average in real-dollar terms, and exceed the budgets of either China or Russia by a ratio of some five to one? If the budget is bloated, how can it be prudently scaled back? Regardless of whether it is excessive, how can proposals for new types of military capabilities be properly evaluated? Only by understanding the components of the defense budget can these questions be seriously addressed.
Before getting into more specific arguments, it is worth noting comments frequently made about America’s defense budget—usually by those wishing to make the defense budget seem high or low. Many who wish to defend the magnitude of Pentagon spending often point out that in recent decades its share of the nation’s economy is modest by historical standards. During the 1960s, national defense spending was typically 8 to 9 percent of gross domestic product (GDP); in the 1970s, it began at around 8 percent and declined to just under 5 percent of GDP; during the Reagan buildup of the 1980s, it reached 6 percent of GDP before declining somewhat as the Cold War ended. In the 1990s, it started at roughly 5 percent and wound up around 3 percent. During the first Bush term, the figure reached 4.0 percent by 2005 and stayed there through 2007; it grew towards 4.5 percent by 2009. Seen in this light, current levels (including war time supplemental budgets) seem relatively moderate, if hardly low.1
By contrast, those who criticize the Pentagon budget often note that it constitutes almost half of the aggregate global military spending (to be precise, 41 percent in 2006, according to the estimates of the International Institute for Strategic Studies).2 Or they note that estimated 2009 national security discretionary spending of some $670 billion exceeds the Cold War inflation-adjusted spending average of $450 billion (expressed in 2009 dollars, as are all costs in this chapter) by almost 50 percent once war costs are included (and exceeds the Cold War average modestly even without war costs). Or they note that it dwarfs the size of America’s diplomatic, foreign assistance, and homeland security spending levels (roughly $13 billion, $25 billion, and $44 billion, respectively, in 2009).3
These observations are all simultaneously true, and as such they are probably inconclusive in the aggregate. The U.S. defense budget is and will remain large relative to budgets of other countries, other federal agencies, and even other periods in American history. Yet at the same time, it is modest as a fraction of the nation’s economy at least in comparison with the Cold War era. As such, while informative at one level, these observations are of little ultimate utility in framing defense policy choices for the future. We must look deeper.
An initial conclusion of this book is that broad arguments about the adequacy or excess of American defense spending, based on overall observations about the size of the defense budget in a historical, international, or economic perspective, are analytically suspect. In my judgment, it is not possible to reach a definitive judgment about whether the United States overspends on defense through the methodologies of defense analysis or the science of studying warfare. And it is especially difficult to do so using arguments, analogies, and comparisons that invoke the overall size of the defense budget. Only by looking more carefully at how defense dollars are spent can we decide if the budget is excessive (or insufficient); the key is to try to identify missions that are not needed, or weapons modernization plans that are too fast and indiscriminate, or war plans that are excessively cautious and conservative. Again, to offer my own perspective, a number of proposed or ongoing defense programs are quite debatable, potentially unnecessary, and/or wasteful. Pentagon belt tightening is feasible in a number of areas, based on reasoned analytical assessments of America’s defense needs. (I am not focusing primarily on the costs of ongoing wars in Iraq and Afghanistan in this specific discussion; those operations are clearly debatable on their own merits. My focus here is on the underlying and enduring American defense establishment.) This book is designed in part to help people identify, study, and debate such possible cutbacks. Then again, because so many of the Pentagon’s costs are rising faster than the rate of inflation, economizing in a number of areas may do more to slow the growth of real defense spending than to allow actual cuts.
This chapter first explains different ways the defense budget can be categorized, broken down, and defined. It presents several tools for understanding the budget in greater detail—which is necessary for calculations that seek to understand the fiscal implications of any proposed changes to military force structure, weapons procurement plans, or military operations. One approach, developed largely by former Pentagon official and Brookings and MIT scholar William Kaufmann, is essentially “top-down”—it starts with the total defense budget, and then subdivides it into various pieces. Another approach, used in the executive branch, the Congressional Budget Office (CBO), and elsewhere, takes a “bottom-up” approach. Costs per unit are estimated directly, based on what the unit in question comprises in terms of people, equipment, and training time. Largely because Kaufmann’s method is more comprehensively and clearly presented, it forms the basis for my approach—but CBO’s observations are used as well, and the numbers presented are thus modified from Kaufmann’s original versions (as well as being updated for inflation and cost growth).
After presenting these methods, the chapter then turns to the question of military readiness. Readiness is a major strategic issue, a hot-button political subject, and also effectively the defense activity that consumes at least half the defense budget (since it includes spending on people, as well as their training and other operational necessities). Finally, addressing matters of frequent political and policy discussion, the chapter places U.S. defense spending in a global context, with a particular eye on how American spending stacks up against that of China.
UNDERSTANDING THE DEFENSE BUDGET:
BASIC CATEGORIES
In discussing the defense budget, it is critical to start with a clear use of terminology. Words can be very confusing in the field of defense analysis.
By official definitions, the U.S. national security budget does NOT capture all major government activities that influence American security, since it includes neither diplomacy, nor foreign assistance, nor Department of Homeland Security operations. It does, however, include the Department of Defense, the intelligence community, and the Department of Energy’s nuclear-weapons-related activities.
The national security budget is officially known as the 050 function in the federal budget. International affairs programs, including diplomacy and foreign assistance, are labeled as the 150 account; veterans’ benefits are found within the 700 function; and homeland security is distributed among a range of accounts.
It is important to know the distinction between budget authority and outlays. Outlays amount to actual spending—checks written by the Treasury and cashed by individuals or corporations. Budget authority by contrast is a new legal power to enter into contracts, granted by the Congress through appropriations bills. It can loosely be thought of as putting money into the Pentagon’s hands (or the Pentagon’s account at the U.S. Treasury), which is then gradually spent over the ensuing months and years as contracts are signed and products delivered. Authorized budgets slated for a given fiscal year may be spent very quickly, as with salaries for troops, or slowly, as with contracts for aircraft carriers or other large types of equipment (which take years to build and are fully paid for only when complete).4
A given year’s national security budget authority tends to exceed outlays when defense budgets are rising, and to fall below outlays when budgets are declining. (For example, if defense budgets have been going up—a given year’s budget authority might be $500 billion, but actual spending or outlays for that year might be $480 billion, for example—it is the latter figure that then enters into federal deficit calculations.) The reason is that outlays, as noted, show a lag effect because they reflect the effects of the previous year’s budgets, as well as the current year’s. There can also be differences, to the tune of several billion dollars a year, between discretionary budgets and overall budgets (the latter count mandatory spending as well). Mandatory accounts do not require annual budget action from the Congress (entitlements like Social Security and Medicare are the most important examples of mandatory spending in the federal budget). However, almost all defense spending is discretionary, meaning no funds are available in a given year unless explicitly appropriated for that year by law.
Unlike other federal agencies, the Department of Defense routinely provides Congress and the public with detailed information about its longer-term plans each time it submits a budget request. As part of its so-called future years defense program (FYDP), it projects roughly five years into the future. During the Rumsfeld tenure at the Pentagon, such information was less forthcoming than had been the norm, but even then the available information was considerable. The Department of Defense also publishes unclassified information on its major weapons programs through documents such as selected acquisition reports (SARs). These show total cost projections and planned numbers of equipment purchases for several dozen of the Pentagon’s largest acquisition programs over an even longer time horizon (extending as far out as the relevant acquisition program is expected to continue).
A brief word is in order about the way the Pentagon budget is constructed, which centers around the so-called PPBE process (for planning, programming, budgeting, and execution system). As explained by retired Colonel M. Thomas Davis, the Pentagon begins its efforts to create a budget proposal for Congress with reference to the National Security Strategy (which is itself coordinated and published by the National Security Council, generally under the signature of the President). The Joint Staff at the Pentagon then writes a National Military Strategy based on the National Security Strategy. This is used in the first, or planning, phase of the PPBE process to create documents known as Guidance for Development of the Force and Joint Programming Guidance (formerly done via the Defense Planning Guidance document), which prioritize certain military missions and begin to translate broad strategy into more concrete decisions.
However, it is in the second, or programming, phase where specifics really emerge. The services and defense agencies take the lead at drafting program objective memoranda (POMs) during this phase, subject to the all-important budget ceilings provided by the Office of Management and Budget (OMB). The POMs are then vetted and revised by the Office of the Secretary of Defense. Once so edited, they are used to produce the Future Years Defense Program (FYDP), as noted earlier, a half-decade-long budget plan. Even greater detail is provided about the first two years of that FYDP in the budget request submitted to Congress (however, Congress only appropriates funds for the first of the two years, once it sees the budget proposals and acts on them).5
At present, the Department of Energy’s share of national security or 050 spending is about $18 billion a year. Another $6 billion is spent here and there by other agencies, but the bulk of the money is for the Department of Defense (DoD).
As for the DoD, estimated spending in 2009 is roughly $650 billion. Viewed in terms of budget authority, the underlying “peacetime” budget for that year in the administration’s request was $515 billion. Another $70 billion was requested as an initial “supplemental” for the final months of calendar 2008 (that is, the initial months of fiscal year 2009, prior to the inauguration of the new president). Supplemental budget authority had totaled about $170 billion in 2007 and $195 billion in 2008 (again, in 2009 dollars).
The following section first explains the so-called peacetime budget, breaking it down by category and function, and then discusses supplemental budgets such as those presently used to fund the Iraq and Afghanistan wars.
Breakdowns of the Department of Defense Budget
The U.S. military, in its official budget documents, breaks down its overall budget several ways. Two basic methods show spending by appropriations “title” and by military service (neither includes Department of Energy [DoE] nuclear costs). The service-by-service approach merges Marine Corps costs within the Navy’s (since technically the Marine Corps is a separate service but is not a separate department). Many intelligence community costs are found within the Air Force budget when costs are subdivided this way, since it launches and operates many of the rockets and satellites used in intelligence as well as many of the aerial assets.
Another method, devised by former Secretary of Defense Robert McNamara, subdivides spending by what he called military “programs.” Rather than allocate the defense budget based on military service, or type of activity, he sought to use broad categories defined in terms of the overall functions they served. They include strategic nuclear capabilities, main combat forces, transportation assets, administrative and related support activities, National Guard and reserve forces, and intelligence, as well as several smaller areas of expenditure. This method is not completely accurate. Many military forces are usable for both nuclear and conventional operations, for example. Should they be viewed as strategic nuclear capabilities or main combat forces? And how to allocate expenditure for equipment first bought for active forces but later transferred to the reserves? Moreover, these categories are broad enough that, even if accurate, they may have only a modest bearing on informing policy choice. Nevertheless, they still give at least an order-of-magnitude sense about how different types of military objectives or main activities translate into costs.
Table 1.1, 1.2, and 1.3 reflect the Bush administration’s request for 2009, which is slightly different but not enormously different from what Congress approved (the following does not count any war costs). The budget authority figures include about $2.9 billion in so-called mandatory spending but are predominantly discretionary in nature.6
Available documentation, found at www.budget.mil and updated in fresh documents each February with the Pentagon’s budget request, provides a great deal of detail for those wishing it. For example, within the military personnel accounts, information can be found on the costs of officer pay versus enlisted pay, on salaries versus accrual financing for the future retirement of current troops, of active versus reserve compensation, and of travel and moving allowances, to name but a few subcategories. Within the procurement budgets, out of which most equipment purchases are funded, subcategories exist for aircraft, vehicles, ammunition, missiles, and other groupings of assets by military service.
TABLE 1.1
DoD 2009 Budget Authority Request by Title (Billions of Dollars)
Military Personnel |
128.9 |
Operations and Maintenance |
180.4 |
Procurement |
104.2 |
Research, Development, Testing, and Evaluation |
79.6 |
Military Construction and Family Housing |
24.4 |
Management Funds, Transfers, Receipts |
0.8 |
TOTAL |
518.3 |
Sources: Under Secretary of Defense (Comptroller), Military Personnel Programs (M-I), Operation and Maintenance Programs (O-I), Department of Defense Budget, Fiscal Year 2009 (Washington, D.C.: Department of Defense, February 2008), pp. 18, 20; Under Secretary of Defense (Comptroller), Construction Programs (C-I), Department of Defense Budget, Fiscal Year 2009 (Washington, D.C.: Department of Defense, February 2008), p. iv; Procurement Programs (P-I), Department of Defense Budget, Fiscal Year 2009 (Washington, D.C.: Department of Defense, February 2008), p. II; and RDT&E Programs (R-I), Department of Defense Budget, Fiscal Year 2009 (Washington, D.C.: Department of Defense, February 2008), p. II.
TABLE 1.2
DoD 2009 Budget Authority Request by Service (Billions of Dollars)
Army |
139.0 |
Navy |
149.0 |
Air Force |
143.7 |
DoD-wide |
86.6 |
TOTAL |
518.3 |
Source: Under Secretary of Defense (Comptroller) Tina W. Jonas, “Fiscal Year 2009 Budget Request: Summary Justification,” Department of Defense, Washington, D.C., February 4, 2008, p. 8.
TABLE 1.3
DoD 2009 Budget Request by Program (Total Obligational Authority, Billions of Dollars)
Strategic Forces |
9.9 |
General Purpose Forces |
201.9 |
Command, Control, Communications, Intelligence, and Space |
77.6 |
Mobility Forces |
13.5 |
Guard and Reserve Forces |
38.4 |
Research and Development |
52.8 |
Central Supply and Maintenance |
22.0 |
Training, Medical, and Other |
70.6 |
Administration |
18.8 |
Support of Other Nations |
2.2 |
Special Operations Forces |
9.0 |
Other |
0.1 |
TOTAL |
516.8 |
Sources: See Under Secretary of Defense (Comptroller), National Defense Budget Estimates for FY 2008, pp. 1–2, 81; Office of Management and Budget, Budget of the U.S. Government, Fiscal Year 2008: Historical Tables (Washington, D.C.: Government Printing Office, 2007), pp. 89, 164; and Allen Schick, The Federal Budget: Politics, Policy, Process (Washington, D.C.: Brookings, 2007), p. 57.
Note: Here, the figures add up to a slightly different total because what is presented is total obligational authority, not budget authority. The difference in these two concepts is quite small for our purposes, but it has to do with the possibility that some funds and obligations can carry over from one year to the next (or lapse or be eliminated before being obligated), creating a slight difference between budget authority and total obligational authority in any given year.
Another detail worth noting here concerns the distinction between discretionary budgets and overall total budgets. Discretionary funds must be appropriated each year by Congress. Overall budgets also include mandatory programs and spending, which do not require yearly attention (entitlements are the largest example of mandatory programs in the federal budget). Almost all military spending is discretionary. Mandatory accounts can be positive or negative since they can involve trust funds, user-fee programs, and the like. For example, in 2008 the administration’s request for all DoD funding was $643.7 billion, while the discretionary request was for $647.2 billion, meaning the mandatory funding request was “negative.”
Which category is most useful for understanding a given policy challenge or framing a given policy choice depends on the issue at hand. For example, imagine someone wishes to know if the country should move to a smaller but more mobile military. One notional way to create this might be to double the budget for U.S. mobility forces, using savings from a smaller combat force structure to fund the expanded transportation programs. McNamara’s categories shown in Table 1.3 would give some sense of how much of the combat force structure might have to be cut to make this possible. The same table could help one estimate how much deep cuts in nuclear programs might save. (In doing so, one would have to remember that DoE nuclear warhead costs, and many missile defense costs, are not captured in the strategic forces category used here. It is also important to know that, historically, nuclear-related costs were far higher than today, typically reaching $80 billion a year in 2009 dollars and perhaps even more by some estimates.)7 Or if the question was how a 5 percent across-the-board increase in military compensation would affect the defense budget, the information in Table 1.1 could come in handy (incidentally, civilian pay for DoD employees, which totals a bit more than $60 billion a year at present, is located within the Operation and Maintenance budget).8
Generally, however, more refined budgetary tools are typically needed for these and other purposes. The previously mentioned tools are informative but not that powerful analytically, nor that accurate. In subsequent sections, we will therefore develop finer tools of defense budget analysis to allow more detailed assessments. First, however, we shall examine the question of war costs and supplemental appropriations.
The Wartime Supplementals
Through fiscal year 2008, Congress had appropriated about $750 billion for military expenses in the broadly defined global war on terror, including $35 billion for Iraqi and Afghani security forces, plus another $40 billion for diplomatic activities and foreign aid for the broader effort. (By comparison, in 2009 dollars, the Korean War cost about $480 billion, Vietnam about $680 billion, and Desert Storm about $90 billion, with 90 percent of the latter costs paid by U.S. allies.) In rough terms, total funding for Iraq approached $600 billion by late calendar year 2008, Afghanistan $150 billion, and DoD homeland security efforts $50 billion.9
The preceding figures refer to the additional costs of various military missions. These are the costs above and beyond those already incorporated for the forces in question in the standard defense budget (such as regular salaries and training costs). These incremental costs per troop for current military operations have been very high—over $500,000 a year on average. This was more than twice what had been projected back in 2002. Even once costs for activities such as training Iraqi security forces are removed, costs per U.S. troop deployed have exceeded $400,000 a year.
This was not the first time the Pentagon failed to accurately predict the cost per troop in a major operation (note that this has nothing to do with how long the mission lasts nor how many troops are deployed, but rather the annual cost per troop deployed). DoD has been notorious for its failure to understand deployment costs accurately in the recent past. For example, in the Bosnia mission, initial estimates for the cost of deploying 20,000 troops to the region for a year were $1.5 billion to $2 billion, but actual costs were at least twice the upper bound (in other words, at least $200,000 per deployed troop per year).10 This is a good case study in how even sophisticated defense analysis tools are often crude, and thus how generalists armed only with fairly simple methodologies can often be nearly as accurate as Pentagon planners.
Why are these numbers so much higher, on a per-troop basis, than those from past wars or from earlier estimates for this very war? It is one thing to have uncertainty in projections of war costs based on not being sure how long a war will last or how hard the fighting will be. (For example, prior to Operation Desert Storm in 1991, the Congressional Budget Office [CBO] estimated that war costs could be roughly $45 billion to $140 billion, when translated into 2009 dollars; actual costs were about $95 billion in 2009 dollars, with just under 90 percent paid by friends and allies.)11 It is something else to be off by 100 percent or more when the duration and troop strength of a given mission are not in doubt. Numerous reasons help explain this. Military facilities have been developed to be high-tech, comfortable, and useful over a sustained period. Equipment has been worn down by intense operations and damaged by enemy action far more than predicted. New types of equipment to handle specific challenges of recent wars, notably the Mine Resistant Ambush Protected (MRAP) vehicles, have been developed and produced in large numbers. Contractors have been hired to support operations in very large numbers. And quite unabashedly, DoD has added a number of costs for activities not strictly related to the war—such as restructuring its Army brigades—in supplemental requests since about 2005.12 Accordingly, supplemental procurement funding averaged only about $10 billion annually through 2005, but rose to $27 billion in 2006 and $54 billion in 2007—with a whopping $74 billion requested for 2008 (again, expressed in 2009 dollars).13
UNDERSTANDING THE DEFENSE BUDGET:
KAUFMANN METHODS
While the Pentagon’s basic budget categories are helpful, those summarized earlier do not suffice when one seeks to analyze policy alternatives. The breakdowns by service and title fail to fully describe the missions to which defense dollars are devoted. Even the McNamara program elements fail to give any insight into the force structures designed to carry out those missions or the costs associated with them on a per-unit basis.
In recognition of this dilemma, longstanding Pentagon advisor and Brookings scholar, the late William Kaufmann, created two additional breakdowns of force structure, weapons purchases, and related Pentagon costs that can complement the McNamara method and in some cases provide more useful analytical tools. One approach subdivided costs by geographic region of the world, the other by main combat formations of each of the military services. The second is especially useful.
Implicit in the Kaufmann methodology, by the way, is the notion that to save money in the Department of Defense budget, one probably needs to cut real capabilities. That view is in contrast to the hopes of many that a savvier way of buying weapons, or more restrictions on “revolving door” practices between defense industry and the military services that have sometimes fostered corruption, or privatization of more defense support activities will lead to big savings without painful cuts in military muscle. While such savings are sometimes attainable, and while all these policy issues are worthy of attention in their own right to foster good governance, experience suggests that they are unlikely to yield big windfalls of savings that can then be redirected. Perhaps the best summary of how to think about such savings comes from a report written by former Congressional Budget Office Assistant Director and current Pentagon Comptroller Robert Hale. Aptly titled, “Promoting Efficiency in the Department of Defense: Keep Trying, But Be Realistic,” the report suggested that creative and effective defense reforms might save hundreds of millions or even a few billion a year if done well. This is real money, to be sure, but it is a far cry from the tens of billions believed achievable by some in the past. (For example, some Pentagon critics have noticed the 20 to 30 percent savings that resulted when DoD outsourced certain totally nonmilitary activities like paycheck processing and assumed that such monies could be found more generally throughout the budget. That has generally proven quite hard to achieve on a broader scale in the normal peacetime defense budget.)14
By contrast, there is something of a consensus among analysts that private contractors can save DoD money in overseas operations, and perhaps even more than 20 to 30 percent, with some estimates approaching 50 percent. Such savings are achievable partly due to lower per-person costs when contractors are able to hire lower-wage non-Americans, and partly due to the fact that contractors need not be paid when operations end (whereas comparable numbers of uniformed personnel would presumably be retained in the force structure even after an operation was over). However, this benefit must be weighed against the risk that contractors may not be quickly available, may not be willing or able to operate in dangerous settings, and may not be sufficiently well regulated and disciplined to be compatible with some missions requiring great care in the use of force. In addition, American contractors are generally not individually cheaper than government employees, according to a recent study by the Office of the Director of National Intelligence, and in fact may be at least 50 percent more expensive per year.15 On balance, contractors do have an important role for the U.S. military. But this fact is already being exploited, since in Iraq the United States is using one contractor for every uniformed man or woman—2.5 times the highest rate of any previous large conflict, relative to the number of deployed troops.16
Kaufmann’s Geographic Breakdown of American Defense Costs
Kaufmann’s geographic approach subdivided America’s military missions by overseas theater—Europe, the Atlantic sea lanes, the Far East, the Persian Gulf, Latin America, and Africa. Most combat formations were assigned accordingly, though some were attributed to American territorial defense, or to missions such as nuclear deterrence and intelligence. Kaufmann created a taxonomy that, like McNamara’s, had about ten main categories. Again as with McNamara’s, the allocations were constructed such that their sum total equaled the aggregate defense budget.
The basic logic of Kaufmann’s allocation scheme is simple and appealing. And if accurate, it can allow for a clear method of assessing the fiscal implications of various American security commitments, such as protection of Persian Gulf oil or of key allies like Japan, Korea, or the countries of Western Europe. But is it right?
While not lacking merit, Kaufmann’s geographically oriented defense budget breakdown is probably the most controversial of the major methods considered here. Some military assets are indeed designed primarily for one type of operation in just one or two places—such as frigates designed to protect ships as they traversed the Atlantic and Pacific during the Cold War, or aircraft carriers and Marine Expeditionary Units routinely deployed to specific regions like the Persian Gulf and Western Pacific. In these cases, it is not hard to apportion costs on a regional basis, at least roughly speaking. (For ships and ship-based Marines, examining their homeports and typical deployment patterns can further help determine if they should be assigned primarily an Atlantic/Mediterranean or a Pacific/Indian Ocean orientation.) During the Cold War, Europe was the primary intended heavy combat theater for air-ground operations, making it logical to attribute the costs of most Army and tactical Air Force units to that region (at least until the Vietnam War upset the assumptions, and the calculus). After the Cold War, the focus for warfighting planning shifted to the Persian Gulf and Korean peninsula, as Pentagon documents such as the bottom-up review explicitly noted. So this method has a certain logic to it.
Kaufmann’s last breakdown was done in 1992, when the Pentagon still worried about a possible Russian resurgence and the resulting hypothetical danger to countries like the Baltic states. As such, he estimated that a substantial fraction of the overall defense budget was for the defense of Europe. It is not clear if he would reach the same conclusion today, though in the aftermath of Russia’s August 2008 invasion of Georgia, arguably one could. (His amounts are shown in Table 1.4, displayed as percents of the overall Department of Defense budget. Only the budgets for nuclear forces as well as national communications and intelligence are not divvied up by region.) The share of the defense budget focused on the Western Pacific, with an eye not only towards North Korea but also the rise of China, and the more general ascendance of that part of the world in economic and strategic terms, conceivably might change if Kaufmann were to redo his estimates today. Kaufmann’s figures were quite modest; a larger estimate might be appropriate now. In addition, the United States has also been more active in Africa in the years since Kaufmann last wrote, beginning with the ill-fated Somalia mission but also including refugee relief in Central Africa, counterterrorism cooperation with countries in the Sahel region, and now the creation of Africa Command (AFRICOM). But Kaufmann’s numbers nonetheless do reflect his initial assumptions about the state of the post–Cold War and post-Soviet world, and as such still have some relevance today.
TABLE 1.4
Kaufmann Estimates of DoD’s Spending by Geographic Region under the “Base Force” of the First Bush Administration (in Percent)
Strategic Nuclear Deterrence |
15 |
Tactical Nuclear Deterrence |
1 |
National Intelligence and Communications |
6 |
Northern Norway/Europe |
5 |
Central Europe |
29 |
Mediterranean |
2 |
Atlantic Sea Lanes |
7 |
Pacific Sea Lanes |
5 |
Middle East and Persian Gulf |
20 |
South Korea |
6 |
Panama and Caribbean |
1 |
United States |
3 |
Source: William W. Kaufmann, Assessing the Base Force: How Much Is Too Much? (Washington, D.C.: Brookings, 1992), p. 3.
The resulting budget tools Kaufmann created are most useful when trying to assess how much the country spends defending various overseas interests and allies—and how much it might save if it reduced certain commitments (or how much it might add if it increased commitments). This framework can help in debates over allied military burdensharing, in discussions of how much the United States spends defending Persian Gulf oil, and so forth.
However, it would be a mistake to take this framework too literally, because it is not really how the Pentagon plans its forces or budgets. It is very rare to have a combat formation that could only be usable in one part of the world. To be sure, some headquarters capabilities, some planning staffs, and some intelligence assets are devoted to a specific region. Also, the occasional combat unit, such as the Army’s 2nd infantry division in Korea, is sometimes very closely associated with one region. But most American combat forces are flexible, as they must be. The United States has too many global allies and interests to create, in effect, separate force structures to defend each. The cost of doing so would be prohibitive.
Most American forces are based in the United States and are deployable to whatever region national command authorities might need to send them. Even formations thought of as devoted primarily to one place may, when a crisis erupts, wind up going elsewhere. The Army drew down large numbers of European-focused forces to fight in Vietnam (not to mention Desert Storm in 1991), and one of two brigades previously based in Korea was sent to Central Command’s operations in the Iraq/Afghanistan theater in recent years (after which it returned to the United States, not Korea). Speaking of Afghanistan, it is a place for which no combat units had been designated prior to 2001. Looking to hypothetical scenarios in the future, there are no dedicated American forces for addressing instability in Pakistan, peacekeeping in Kashmir, humanitarian relief in Africa or South or Southeast Asia, or a range of other possibilities. Yet capacity must be retained for addressing one or more such scenarios at a time, even if each is individually unlikely to occur.
As such, Kaufmann’s framework, while useful, is more notional than precise. The methodology by which he makes his regional cost estimates is also not explained in any detail in his writings. Kaufmann’s personal reputation for rigor and great knowledge in the field make it likely that his estimates are as reasonable as any other—given the inherent limitations of tackling the problem in this basic way. But they are not easily reproducible or confirmable.
A corollary of this observation is that, even if a given overseas interest of the United States disappeared, or was deemed no longer requiring or meriting American military protection, the resulting implied decline in the defense budget would generally be less than Kaufmann’s method suggests. This is because some of the forces he allocated to a given region are also important for other regions—if not in a primary role, then at least as a strategic reserve.
Costs of Individual Combat Units
An additional methodology developed by Kaufmann breaks down the overall defense budget according to the main combat force structure—Army and Marine Corps divisions (active and reserve), special operations forces, Air Force and Marine Corps tactical air wings, Navy carrier battle groups and amphibious ships and sea control forces, airlift, sealift, prepositioning, strategic nuclear forces including submarines, bombers, land-based ICBMs, missile and air defenses, early-warning assets, and national intelligence and communications systems. Again, as with his geographic allocation scheme, Kaufmann’s goal was to create a meaningful yet simple tool, so he sought to avoid excessive complexity. His approach was to assign all DoD costs to these categories so they added up to exactly the overall size of the defense budget.
Then Kaufmann went further, making his approach the most analytically useful of all those considered in this chapter to date. Specifically, he divided the cost for a given category of combat capability by the number of units within that category to provide a cost per unit.
In fact, dozens of forces are designed to support combat formations, especially in the Army with its multiplicity of units and subunits focused on various aspects of logistics, intelligence, administration, air defense, other specialized weaponry, and the like. Kaufmann’s logic, however, was that in the end these capabilities exist only to support combat formations. As such, associating a certain proportionate number of support capabilities with each main combat formation, and allocating costs accordingly, is a reasonable way to create a good framework for policy choices. It allows a rough answer to the question of how much the country would save (or expend) if, for example, we cut (or add) an Army division or Air Force wing to the military force structure.
This approach of Kaufmann’s is extremely useful. And it does a much better job of capturing the real cost implications of many policy alternatives than do most official Pentagon numbers. For example, if the Department of Defense were asked how much it could save by cutting an Army division out of its force structure, it might give (and often has given) a very modest figure. Army divisions only have about 16,000 to 18,000 soldiers in them in general, and in making an estimate of savings, the Army would probably focus on personnel and operating costs (not equipment), since the former are often the only guaranteed expenses for a given military formation in a given year. However, at least 40,000 people are actually associated with a given Army division once support costs are included, and over time the Army will save a certain amount in reduced equipment expenditures when reducing force structure (even if savings in a given year cannot be specified exactly). So Kaufmann’s method would often show a cost about five times greater for a unit such as an Army formation—and there is little doubt Kaufmann’s estimate would be more accurate in general, at least in indicating potential savings in the medium to long term.
That said, the Kaufmann approach has its limitations. It is not entirely accurate. Not all defense costs can be proportionately allocated to main combat formations. Some military costs such as the research and development of new technology, strategic intelligence operations, and combat headquarters operations are to a large extent independent of the size of the force structure. Thus, cutting the number of combat units does not reduce the former costs proportionately, if at all. Yet Kaufmann’s method would often allocate such costs proportionately to combat force structure, implying greater savings than reality would allow. As for the costs of new weaponry, reducing purchases by a given percentage (as implied by cuts to force structure) generally would not allow fully commensurate savings, since economies of scale would be lost as fewer weapons were purchased.17 This effect can be mitigated by encouraging defense companies to consolidate as defense budgets decline, among other means. But that process takes time, and at some point the benefits to the Pentagon and taxpayer of having multiple firms compete for the DoD’s business can be partially lost. In other words, reducing the total purchase of a given type of weapon by say 10 percent does not typically produce savings as great as 10 percent, yet Kaufmann assumes it would.
Kaufmann’s figures may overestimate actual savings from any force structure cuts by 20 to 30 percent, roughly speaking. To see why, we can examine Army force structure as one example. It has historically been divided into three relatively equal groups—combat units, deployable support units, and nondeployable or institutional units. Most support units can be cut proportionately when main combat forces are reduced, but the link between combat forces and institutional capabilities is much less direct.18 So Kaufmann assumes comparable savings in all three types of costs, when in fact major savings are likely to prove possible only in two of the three. Despite the limitations of Kaufmann’s methods, they are typically the most accurate way to get back-of-the-envelope estimates.
In his 1992 book, Kaufmann made the following estimates of the average annual costs of different types of main combat units. These were projections for what those costs would be in the mid-to-late 1990s (expressed here in billions of constant 2009 dollars; Kaufmann had used 1993 dollars, and I multiplied by 1.53 to make the conversion). They include yearly personnel and operating costs, as well as an annualized estimate of how much investment a given unit requires to be outfitted with weaponry and other equipment. (In another part of the book, Kaufmann estimates differences in cost from one type of division or air wing to another, but he himself works with the average numbers shown in Table 1.5 for comparative purposes.)19
Kaufmann’s approach was to multiply each of these unit costs by the number of that type of combat unit in the force structure, add $2.8 billion for special forces, as well as $30.5 billion for national intelligence and communications, and $60.8 billion for nuclear forces ($60.2 billion of that being for strategic forces, just $0.6 billion for tactical forces)—and wind up with the total defense budget exactly. In other words, he assumed that all other types of assets and units in the U.S. military force structure could be associated proportionately and directly with main combat forces, as noted earlier. This assumption is incorrect in a technical sense, yet is hugely useful as an approximation, if one’s main goal is to have a rough gauge of how force structure translates into costs.
Although the preceding figures have all been converted into 2009 dollars, one could argue that they are not really quite ready for use in the present day.20 In fact, defense costs typically rise substantially faster than inflation (as discussed later in the section on the Congressional Budget Office’s methodology). This is largely due to the growing complexity of military technology, which accounts for most cost growth from generation to generation of equipment such as fighter aircraft.21 These considerations would suggest that the preceding figures should all be larger than shown, by as much as 50 percent or so—necessitating one more round of modifications of these numbers to make them usable today.22
However, I have chosen not to modify Kaufmann’s numbers in this way. The main reason is that as noted, Kaufmann’s cost figures back in 1992 actually overestimated the savings that could be obtained by cutting unit x or unit y (and overestimated as well the costs of adding force structure). Kaufmann assumed implicitly that research and development costs could be cut proportionately when force structure is reduced. He did the same for supporting capabilities—reconnaissance units, electronic warfare units, refueling aircraft, air defense and artillery units, and so on, not to mention headquarters and administrative/managerial oversight, and the service’s educational institutions, health care systems, and the like. By his logic, such capabilities would not exist absent combat forces; they exist only to support combat forces, and if combat units are reduced that should imply commensurate reductions in support capabilities. Kaufmann’s method helps focus the mind on the potential for overall savings if efficiencies are pursued throughout the Department of Defense when main combat force structure is reduced, but they are nonetheless too high, because certain activities like research and development and central oversight cannot be reduced in lockstep with changes in force structure. By my best estimate, as noted 20 to 30 percent of the costs Kaufmann associated with a given unit would be very hard to pare back in the event of marginal changes to the force structure.
TABLE 1.5
Kaufmann Estimates for Average Annual Costs of Key Combat Formations (Converted from 1993 to 2009 Dollars, Billions)
Active Army Division |
5.2 |
Active Marine Division |
4.0 |
Reserve Army Division |
1.0 |
Reserve Marine Division |
0.9 |
Active Army Brigade Combat Team |
1.2 |
Reserve Army Brigade Combat Team |
0.2 |
Active Marine Tactical Air Wing |
3.5 |
Active Air Force Tactical Air Wing |
3.0 |
Reserve Marine Tactical Air Wing |
0.9 |
Reserve Air Force Tactical Air Wing |
0.60 |
Navy Carrier Battle Group |
6.4 |
Amphibious Ship |
0.15 |
Sea Control Ship |
0.2 |
Airlift Aircraft |
0.02 |
Sealift or Prepositioning Ship |
0.02 |
Source: William W. Kaufmann, Assessing the Base Force: How Much Is Too Much? (Washington, D.C.: Brookings, 1992), pp. 70, 74, 82.
Note: I have taken the liberty of modifying Kaufmann’s estimate for an active Marine Corps division from $2.6 billion to $4.0 billion because this is the one estimate of his that seems somewhat incongruous with other data and estimates in this chart. I have also taken the liberty of estimating how much a brigade combat team—increasingly the standard unit of organization and measure for the U.S. Army—would cost.
On balance, these imprecisions tend to balance and cancel out each other. As such, Kaufmann’s figures are probably the most reasonable and accessible starting points available, once adjusted, as I have explained earlier. Once all is said and done, they are probably within 10 to 20 percent of more accurate (but generally elusive) estimates in most cases.
UNDERSTANDING THE BUDGET: CBO
The Congressional Budget Office, staffed with several dozen individuals specializing in defense budgeting yet generally writing unclassified documents for a wide audience, has considerable expertise in the field, combined with the goals of achieving simplicity and clarity in its analysis. That combination makes it especially useful in this discussion.
CBO has generally been reluctant to provide comprehensive information on the costs of main combat units. So Kaufmann’s numbers, as adjusted in the previous pages, are perhaps our best guide here. But CBO has offered cost estimates over the years for a number of weapons systems. In Table 1.6, costs are expressed in millions of 2009 dollars and equipment is assumed to have a lifetime of thirty years (except fighter and attack aircraft, which are assumed to have a lifetime of twenty years, and Virginia-class subs, which are assumed to last thirty-three years).23
CBO has also produced a different type of overall defense budget analysis regularly over the years. Looking at the whole DoD budget, it projects the medium-term costs of the Pentagon’s plans. In these studies, CBO analyzes information more by title—operations and maintenance, military personnel, military procurement, research and development and testing and evaluation—rather than by specific combat unit.
To carry out these projections, CBO takes the existing budget. It then adjusts it for several factors: planned changes to the force structure, known policy changes such as pay raises for personnel or purchases of new weapons, and Pentagon plans for reconfiguring bases or support organizations. It also offers its own best estimates of what is politely termed “cost growth”—the degree to which actual costs will exceed predicted ones. This is a major concern in many types of Pentagon budgetary accounts. For example, over recent decades, operations and maintenance costs on a per-troop basis have grown about 3 percent annually in real terms—even when they were expected to hold roughly steady. The rate of growth varies, but it is always greater than inflation, and it has a fairly steady trajectory to it. In recent decades, health care and environmental cleanup costs associated with base closures have been among the largest drivers of increased operations and maintenance (O&M) expenditures, with the former growing from about $15 billion a year to almost twice that from roughly 1988 through 2003, and now reaching $40 billion. Aging weaponry, by contrast, has apparently had only a modest impact to date.24 While it is not a given that such cost-growth trends will continue in the future, absent a radical change in management methods they probably will.
TABLE 1.6
CBO Cost Estimates (Millions of 2009 Dollars)
Sources: David A. Fulghum and Amy Butler, “Passing the Buck,” Aviation Week and Space Technology, February 11, 2008, pp. 24–25; Caitlin Harrington, “Congress Warned Over Additional Raptor Buy,” Jane’s Defence Weekly, February 13, 2008, p. 14; Raymond Hall, David Mosher, and Michael O’Hanlon, The START Treaty and Beyond (Washington, D.C.: Congressional Budget Office, 1991), pp. 139–40; Congressional Budget Office, “Total Quantities and Unit Procurement Cost Tables, 1974–1995,” Congressional Budget Office, Washington, D.C., 1995, pp. A-17 through A-22; Lane Pierrot, A Look at Tomorrow’s Tactical Air Forces (Washington, D.C.: Congressional Budget Office, 1997), pp. 3–5; and Eric J. Labs, Transforming the Navy’s Surface Combatant Force (Washington, D.C.: Congressional Budget Office, 2003), p. 11.
Note: Most numbers here are limited to two significant figures to avoid implying greater precision than available data warrant.
For weaponry, CBO makes use of detailed information about the normal growth of equipment costs as a given system moves through development into production. A detailed literature is available on the subject, created by defense research organizations such as the Institute for Defense Analyses, RAND, and others. Room for judgment always exists about the degree to which future weapons will display different cost-growth patterns than past ones, meaning debate often arises about CBO’s numbers. But to first approximation, assuming cost growth in the future like that which has occurred before is usually a more accurate method of predicting future costs than relying on the DoD’s generally optimistic official figures.
Weapons typically cost more to acquire than the Pentagon initially anticipates. For example, space vehicles as well as Army ground combat vehicles both tend to cost 70 percent more to develop, and ground vehicles also typically cost 70 percent more to produce, than initially estimated. On the other end of the spectrum, ships typically cost about 16 percent more to develop than first projected, and 11 percent more to produce (after adjusting for any changes in numbers of weapons bought). Using such information, CBO can crudely estimate cost growth for each weapon based on where it stands in its overall development and production cycle. On average, weapons cost around 20 to 50 percent more to acquire than initially forecast.25
CBO also has weapons-specific models, most notably for fighter aircraft, that predict cost based on weapons performance characteristics. For fighter aircraft, cost estimates are based on aircraft weight, maximum speed, the amount of advanced materials in the airframe, the number of contractors involved in the effort, the technological maturity of software, the maximum airflow through the engines, engine thrust-to-weight ratios, specific fuel consumption, and avionics. The basic concept is to look at previous aircraft, take their performance characteristics for all these key attributes, run a regression to create a mathematical formula, and then plug in the expected performance features of the new plane to calculate expected costs.26
In its costing methodology, CBO also needs to find a way to estimate procurement costs for small equipment. Cumulatively, such weaponry constitutes roughly 40 percent of all DoD investment spending—that is, the sum of procurement plus research, development, testing, and evaluation.27
But documentation for such smaller equipment is not detailed enough to allow CBO to do a system-by-system assessment as it does for large weaponry. As such, once it has projected cost growth for large weapons using detailed Pentagon documentation (including selected acquisition reports), it then makes an estimate for minor systems on the assumption that they will continue to account for 40 percent of total procurement spending.
CBO often underscores the uncertainty in budget estimates by creating lower and upper bounds on its projections. The former are typically based on Pentagon expectations about costs, the latter on models and historical experience. For example, CBO recently calculated a higher bound for long-term DoD budgets that was about 13 percent higher than its lower-bound estimate. The average over the next two decades for the higher bound would be roughly $605 billion, versus $535 billion a year, expressed in constant 2009 dollars. These figures do not count major war costs or Department of Energy nuclear weapons expenditures. Of the $70 billion in possible cost growth, about $40 billion would be in the operations and support accounts, the remaining $30 billion in the investment accounts.28
MILITARY READINESS
Most defense dollars go to one of two main activities in the United States: preparing forces for combat and other missions over the longer term, through research and development and procurement of weaponry as well as other efforts, and ensuring the near-term combat readiness of forces for missions in the here and now. In other words, the former activities are primarily (though not exclusively) about investment in hardware, the latter more about people and training. The former are addressed in the chapter of this book on military technology; the latter here.
Military readiness refers to prompt and immediate response capability for plausible missions. It is defined by the Joint Chiefs of Staff as the ability of the armed forces to deploy quickly and perform initially in any military contingencies as they were designed to do.29
The question of military readiness is in large measure a budgetary question, since many readiness gauges can be substantially influenced by how defense resources are allocated. It is also a critically important matter in many national security debates and in many American political debates as well.
Indeed, readiness has often been a political football. In the 1970s, America’s military was alleged—in large part correctly—to have gone “hollow.” A substantial force structure existed, but it was not particularly strong within, and it did not hold up very well when called upon to perform. During the mid-to-late 1990s, similar allegations about the U.S. military were made as the armed forces were downsized in the aftermath of the Cold War.30 In the first decade of the twenty-first century, as the Iraq and Afghanistan missions have continued relentlessly, many have spoken of a U.S. Army that has become broken.
What are these criticisms about? What do terms like “hollow” and “broken” even mean in the context of military preparedness?
It is important to differentiate military readiness from other important criteria for assessing military preparedness. In fact, readiness for assigned missions is only one concern that strategists and policymakers must emphasize.
Policymakers also need to determine if they are worrying about the right possible missions. Since many wars historically come as surprises to countries fighting them, this is a very difficult and crucially important task. Being ready for a war or other important mission that does not happen is no great solace if a country’s armed forces prove unready for what they are ultimately asked to do.
Planners also need to worry about future threats. (This is sometimes called “long-term readiness,” though that is a confusing use of terminology best avoided in favor of a better term like investment or modernization.) Long-term planning can compete with the near-term ability to carry out key functions—or readiness. Emphasizing the latter requires lots of money and time for training, for spare parts and fuel and ammunition, and for focus on threats or military missions that are already recognized as important. Emphasizing the longer-term, by contrast, tends to require more money and time devoted to research and development, to professional military education, and to experimentation with new warfighting concepts and technologies.31 Dollars spent on R&D cannot be spent on fuel or ammunition; officers’ time devoted to running future-oriented exercises and experiments is not available for drilling forces for near-term missions; units trying out new concepts are not as able to practice on more immediate tasks or potential tasks. So it is important to think in terms of tradeoffs.
Indeed, readiness for one mission can also compete with readiness for another. The Balkans wars were seen as distractions from “real” warfighting priorities and readiness requirements in the 1990s, since formal war plans were more focused on Northeast Asia and the Persian Gulf. The Afghanistan war was not anticipated or planned very well before it occurred, in part because of the preoccupation with what were considered more immediate threats requiring that readiness efforts be focused on them. The U.S. military prepared much better for the invasion phase of the Iraq War than for the post-invasion phase, suggesting that readiness considerations had emphasized classic invasion scenarios at the expense of counterinsurgency and stabilization missions. Also, in the last few years the ledger has shifted, with the constant preparation for ongoing operations in Iraq and Afghanistan dominating the time and attention of soldiers and Marines to the detriment of training for other missions.
Even once one defines readiness fairly narrowly and leaves broader strategic issues for separate debate and analysis, challenges remain in assessing it. The military is always short certain capabilities as defined by its doctrine and its process for determining military “requirements.” How much does one shortfall matter, in a given area of personnel or equipment or training, when other capabilities and other resources are robust? Just as with the debate over metrics in Iraq, it is always possible to find readiness metrics that make a predetermined political case—that seem to suggest that the military is either in fine shape or is falling apart. It is for such reasons that Richard Betts, the Columbia University professor, subtitled one chapter in his seminal book on the subject of readiness “lies, damn lies, and readiness statistics.”32
It is not particularly helpful when the Pentagon limits its public discussion of readiness metrics to a few categories that it selects, especially when it fails to present a historically significant period of time to provide perspective on these categories. For example, in its 2009 budget request, the Pentagon showed data for just six categories of readiness, and in each case only showed two years’ worth of data.33
Yet we must not throw up our arms in frustration. It is possible to think systematically enough about readiness to shed at least some light on the question of how well prepared the American military (or another military) may be for the likely missions it is designed to conduct. Otherwise, we are left to resorting to the selective use of statistics and anecdotes, which is not an effective way of determining how to allocate military resources.
The most dependable way to evaluate readiness is to fight real enemies, since war is always a major learning experience, and a great judge of capability. However, this is not a reliable or desirable way to assess readiness. The downsides are obvious—the possibility of suffering a major defeat before learning what went wrong, not to mention the inherent costs in blood and treasure. However, it is still true that the United States employs its military often enough that real operations are significant sources of information and important barometers of preparedness.
Consider just a few examples from the American experience of the last three decades. In 1980, the United States failed to rescue hostages in Iran, and lost eight servicemen when an accident during refueling in the Iranian desert led to the cancellation of the mission. Poor coordination among the different military services involved and a lack of realistic training may have contributed to the tragedy. In Lebanon in 1983, the U.S. military learned about the importance of force protection and the difficulty of maintaining neutrality during a peacekeeping mission when the Marine barracks were bombed. In short, it found out it was not ready to protect its forces against the types of threats that presented themselves on that battlefield (which, of course, foreshadowed the types of suicide truck bombings that became tragically commonplace in the ensuing quarter century). The Kosovo War in 1999 was impressive in many ways, and debunked the notion that the U.S. military had somehow been rendered hollow by its post–Cold War downsizing. But it also revealed how immobile the Army was when it tried to send a modest number of Apache helicopters to nearby Albania. The Iraq War required major learning—or perhaps the relearning of past lessons since forgotten—about counterinsurgency operations, with the U.S. military more impressive from 2007 on than in the earlier years of the war.34
More happily, in the 1986 bombing of Libya, as well as the 1989 invasion of Panama, Ronald Reagan’s military buildup was at least partially vindicated as U.S. forces (and equipment) performed impressively. In the Panama case, improvements in military command and control instituted in the 1986 Goldwater–Nichols act helped also ensure a well-directed mission.35
But clearly, waiting for the next war to evaluate a new readiness initiative or take stock of what traditional efforts have accomplished is not a preferred, or reliable, way to evaluate readiness. As such, other more mundane methods must be employed. They typically begin with dividing the subject into three broad categories: personnel, training, and equipment. Within each of these three umbrella groupings, one can then break things down further by looking at the different subcategories of troops and various types of military specialists, assessing the capabilities of each key subunit. One can do similar things when evaluating training for various units, as well as the availability and condition of equipment—not only in regards to major vehicles but also spare parts and ammunition.
Three challenges stand out when performing such accounting tasks, which are in and of themselves generally straightforward (if somewhat tedious and labor-intensive). First, has the military established proper benchmarks for readiness within each area? If standards for training or personnel are too low, for example, reaching those standards in a given month or year may not ensure true readiness. The error can work in the other direction as well. If a certain amount of ammunition is deemed necessary by the Pentagon, any shortfall relative to that requirement then becomes, or at least appears to be, a readiness problem even when it may not be. A case in point was the U.S. cruise missile inventory several years ago—consisting of almost ten times the number of cruise missiles used in Operation Desert Storm, yet somewhat below what the DoD “requirement” specified.36 A resulting political storm resulted over the alleged failure of the Clinton administration to be properly ready for combat, when in fact the armed forces faced a shortfall only relative to what was perhaps an overly ambitious goal.
Second, and in some ways even harder, how does one determine the relative military importance of any given shortfall? If nine out of ten readiness measures for a given unit are fine, but the last involves a mission-critical resource (such as ammunition for guns), a severe shortfall in one area of readiness could trump the adequacy of other preparations. This type of situation can lead to failing readiness grades for a whole division or air wing (often called the “C Rating”) when that unit has only one specific problem. Sometimes that is appropriate, but quite often it is not, because shortages of one type of capability can often be redressed by substituting it with another.
Relatedly, in the late 1990s, two entire Army divisions were deemed “unready” because they had each sent one brigade to the Balkans. In fact, each still had two brigades that were available for other missions.37
Even more strangely, consider a case where a unit is supposed to acquire a new type of fighter by a given date. If the deliveries of that fighter are slower than expected, the reported readiness of the unit can actually decline even while in the process of getting better equipment. That happens, for example, when the formal readiness accounting standard changes faster than new equipment can be delivered. There is also subjectivity in some of the integrated readiness measures that makes them vulnerable to a certain political influence at times. It might be preferred to emphasize how good readiness is (for those wanting to show a “can do” attitude and document their service’s good use of the defense resources it has been given). Alternatively, a given military service or official may emphasize how bad readiness is, relative to some standard (for those wishing additional resources, for example).38
The best antidotes to these problems are to present data in a relatively raw form, use consistent standards over time in evaluating it, and avoid sweeping generalizations about “passing” and “failing” readiness grades. Generally, it is best to show statistical trendlines, and use qualitative judgments to supplement these numbers rather than replace them.
A third problem for an outside specialist when assessing readiness is that not all information is easily available. Some is quite appropriately classified; a key shortage in a key type of equipment is not the sort of thing to broadcast (especially if those within the system recognize the problem and address it without outside prodding). Other information that is less sensitive should be shared; it should enter into political debates about properly resourcing the defense budget and wisely allocating funds among competing accounts.
Consider several examples from each area of readiness in 2008. My purpose in providing all the details is not to imply that today’s readiness problems are unique or monumental (though some are serious, it is true). Rather, the goal is to use 2008 as a detailed case study to show how different indicators can provide different suggestions about the severity of a readiness problem, necessitating some effort to integrate the indices when seeking an overall evaluation.
Personnel
Key measures of personnel readiness include the experience and aptitude of typical troops, the availability of individuals with critical specialized skills, and the ability of the military to recruit new members as well as retain those already in the service. Many of these can be measured. Admittedly, some key readiness metrics, such as the aptitude and creativity of flag officers, cannot be easily understood by peacetime gauges of preparation. However, data and analysis can help assess most types of readiness trends.
In recent times, there has been considerable concern about a lowering of personnel standards. It is worth reflecting on this situation in some detail as an illustration of the types of considerations that arise in such debates. One general rule is that there are always strains on the military, shortfalls in its equipment, skill sets lacking among its people, and specific weaknesses in near-term combat training. However, sometimes these weaknesses add up to a serious problem and sometimes they do not. Hence, each individual readiness statistic or measure needs to be placed in perspective.
Consider several trends in U.S. military personnel that arose during the George W. Bush administration. The military accepted more recruits with general equivalency degrees rather than high school diplomas; it enlisted a higher percentage of applicants scoring very low on its aptitude tests; and it also took on more individuals over forty years old as first-time military personnel than before.
The trendlines on age and G.E.D. degrees have begun to cause considerable concern in the last year or two—though the nature of the problem needs to be kept in perspective. The G.E.D. is considered academically equivalent to a high school diploma, and certainly the military can ensure that anyone with such qualifications is up to par by testing them in other ways, too. Moreover, as of 2005, 90 percent of DoD recruits continued to have high school diplomas, comparable to the 1985 figure at the height of the Reagan buildup. And the typical recruit scored better on the Armed Forces Qualification Test (AFQT) in 2005 than in 1985. That said, while figures for the other services have remained good, the Army has experienced some problems, with the high school graduation figure for 2005 dipping to about 80 percent (worse than the 1985 figure of 86 percent, though still much better than the typical level of around 55 percent in the 1970s).39 By 2007, the percentage of high school graduates had declined to below 80 percent; it rebounded modestly to 83 percent in 2008, but further progress is still needed.40
There has been a recent rumor that West Point graduates have been leaving the service at drastically increased rates as soon as their minimal obligations are satisfied. In fact, this appears not to be true. The last year for which data is available as of this writing (the class of 2002, which was eligible to leave the service as of 2007), showed a 68 percent reenlistment rate, only 4 percentage points below the 1990s average.41 More generally, company-grade officers (first and second lieutenants as well as captains) have not been leaving the force at a greater than normal rate either; the average rate during the Iraq War has been less than the average rate of the late 1990s, for example.42 A similar conclusion is true of majors.43 Nonetheless, the Army is now short several thousand officers in aggregate.44
This is largely because the Army is increasing the number of officers needed as it enlarges the number of brigades in its force structure. In addition, the Army did not retain enough young officers in the early 1990s, meaning the current pool of officers from which to recruit for mid-level positions is too small.45
Other matters are more worrisome as well. While the number of individuals scoring relatively high on aptitude tests remains better than the 1980s, trends are in the wrong direction.46 Moral waivers for matters such as criminal history have increased substantially in recent years, with a total of 860 soldiers and Marines requiring waivers from convictions for felony crimes in 2007, up by 400 just from the year 2006. While most of the convictions were for juvenile theft, and the aggregate total is modest compared with the size of the force, only by arresting such trends will the quality of the force be ensured.47
Divorce rates have leveled off somewhat at about 3.5 percent, after reaching 3.9 percent in 2004, and are not worse than in the general population—but they are still above the 2.9 percent level of 2003.48 Suicide rates are a significant problem for the military, and an extremely tragic reality for so many troops and their families. The rate reached 17.3 per 100,000 soldiers in the U.S. Army in 2006. That is not far off from the age-adjusted and gender-adjusted average for the U.S. population on the whole (for males, for example, the rate is 17.6 per 100,000), but rates are at record levels today—much higher, for example, than the rate of 9.1 per 100,000 soldiers in 2001—and as such a serious reason for worry.49
For one group of soldiers surveyed in 2008, among those who had been to Iraq on three or four separate tours, the fraction displaying signs of post-traumatic stress disorders was 27 percent (in contrast to 12 percent after one tour and 18.5 percent after two). As of early 2008, among the 513,000 active-duty soldiers who have served in Iraq, over 197,000 had served more than once, and over 53,000 had deployed three or more times so many are now at such risk.50
Some of the problems mentioned here for the Army are hardly without precedent. For example, in the late 1990s the Air Force pilot shortage was more than 10 percent, as no-fly zones over Iraq taxed airmen and airwomen’s patience, and high pay rates in the civilian economy enticed many to doff their uniforms. This would clearly have been a problem for any operation requiring a large fraction of the force structure, such as a major war. Through careful monitoring of the deployment burdens of various units and individuals, however, it was at least temporarily manageable for the challenges of the 1990s. In addition, certain other military specialties were overused, and various individuals were overdeployed, because the force structure had not been properly constructed for frequent deployments that were especially demanding of a particular type of military skill set. Shortages of certain “low density / high demand” military specialties also arose. These specialties existed within the military in only modest numbers, but were frequently needed in certain kinds of operations, and included skill sets such as computing, aviation, foreign languages, policing, and other military activities that could become particularly (and unexpectedly) important in a given mission. Individuals with such skills could often find better opportunities in the private sector. It can be hard to anticipate shortfalls in such areas without access to more detailed information on military manpower than most analysts are generally able to access, especially since the Department of Defense is not always as forthcoming as it could be with relevant information.
On top of that, some units were undermanned and had to be reinforced by individuals borrowed from other units when preparing to deploy. This was acceptable on the whole, but led to situations where some people were deployed with a different unit than their own.51 If their own unit was itself later deployed, the same person might wind up deploying abroad twice in a relatively short time; the military personnel system did not protect them from such inadvertent overuse. This was more a matter of fairness to military personnel, and sustainment of good morale, than a crisis for warfighting readiness, but it was important to address just the same. Again, the moral of the story is that not all problems with military readiness are of equal concern, though most merit attention and remedial action.
In the preceding cases, targeted policy interventions usually occurred. For example, pilot bonuses were increased.52 In addition, in some cases a higher level of future attrition was simply assumed to be likely (especially at a time when a strong civilian economy created many high-paying commercial jobs for pilots), and more military pilots were recruited and trained to compensate. Recruiting and advertising budgets, as well as the number of personnel assigned to the recruiting task, can also be varied. In fact, there is a fairly detailed econometric literature on just how well each type of policy tool tends to work historically (in terms of increasing recruiting or retention per dollar spent). Such measures tend to be capable of addressing most discrete, specific problems within two to five years—if that much time is available.
Other changes can be considered as well. More military specialties can be rewarded with differential, added pay (rather than linking compensation so linearly to military rank). Because military pay, while never truly enough to compensate those who actively risk their lives for their country, is nonetheless reasonably good by comparison with private sector jobs for individuals of comparable age, experience, and education in the U.S. population, the idea of selective bonuses for certain specialties would seem reasonable. Equity and fairness considerations need not preclude such an approach of selective targeted raises and bonuses.53
Military pension plans could also be modified. The military retirement system is essentially an “all or nothing” operation. Stay in for twenty years and become fully vested and immediately eligible for full benefits; stay in a day less and receive nothing. This approach probably hurts retention for those considering whether to stay in the military or not as they approach five to ten years of service. That is the period when another few years of military employment promises no accrued pension benefits whatsoever (unless personnel wind up staying the full twenty years), compared to many private sector jobs that would begin to vest them quickly. These types of reforms may be considered simply as a matter of fairness, or of keeping up with the times and the changing nature of the American economy. Yet they may not be given sufficient attention absent a problem in readiness that motivates innovation.54
Details of the exact state of the U.S. military have only limited bearing on the purpose of this textbook, of course. But they are provided in some detail to give a sense of the kind of data that are available—and of the kinds of policy challenges that need to be faced. The broad analytical conclusion from the preceding survey is that one always needs to ask how serious a given shortfall might be when measured against the overall capability of the military (and likely demands of war); what policy recourse might be available to address a given problem as well as how long it might take to implement the repair; and what, if any, immediate stopgap solutions can be adopted to mitigate the strains on people in the short term. It is also important to study trends; problems may either get worse or ameliorate with time. Those that are worsening naturally demand the greatest attention.
If an overall assessment was to be offered of the U.S. ground forces in 2008, it is that of an Army and Marine Corps under serious strain but collectively holding up. Most indicators are not worsening, though they are less healthy than in most periods of the 1980s and 1990s. As such, complacency is hardly in order; we should be concerned that the reasonably good readiness of the military today is fragile and not sustainable indefinitely, even if at present it seems to be sustainable for the foreseeable future. Moreover, at the individual level, many soldiers and Marines are facing enormous hardship, raising fairness and equity issues for a democracy at war, and asking so few to do so much for so long on behalf of the nation.55
Training
In the Vietnam and immediate post-Vietnam eras, military training was not nearly rigorous enough. However, great attention was focused on this matter in the 1980s, demanding regimens were developed, resources were amply provided for training, and subsequent military performance was seen to improve greatly, as validated not only in training but in certain military operations such as the 1986 air attack on Libya, the invasion of Panama, and eventually Operation Desert Storm. Ever since that period, Reagan-era standards have remained important in determining how forces should be prepared—from basic training to specialized training to unit training at main bases to large-formation training at the various weapons schools and combat training centers.
Challenges remain, however. In normal peacetime, one concerns the growing role of simulators. As flight simulators and even tank simulators become much more realistic, to what extent can they replace the need for true training? Military personnel need to work with real weapons, realistic (if simulated) battlespaces, and real ammunition—not only at a technical level, but to acclimate people to the pressures and fear of combat. That said, do we really know for a fact that tank crews need 800 miles a year of driving their vehicles (rather than 600, with many more hours on simulators)? Or that pilots need twenty to twenty-four hours in the air per month, the late–Cold War norm, rather than today’s fourteen to eighteen?56
These questions underscore the degree to which tracking readiness cannot be purely a technical or an accounting exercise. Judgment is always needed. Mistakes are always possible, and sometimes they are not appreciated until a force engages in actual combat. One reasonable guideline is that we should be wary about making rapid major changes in how we train, absent periodic validation under stressful realistic circumstances (such as actual war time operations) that combat skills are remaining strong.
An even more difficult problem has been faced by the Army and Marine Corps during the Iraq and Afghanistan wars. Soldiers and Marines have virtually no time to do anything more than deploy to the theater of combat, return, rest, and then prepare to go back. Generalized training in other types of combat, besides the counterinsurgency and counterterrorism now being carried out in Iraq and Afghanistan, is by necessity being neglected. The assumption is that forces who performed so brilliantly in classic combat in 2003, and who have been so hardened by ongoing combat of a different type since then, will remain proficient for the full range of possible missions for the foreseeable future even without the full range of training as required by official doctrine. But that is an assumption, not an obvious truth. The assumption can be periodically tested by asking modest numbers of troops to be subjected to tough assessments of their skills in other types of combat on the training ranges (even if there is not enough time nor enough resources to do so for most formations). But again, a level of uncertainty exists in measuring readiness that is hard to eliminate entirely.
Equipment
Combat units clearly need vehicles in working order, enough ammunition to fight for a certain time, and enough spare parts to fill projected demands. The question is not just instantaneous readiness, but the ability to sustain operations for some reasonable period.
Assessing equipment readiness is in the end probably easier than doing so for people or even training. It is generally a question of countable hardware, not human skills. One challenge is in evaluating the importance of a shortfall in which equipment inventories drop by a certain moderate amount, say 10 to 15 percent relative to nominal requirements of “mission capable” aircraft or tanks or trucks or ships. At what point does the cohesion of a tank unit—its ability to carry out basic operations such as maneuver and attack in a coordinated fashion—suffer in that case, for example? Put differently, at what point does expected combat performance (number of accurate rounds delivered on target, attrition rates over a certain period of time in battle, and so on) decline precipitously and by a much larger amount than the reduction in available equipment or manpower might suggest? To some extent, the answer to this question will be scenario-dependent of course, making the calculation quite hard in some cases. As a broad rule of thumb, however, the military’s own collective judgment and official rating systems tend to suggest that shortfalls in the range of 5 to 15 or 20 percent of most types of equipment are tolerable, but that larger deficits quickly become of serious concern.
Sometimes a shortfall in one area can be balanced out by surpluses elsewhere. Consider the state of the U.S. Army in recent years. For most major types of vehicles—all classes of helicopters, Abrams tanks and Bradley fighting vehicles, medium weight trucks—there has been no major crisis due to the wars in Iraq and Afghanistan. No more than 20 percent of the total inventory of most weapons has been in the Central Command theater at a time (according to Congressional Budget Office data published in 2007). For most major fighting vehicles and helicopters, there was no shortage of usable equipment for forces based back in the United States. There were, however, notable shortfalls of up-armored HMMWVs, MRAPS, Strykers, and two of seven types of trucks. For the trucks, since there were substantial surpluses in some of the other five categories of trucks, there was probably little major problem. For the armored vehicles, however, there would clearly be great difficulty in finding a way to deploy many to any new scenario that might develop fast. So the Army equipment readiness issue is quite specific—potentially serious for some scenarios, much less so for others. On balance, while Iraq and Afghanistan have taxed the equipment inventories of the U.S. ground forces in particular, the far greater strain at this point is on people, not weaponry.57
AMERICA’S BUDGET IN INTERNATIONAL PERSPECTIVE—
AND THE QUESTION OF CHINA’S MILITARY SPENDING
Most of the previously mentioned budgetary tools are focused on the U.S. defense budget to allow evaluation of possible specific policy alternatives. But it is also important to understand America’s defense budget in a global context. Comparative military spending levels are hardly definitive for measuring relative power or predicting combat outcomes; that is why a later section of this book focuses on combat simulation and modeling. But budget comparisons are an important and salient gauge of the effort and resources a country puts into its armed forces, if nothing else. While outputs matter more than inputs, the latter are hardly insignificant.
At this moment in history, two broad questions stand out. First, why is American defense spending so high compared with the military budgets of other countries? Second, and somewhat relatedly, how should one interpret the rapid growth in the defense resources of the People’s Republic of China?
On the first point, U.S. military expenditures are indeed remarkably high. There is no other way to put it. As noted earlier, according to the most commonly accepted comparative measurement by the International Institute for Strategic Studies in London, U.S. expenditures represented 41 percent of the world’s total in 2006. In fact, America’s share grew further in the next couple of years, given the high costs of war supplementals.58
As noted earlier, in 2009, U.S. defense spending as a fraction of the nation’s gross domestic product is estimated at roughly 4.5 percent—hardly a small number, and significantly more than the 3.0 percent total to which defense spending dropped in 2000 before beginning its upward trend. But again, that 4.5 percent figure is about half the norm of the 1950s and 1960s, and about three-fourths the average figure during the Reagan years. U.S. defense spending is also now about 20 percent of federal government outlays, in contrast to nearly half in the 1960s, for example.59
Nonetheless, American defense spending is enormous. It easily exceeds the Vietnam and Reagan-era peaks in real-dollar (inflation-adjusted) terms. It is nearly half the world’s total military spending, whereas U.S. GDP is only one-quarter of the global aggregate. It dwarfs China’s military spending of about $135 billion (see the following pages for more on this), or Russia’s spending of about $75 billion (these are 2006 figures, expressed in 2009 dollars). Most of the next tier of top military spenders are American allies (France and Britain each spent about $60 billion in 2006, Japan $45 billion, Germany $42 billion, and Italy $34 billion). Saudi Arabia and South Korea rounded off the top-ten list at $33 billion and $27 billion, respectively, with India next at $25 billion. Then came another slew of American partners and allies including Australia ($19 billion), Brazil ($18 billion), Canada ($17 billion), Spain ($16 billion), Turkey ($13 billion), Israel ($12 billion), the Netherlands ($11 billion), the UAE ($10 billion), and Taiwan ($9 billion). Among major American worries, the lead spenders are Iran ($8 billion), North Korea (about $2 billion to $5 billion, though estimates are difficult to obtain), Venezuela ($3 billion), Cuba ($2 billion), and Syria ($2 billion).60 All these latter allocations are extremely modest—as are the working budgets of groups such as al-Qaeda and various extremist, overseas militias (which measure in the tens or hundreds of millions of dollars a year at most).61
For some, the preceding information is enough to conclude that U.S. defense spending is not only large but exorbitant and unnecessary—especially in an era of large U.S. budget and trade deficits. Indeed, when NATO and East Asian allies are figured in, the Western alliance system accounts for about 70 percent of global military spending, and when other allies in places like South America are also included, as well as countries having security partnerships with the United States such as Taiwan, Israel, and the Persian Gulf sheikdoms, the broader U.S.-led global alliance system’s military spending exceeds 80 percent of the world’s total.
However, economists are divided about whether U.S. budget and trade deficits are deeply harmful of the American economy. Certainly, most would prefer smaller deficits, but even if that is the case, whether the military should take the lead on producing savings is a debatable proposition. (In my own work, I have used the budget deficit as motivation for seeking more painful reductions in defense modernizations than the services would prefer, but there is no conclusive way to insist that this must be done as a simple matter of long-term national security.)
Moreover, simple comparative metrics are hardly adequate to determine proper force sizing or budget levels. There are multiple reasons for this, beginning with the simple observation that American defense spending would look huge comparatively even at $200 billion or $300 billion—and it is hard to find proposals for cutting American spending below $200 billion among serious defense scholars (indeed, it is increasingly hard to find proposals for amounts below $300 billion). This suggests that the comparative approach is more useful for establishing context than reaching conclusions.
An important reason for America’s high defense spending is its large number of overseas interests and allies. Allies add to the strength of the Western alliance on what might be called the supply side, but they also potentially add burdens to the United States on the demand side. With several dozen formal security partners, and large military deployments in three main regions of the world (East Asia, Europe, and the broader Middle East) as well as smaller commitments in numerous other places, the United States has many actual and potential military obligations. Moreover, to be reached by American forces, these distant theaters all require a substantial effort from the U.S. military, adding to the difficulty of potential missions—and limiting the utility of comparative defense budget analysis, since potential enemies would generally be fighting on or near their home turf.
Another important explanation for the disproportionately large U.S. defense budget is that the United States seeks a major qualitative advantage in military capability. It is not interested in a “fair fight,” that is to say an even competition. Rather, it seeks a major military advantage. Such superiority, so the logic goes, should enhance deterrence by reducing the likelihood that other countries will choose to challenge the U.S. military. To put it more negatively, history is full of examples in which the smaller military, and quite often the less expensive one, prevails in battle. While high spending cannot totally overcome the distinct possibility that the underdog will win in war, it can certainly make the underdog’s job much harder.
Qualitative superiority also helps compensate for the modest size of the U.S. armed forces, which have been severely strained in the process of trying to stabilize two mid-sized nations of about 30 million people each in recent years. The active-duty American military, at about 1.5 million uniformed personnel, is certainly not large in a historical or current international perspective. Not only is it down from the range of 2.0 to 2.25 million that characterized the post-Vietnam Cold War military (and much higher levels during Vietnam and Korea), it represents less than 10 percent of the world’s total of 24 million individuals under arms. China leads the way at 2.1 million. India at 1.3 million, North Korea at 1.1 million, and Russia at 1.0 million are not too far behind the United States by this measure. (South Korea, Pakistan, Iran, Turkey, Egypt, and Vietnam occupy the next tier in terms of size, with armed forces ranging from 450,000 to 700,000 personnel each.)62
The premium that the United States places on operating globally, and with a high-tech advantage, leads among other things to very large budgets for the Navy and Air Force. Each of these has a budget comparable in size to that of the U.S. Army; it is unusual for a country’s air and naval capabilities each to cost as much as its ground forces.
The mention of China, moreover, underscores the point that American defense planning must look to the future, not just the present. A country with 2.1 million persons under arms, a military spending level exceeding $100 billion a year and growing fast, and irredentist claims on a key American economic and security partner (Taiwan) demands at least some concern. Yet it is also important to place this spending in perspective and not be more worried about it than the evidence would warrant.
The official defense budget of the People’s Republic of China (PRC) at market exchange rates was $47 billion in 2007. But that figure is just the starting point for gauging China’s military resource allocations.
The PRC official number substantially understates actual military resources for two main reasons. First, due to limited transparency in its official papers and budgets, it fails to include many defense activities commonly recognized as intrinsic to a military budget by standard NATO definitions (or common sense). These feature foreign arms purchases, military-related research and development, nuclear weapons activities, and paramilitary forces. Altogether, the absence of these categories of expenditure creates a significant error in the official budget. Thankfully, corrections can be made with reasonable accuracy. For example, China’s arms purchases from abroad (largely from Russia) in recent years have reached roughly the $2 billion to $3 billion annual level, and its nuclear costs are probably in a roughly comparable range, as are its research and development costs. Various types of industrial subsidies given to the defense industry may total $5 billion to $10 billion a year. Overall, when adjustments are made to the official budget to account for such undercounting, the total budget probably increases $15 billion to $20 billion a year.63
Second, as a developing economy, China (like many other countries) has a number of military-related costs that are quite inexpensive by Western standards. When China’s defense budget is expressed in dollars, therefore, and a comparative sense of its military resource allocation is sought, it is more useful to express figures in terms of purchasing power parity than straight exchange-rate comparisons. Making such conversions is difficult. While the fact that a Chinese soldier can be fed and housed much more cheaply than a Western soldier is an advantage for the PRC, and should probably be acknowledged in any careful comparison, the fact that a typical Western soldier’s education and salary are much more expensive reflects a qualitative advantage of that soldier. So not all costs can be converted the same way.64
When all is said and done, the United States government has typically estimated China’s actual military resource levels as two to three times its official numbers. A reasonable midpoint estimate for China’s spending level in 2006 is thus about $135 billion, and the 2007 level may have reached close to $150 billion.65
As noted, this only provides a general context. For a possible war, say over Taiwan, would China’s greater geographic proximity and greater sense of national vital interest allow it to fight the United States to a draw—or at least have a chance to do so—in any future war? It is hard to determine the answer from defense budget comparisons alone. Trying to reach judgments on this complex subject will have to await the chapter on modeling. For now, the point is simply to establish some sense of the scale of China’s effort and the pace at which it is accelerating that effort.
ILLUSTRATIVE COST CALCULATIONS AND COMPARISONS
To illustrate how the preceding defense budget costing tools can be used, and to address several issues important in their own right as well, it is useful to consider a number of specific questions in which budgetary methods can answer a question—or at least inform a broader debate. They are examples of important questions, and my approaches to addressing them are examples as well; other approaches can be imagined in most cases.
QUESTION 1: What is the most economical way to destroy 100 fixed aimpoints within a given country?
ANSWER: This is a very “nuts and bolts” type of question, less strategic than tactical and even technical, but it is nonetheless an example of the kind of practical matter that force planners must consider frequently.
Of course, this question can have many variants, depending on which weapons are available and capable of carrying out the attack in question. There is no right answer across all scenarios.
To simplify the problem, though, let us focus on one particular choice: whether to use a stealthy, “penetrating” bomber like the B-2 (carrying cheap, “dumb” bombs) to attack targets directly, or whether to use aircraft from standoff distances that do not penetrate the airspace in question, but instead use cruise missiles (such as today’s B-52s, or perhaps a military variant of a 747 in the future). Of course, this comparison only makes sense if both the bombers and the cruise missiles can successfully reach their targets. Or, if some losses are expected, the costs of replacing the losses should be factored into the calculation, as should the implications of losing bomber pilots—or having them captured. For that and other reasons, simple cost comparisons like the one done here are only a single element of any assessment about which weapons to employ, and which to purchase and maintain in a country’s force structure.
Since the United States already has B-2 and B-52 bombers, this example is somewhat contrived. But it could still be relevant in deciding which type of plane to buy in the future.
To simplify, assume that the B-52 like aircraft can carry ten cruise missiles and the B-2 bomber can carry twenty bombs (either a truly “dumb” bomb with no guidance package, or an inexpensive short-range guided munition such as the joint direct attack munition or JDAM, which is terminally guided by GPS). Further assume that the cruise missiles and bombs are equivalent, in terms of expected effect on the target—their explosive power and their accuracy are comparable, so the same number of weapons must be used on each target regardless of which ordnance is selected. (In fact, assume that two weapons must be dropped on each target.) If the targets must all be destroyed on the first sortie of planes, to maximize the shock value and prevent the enemy from recovery and prompt retaliation, we need a way to deliver 200 weapons to target in a single salvo. Using these assumptions, that would require either ten B-2 bombers or twenty B-52/747 aircraft.
To complete the calculation, some information on costs is needed. From Table 1.6, assume that the marginal production cost of buying B-2s is $850 million an airplane (assuming that the research and development of the plane was already completed—meaning that such costs are “sunk” and as such no longer relevant to future policy choices). And assume the cost of a B-52 or 747 carrying cruise missiles to be $250 million, again not far off from reality. Cruise missiles are assumed to cost $1 million each in the following; JDAM bombs $20,000 apiece.
So buying ten B-2 bombers and 200 JDAMs would cost about $8.505 billion, whereas buying twenty B-52/747 aircraft and 200 cruise missiles would cost $5.2 billion. Clearly, if only one mission of this type were expected over the lifetime of the aircraft, the B-52/747 option would be cheaper—assuming that indeed it would be equally effective militarily. Factoring in estimated operating costs over a thirty-year lifetime (see Table 1.6), the answer would not change much at all, since the added cost per plane of maintaining the B-2 would be largely canceled out by the fact that more B-52s would be needed to accomplish the same mission according to assumptions about likely weapons payload.
Of course, in reality one would have to reach some conclusion about how many such missions might be expected over the lifetime of the planes, and how many other types of missions might be assigned to the planes, before reaching any firm decisions about the respective budgetary attractiveness of each option. As such, these kinds of cost calculations are best viewed as tools to aid the military judgment process, not means of arriving at a definitive “correct answer” in and of themselves.
QUESTION 2: What is the most cost-efficient way to carry out the forward presence mission of the U.S. attack submarine force?
ANSWER: This is again a fairly specific, technocratic question—but it is needed to inform a broader discussion of where to base U.S. military assets and how to operate them in overseas theaters.
American attack submarines are used frequently for intelligence-gathering and related purposes in places such as the Western Pacific Ocean and the Persian Gulf. Here we consider different ways of carrying out this mission.
The U.S. attack submarine force has had multiple missions over the years, many secret, and the Navy has generally been extremely reluctant to describe how it has sized and shaped that part of its force structure. But in recent years, Eric Labs at the Congressional Budget Office and others have been increasingly successful in gaining information about the employment of the attack submarine force. It now appears that the main mission of much of that force—or at least, the mission that is the most taxing on the Navy, and hence the one that determines the size of the attack submarine fleet—is maintaining presence in theaters such as the Western Pacific and the Persian Gulf/Strait of Hormuz. Understanding the operations of navies such as those of the PRC and Iran, as well as understanding the properties of the waterways near these strategically crucial regions—and being ready for rapid response in the event of crises in such places—have become the main missions that determine the size of the attack submarine (SSN) force.
As such, to give a concrete example of policy choices, consider the following: How much could the Navy save by basing more SSNs at Guam? Three are homeported there at present, and focused naturally on missions in the Western Pacific. The Navy could reportedly find room for at least another half dozen SSNs in the island’s port facilities (once suitably upgraded). Of course, other considerations must enter into the final decision-making as well, ranging from the somewhat mundane matter of the quality of life for Navy crews based at Guam, to the more strategic matter of the relative vulnerability of attack submarines to surprise attack when based on Guam or Hawaii or California. But assuming such considerations do not decidedly argue against Guam, the cost comparison becomes of great interest.
An SSN operating from a mainland U.S. port cannot maintain much time deployed in places where it can carry out reconnaissance, train, and be prepared for crisis response. In fact, its ability to stay on station is surprisingly limited, much less than for surface ships apparently—only about thirty-six days a year. By contrast, an SSN homeported on Guam can do about three times as well (roughly 106 days a year). Employed in this way, it will spend slightly more time per year at sea than a U.S.-based submarine, but manage triple the useful time.
Using the SSN more frequently in this way reduces its expected service life somewhat, from thirty-three to twenty-seven years, and increases annual operating costs by about $3 million, from $41 million to $44 million. In addition, about $10 million in added infrastructure would need to be built for each SSN added to Guam.
As such, the cost comparison can be approximated in this way. One SSN on Guam can do the work of three based in California, according to CBO. With a procurement cost of $2.6 billion per sub, the annual average cost of operating one SSN from Guam is about $140 million (dividing the procurement cost by twenty-seven years and then adding in average annual operating and infrastructure costs, in the same sort of way as employed in Table 1.6 earlier). By contrast, the annualized cost of operating three SSNs from the mainland United States is about $360 million. Clearly, if cost is the main issue, Guam wins the comparison hands down. But again, such calculations are aides to decision makers, not final answers themselves, as many other factors typically must be considered when determining force structure and basing.66
Of course, this option could not be pushed too far; not only would Guam run out of space for Pacific-oriented submarines, but placing too many in one place would increase the dangers of a successful enemy surprise attack. Moreover, forward presence is not the only mission of the submarine fleet, so a certain number of submarines are needed for surge purposes in a crisis or conflict regardless of the efficiencies associated with forward homeporting. But for a modest additional number of submarines, Guam basing is an interesting option.67
QUESTION 3: How much per year, in peacetime, does the United States spend defending Persian Gulf oil?
ANSWER: Since the Cold War ended, possible military scenarios in the Persian Gulf have typically represented one of the two main sets of possible operations around which the Pentagon has built its combat force structure (the other set being in East Asia). In addition, the Navy and parts of the ground forces (especially the Marines, and some Army light forces) have focused more broadly on maintaining presence and quick-response capability across multiple theaters. That would seem to suggest three chief overall missions for main American combat forces, two geographically specific and one more general.
Given that the current peacetime defense budget of the United States is just over $500 billion, this framework might seem to imply costs of about $100 billion a year to defend the Persian Gulf. That estimate can be reached by first factoring out $200 billion of the defense budget devoted to research and development, intelligence, homeland defense, nuclear forces, and other such costs not easily attributable to any geographic theater, as well as the costs of a core military infrastructure including major commands and educational institutions and the like in the United States (see Table 1.3). Then, the remaining $300 billion can be divided into three relatively equal parts.
This is clearly a very rough way to gauge costs. Another way might invoke the Bill Kaufmann breakdown of U.S. military spending by region. He estimated in 1992 that the United States spent 20 percent of its peacetime defense budget on defense of the Persian Gulf and broader Middle East (see Table 1.4). Applying that same percentage today would also suggest a cost of about $100 billion a year at present (again, not counting the costs of the ongoing Iraq War).
However, these figures may be too high. Many forces that would be assigned to Central Command in wartime are available for other purposes as well. (This is the obverse of the current situation, in which forces that could otherwise be used in Europe or East Asia or elsewhere are taking their turns on deployment to Iraq and Afghanistan.) A figure of $100 billion a year may not be a bad estimate of the costs of forces that would be most likely deployed to operations in the Persian Gulf, as their most probable main mission. However, this overlooks the fact that many of them could have secondary purposes as well.
So the real question is how much less would the United States spend on its overall military if the Persian Gulf were somehow dropped from the list of overseas commitments and possible wartime theaters? That requires an effort to estimate what force posture the United States would want to keep and thus is inherently somewhat interpretive and subjective. Regardless, however, the savings would be somewhat less than $100 billion a year, since some of the forces normally assignable to the Persian Gulf might need to be kept for other possible scenarios (such as stabilization of a collapsing Pakistan or Indonesia, multilateral enforcement of a peace deal between India and Pakistan over Kashmir, a temporary international trusteeship for an independent state of Palestine, a sustained air patrol in the Taiwan Strait if China/Taiwan tensions heat up again and remain hot over an extended time, or a possible Russian menace to the Baltic states, now NATO members).
This question is too complex to lend itself to a precise answer. But figures in the general range of $50 billion a year are probably reasonable answers to the question posed. Such a figure is probably what should be used when, for example, one wishes to compare the costs of alternative energy policies to the status quo.
QUESTION 4: After current wars are over, should the United States reduce its military readiness to fund transformation?
ANSWER: This question again goes well beyond budgetary matters, but budgetary calculations can help inform the answer to the question.
To be specific, and dramatic, let us assume that half of the entire U.S. military is put into a second-tier readiness status, such that its training activities are scaled back dramatically—perhaps to a level only slightly greater than the normal training and readiness level of the National Guard and Reserve. That might suggest a two-thirds reduction in readiness activities and resources, including those to buy fuel, ammunition, spare parts, and other such supplies.
To estimate savings, we could begin with that part of the operation and maintenance budget focused on training, maintenance, and the like, and reduce it accordingly. (That is, we would exclude those O&M expenses related to civilian pay, military health care, environmental cleanup, and the reserve component; we would also assume that neither military pay nor investment accounts would decline under this option.)
As shown in Table 1.1, the current O&M budget is about $180 billion a year; civilian pay is about $60 billion of that total. Cumulative costs in the O&M budget for health care, environmental cleanup, basic base maintenance, recruiting, and the like, as well as the military reserves and National Guard, total another $60 billion or so. That leaves $60 billion for what might be called core readiness activities for the active forces. The respective amount spent by the half of the force structure in question is about $30 billion. Reducing this by two-thirds would save $20 billion a year.
More detailed answers to more specific proposals, say about reducing the readiness of one service or another, could be calculated by looking at the respective O&M budget of the service in question and breaking it down similarly. Precise answers would obviously require much more information, including detailed information on which types of supplies could be cut back by what percent in light of a given reduction in readiness. Such precise answers are ultimately needed if a policy change is being budgeted for. But they are often so difficult to produce that it is useful, for brainstorming purposes, to have a way of obtaining a general and approximate estimate like that provided here.
QUESTION 5: How much would the United States need to cut back the Navy or Air Force to add two divisions to the Army?
ANSWER: Using Bill Kaufmann’s estimates from Table 1.5, we can estimate the annual average cost of an active Army division (with about 16,000 troops in the division, including four brigade combat teams, and 25,000 to 30,000 more soldiers supporting it indirectly) at about $5.2 billion. Providing two more would therefore cost $10.4 billion a year (discounting economies of scale to first approximation, as Kaufmann does). Consultation of Kaufmann’s list of the costs of other major formations leads to several possibilities for finding that money. The core of one approach could cut three wings of fighter aircraft (at seventy-two aircraft per wing), or two aircraft carrier battle groups (each with about seventy-two aircraft itself, as well as the carrier and about four major warships as escorts), or some combination thereof.
As noted before, Kaufmann’s methods, while extremely useful as simplifications and approximations, have their limits. They overestimate savings when cutting forces and overestimate costs when expanding them (the reason for the latter is that some fixed or semi-fixed costs like R&D, central oversight, intelligence, professional educational institutions, and early expensive parts of weapons production runs need not be repeated—or at least need not be incurred at earlier cost rates—when additional forces are added). But other cost imprecisions tend to balance these out. So likely errors are in the range of 10 to 20 percent.
It is important to note that the previously mentioned costs are average annual costs, smoothed out over the entire lifetime of the typical weapon in a given unit. Because of political and budgetary realities, however, planners are usually worried less about costs averaged over twenty or thirty years, and more concerned about the next few years at any given moment. As such, it is important to know the details of what would need to be bought for a given policy change. If, for example, the United States has a surplus of tanks in storage that could be restored to good working condition fairly cheaply, investment costs for creating two new divisions might be relatively modest. But the situation could easily be the reverse as well, depending on the circumstances.
NOTES
1. Office of Management and Budget, Historical Tables: Budget of the U.S. Government, Fiscal Year 2009 (Washington, D.C.: Government Printing Office, 2008), p. 137.
2. International Institute for Strategic Studies, The Military Balance 2008 (Oxfordshire, England: Routledge, 2008), pp. 443–49.
3. Office of Management and Budget, Historical Tables, pp. 61, 79, 134–35.
4. To be somewhat more specific, about 90 percent of outlays for personnel take place in the first year and almost all the remainder in the second. About 48 percent of outlays for O&M take place in the first year (averaging across the services), another 40 percent in year two, about 5 percent in year three, and small amounts in years four through six. Procurement dollars are spent more slowly—only about 20 percent in year one, 40 percent in year two, 27 percent in year three, 8 percent in year four, and smaller amounts in the next three years. For research, development, testing, and evaluation, the figures are roughly 45 percent in year one, 42 percent in year two, 7 percent in year three, and small amounts thereafter. See Under Secretary of Defense (Comptroller), National Defense Budget Estimates for FY 2009 (Washington, D.C.: Department of Defense, March 2007), pp. 54–55.
5. M. Thomas Davis, Managing Defense After the Cold War (Washington, D.C.: Center for Strategic and Budgetary Assessments, 1997), pp. 1–3.
6. See Under Secretary of Defense (Comptroller), National Defense Budget Estimates for FY 2009, pp. 80, 102.
7. Raymond Hall, David Mosher, and Michael O’Hanlon, The START Treaty and Beyond (Washington, D.C.: Congressional Budget Office, 1991), . pp. 62, 135; Office of the Under Secretary of Defense (Comptroller), National Defense Budget Estimates for FY 2009 (Washington, D.C.: Department of Defense, 2007), p. 43, available at www.defenselink.mil/comptroller/defbudget/fy2009/fy2009_greenbook.pdf [accessed September 2, 2008]; and Stephen I. Schwartz, ed., Atomic Audit: The Costs and Consequences of U.S. Nuclear Weapons Since 1940 (Washington, D.C.: Brookings, 1998), p. 3.
8. Naturally, more detailed questions require more detailed analysis. For example, when determining compensation levels needed to ensure a given level of recruiting success, comparison between military and civilian pay levels for individuals of comparable age and skill are needed, as is historical data on the typical correlation between a given pay increase or other improvement in compensation and improved recruiting results. Such comparisons need to go beyond broad metrics to examine various types of compensation levels, and various types of occupational specialties. For example, on balance there is no systematic civilian–military pay gap (especially when all of DoD’s compensation, including tax advantages and housing and family subsidies are counted, without even including retirement or health benefits since they do not accrue to all). Counting all direct benefits, a typical twenty-two-year-old person with four years experience and an E-4 ranking earns the equivalent of $75,000 (in 2009 dollars) if unmarried, and about $90,000 if married with two children (for the former, for example, 33 percent of compensation is in basic pay, 18 percent in basic allowances for housing and subsistence, 3 percent in tax advantages, 8 percent in health care and other noncash benefits, and a total of 38 percent in deferred benefits [depending on the person], which include veterans’ benefits, retiree health care, and retirement pay). However, certain individuals with specific technical skills may well make substantially less in the military than they would in the private sector. See Carla Tighe Murray, Evaluating Military Compensation (Washington, D.C.: Congressional Budget Office, 2007), pp. 22, and 31–32; and Richard L. Fernandez, What Does the Military ‘Pay Gap’ Mean?” (Washington, D.C.: Congressional Budget Office, 1999), pp. 1–7.
9. Congressional Budget Office, “Analysis of the Growth in Funding for Operations in Iraq, Afghanistan, and Elsewhere in the War on Terrorism,” U.S. Congress, Washington, D.C., February 2008, p. 1, available at www.cbo.gov [accessed September 2, 2008]; Steven M. Kosiak, “The Cost of U.S. Operations in Iraq and Afghanistan and for the War on Terrorism through Fiscal Year 2007 and Beyond,” CSBA Update, Center for Strategic and Budgetary Assessments, Washington, D.C., September 12, 2007, pp. 1–5; and Testimony of Steven M. Kosiak before the Senate Budget Committee, “The Global War on Terror (GWOT): Costs, Cost Growth and Estimating Funding Requirements,” Center for Strategic and Budgetary Assessments, Washington, D.C., February 6, 2007, pp. 1–8.
10. Kosiak, “The Global War on Terror,” pp. 3–6.
11. Congressional Budget Office, “Costs of Operation Desert Shield,” CBO Staff Memorandum, Congressional Budget Office, Washington, D.C., January 1991, pp. 1–20; and Department of Defense, Conduct of the Persian Gulf War: Final Report to Congress (Washington, D.C.: Department of Defense, 1992), p. P-2.
12. Kosiak, “The Global War on Terror,” p. 6.
13. David Newman and Jason Wheelock, “Analysis of the Growth in Funding for Operations in Iraq, Afghanistan, and Elsewhere in the War on Terrorism,” Congressional Budget Office, February 11, 2008, pp. 1–3.
14. Robert F. Hale, Promoting Efficiency in the Department of Defense: Keep Trying, But Be Realistic (Washington, D.C.: Center for Strategic and Budgetary Assessments, 2002).
15. Elise Castelli, “Study: Intel Contract Employees Costly,” Defense News, September 1, 2008, p. 24.
16. Matthew Goldberg, Logistics Support for Deployed Military Forces (Washington, D.C.: Congressional Budget Office, October 2005), pp. 27–44; and Daniel Frisk and R. Derek Trunkey, Contractors’ Support of U.S. Operations in Iraq (Washington, D.C.: Congressional Budget Office, August 2008), pp. 1, 8, 13 (of 22), available at www.cbo.gov/ftpdocs/96xx/doc9688/MainText.3.1.shtml [accessed August 20, 2008]. About 190,000 contractors were working in Iraq in early 2008—just under 40,000 Americans, about 70,000 Iraqis, and about 80,000 third-country nationals. In most other wars, the United States had about one contractor for every five troops; in Korea, the ratio was 1 to 2.5.
17. The greatest inefficiencies occur when a production line is sized, scaled, and built for a certain total procurement buy, only to have that buy reduced in numbers thereafter. But there are penalties even when a production line can be redesigned for a smaller procurement lot. For example, according to one estimate based on CBO data and other sources, a hypothetical two-thirds reduction in the Air Force’s planned purchase of F-35 Lighting II aircraft (also known as the joint strike fighter), from roughly 1,700 planes to 500, suggested a 15 percent unit price increase. Also, a cut in the F-22 purchase from the intended 340 to about 125 was estimated to increase unit costs some 40 percent. See Michael O’Hanlon, “The Plane Truth,” Brookings Policy Brief #53 (Washington, D.C.: Brookings, September 1999), p. 6. More generally, according to CBO, a 50 percent reduction in the annual rate of production for aircraft increases production costs by roughly 10 percent to 35 percent (15 to 30 percent for most planes)—depending in part on whether the lower rate of production still remains above the “minimum economic rate” range or drops below it. A 50 percent reduction in the production rate for Army vehicles implied a 25 to 40 percent increase in unit cost, roughly, and for missiles, the typical unit cost increase was 15 to 50 percent. See R. William Thomas, Effects of Weapons Procurement Stretch-Outs on Costs and Schedules (Washington, D.C.: Congressional Budget Office, 1987), pp. 17–18.
18. See Frances M. Lussier, Structuring the Active and Reserve Army for the 21stCentury (Washington, D.C.: Congressional Budget Office, 1997), p. 3. At that time, for example, the Active Army had 176,000 soldiers in combat units, another 136,000 in deployable support units, and 183,000 in institutional and other nondeployable categories. For the National Guard, the respective numbers in the three categories were 175,000, 152,000, and 40,000; for the Army Reserves, they were zero, 139,000, and 69,000, respectively.
19. William W. Kaufmann, Assessing the Base Force: How Much Is Too Much? (Washington, D.C.: Brookings, 1992), pp. 70, 74, 82.
20. See, for example, Office of the Under Secretary of Defense (Comptroller), National Defense Budget Estimates for FY 2008 (Washington, D.C.: March 2007), p. 78.
21. Mark V. Arena, Obaid Younossi, Kevin Brancato, Irv Blickstein, and Clifford A. Grammich, Why Has the Cost of Fixed-Wing Aircraft Risen? (Santa Monica, Calif.: RAND, 2008), p. xvii.
22. One basic approach to making this estimate is first to see how much each service’s inflation-adjusted budget has grown since the period that Kaufmann studied. Kaufmann’s book was done in 1992, but he took projected forces and costs for 1997 as his focus of comparison, so the question is how much did real budget authority increase from 1997 through 2009 (not counting supplemental costs for Iraq and Afghanistan). The answer is about 45 percent for the Army, 30 percent for the Navy/Marines, and nearly 40 percent for the Air Force. Then, changes in force structure must be accounted for; these increased budgets now generally fund smaller forces, meaning the unit cost growth is even greater. Army force structure has on balance declined by just over 10 percent relative to planned base force levels, Air Force force structure has been reduced about one third, Marine Corps force structure has held relatively steady, and the Navy (excluding the carrier force) has declined in size by about one-third. See Office of the Under Secretary of Defense (Comptroller), National Defense Budget Estimates for FY 2008, pp. 125–27; International Institute for Strategic Studies, The Military Balance 2007 (Oxfordshire, England: 2007), pp. 28–38; International Institute for Strategic Studies, The Military Balance 1997/98 (Oxford: Oxford University Press, 1997), pp. 18–27; and International Institute for Strategic Studies, The Military Balance 1992–1993 (London: Brassey’s, 1992), pp. 18–28.
23. Raymond Hall, David Mosher, and Michael O’Hanlon, The START Treaty and Beyond (Washington, D.C.: Congressional Budget Office, 1991), pp. 139–40.
24. Amy Belasco, Paying for Military Readiness and Upkeep: Trends in Operation and Maintenance Spending (Washington, D.C.: Congressional Budget Office, September 1997), pp. 5, 14; Allison Percy, Growth in Medical Spending by the Department of Defense (Washington, D.C.: Congressional Budget Office, September 2003), p. 2; Gregory T. Kiley, The Effects of Aging on the Costs
of Operating and Maintaining Military Equipment (Washington, D.C.: Congressional Budget Office, August 2001), pp. 1–2; and Secretary of Defense Robert Gates, “FY 2008 President’s Budget for Defense,” Department of Defense, Washington, D.C., February 5, 2007, p. 5.
25. Lane Pierrot and Gregory T. Kiley, The Long-Term Implications of Current Defense Plans (Washington, D.C.: Congressional Budget Office, January 2003), pp. 44–46; Rachel Schmidt, An Analysis of the Administration’s Future Years Defense Program for 1995 through 1999 (Washington, D.C.: Congressional Budget Office, January 1995), pp. 41–44; and Adam Talaber, The Long-Term Implications of Current Defense Plans and Alternatives: Summary Update for Fiscal Year 2006 (Washington, D.C.: Congressional Budget Office, October 2005), pp. 20–22. In the latter report, CBO estimates that total DoD investment costs (development plus procurement) might be about 15 percent more than anticipated, allowing for the risk of growing costs—but this calculation includes all acquisition including ongoing production of systems for which costs are already well understood. For new systems, average cost increases are thus typically 25 percent (or more) in both the development and procurement phases.
26. Lane Pierrot and Jo Ann Vines, A Look at Tomorrow’s Tactical Air Forces (Washington, D.C.: Congressional Budget Office, January 1997), pp. 84–89.
27. Pierrot and Kiley, Long-Term Implications of Current Defense Plans, p. 43.
28. Adam Talaber, The Long-Term Implications of Current Defense Plans and Alternatives: Summary Update for Fiscal Year 2006 (Washington, D.C.: Congressional Budget Office, October 2005), pp. 1–2.
29. Quoted in Deborah Clay-Mendez, Richard L. Fernandez, and Amy Belasco, Trends in Selected Indicators of Military Readiness, 1980 Through 1993 (Washington, D.C.: Congressional Budget Office, 1994), p. 1, citing the definition from the Joint Chiefs of Staff, The Dictionary of Military and Associated Terms, JCS Publication 1 (Washington, D.C.: Department of Defense, 1986).
30. Richard K. Betts, Military Readiness: Concepts, Choices, Consequences (Washington, D.C.: Brookings, 1995), pp. 115–43.
31. Betts, Military Readiness, pp. 43–62.
32. Ibid., pp. 87–114.
33. Under Secretary of Defense (Comptroller) Tina W. Jonas, “Summary Justification: Fiscal Year 2009 Budget Request,” Department of Defense, Washington, D.C., February 4, 2008, p. 14.
34. Thomas E. Ricks, Fiasco: The American Military Adventure in Iraq (New York: Penguin, 2006), pp. 149–202; Susan L. Marquis, Unconventional Warfare: Rebuilding U.S. Special Operations Forces (Washington, D.C.: Brookings Institution, 1997), pp. 1–5; Frederick W. Kagan, Finding the Target: The Transformation of American Military Policy (New York: Encounter Books, 2006), pp. 92–100; and Ivo H. Daalder and Michael E. O’Hanlon, Winning Ugly: NATO’s War to Save Kosovo (Washington, D.C.: Brookings, 2000), pp. 125–26.
35. Some wartime operations are less vindications of readiness—again, the performance of tasks and missions that have in some sense been anticipated and planned for—than of real-time military innovation. In Operation Desert Storm in 1991, the U.S. armed forces developed an aerial tactic known as “tank plinking” in which laser-guided bombs were used against stationary Iraqi tanks that had been located on the desert floor by infrared detectors not originally intended for such a purpose. In short, the military learned it was not as ready to attack dug-in tanks as it might have been, but it soon found a way to become more effective. The Afghanistan mission of 2001 required real-time development of new tactics involving special forces, as well as local allies, in conjunction with another air-dominant American campaign plan. See Stephen Biddle, Military Power: Explaining Victory and Defeat in Modern Battle (Princeton, N.J.: Princeton University Press, 2004), pp. 199–201; and Michael R. Gordon and General Bernard E. Trainor, The Generals’ War: The Inside Story of the Conflict in the Gulf (Boston: Little Brown and Company, 1995), pp. 322–23.
36. David Von Drehle, “Cheney Steps Up Criticism of Military Readiness,” The Washington Post, August 31, 2000, p. A1.
37. Steven Lee Myers, “What War-Ready Means, in Pentagon’s Accounting,” The New York Times, September 4, 2000, p. A11.
38. Clay-Mendez, Fernandez, and Belasco, Trends in Selected Indicators of Military Readiness, pp. 9–14.
39. Heidi Golding and Adebayo Adedeji, Recruiting, Retention, and Future Levels of Military Personnel (Washington, D.C.: Congressional Budget Office, 2006), p. 6; and John Allen Williams, “Anticipated and Unanticipated Consequences of the Creation of the All-Volunteer Forces,” in McCormick Tribune Conference Series, The U.S. Citizen-Soldier at War: A Retrospective Look and the Road Ahead (Chicago, Illinois: McCormick Tribune Foundation, 2008), pp. 37–38.
40. Anita Dancs, “Military Recruiting 2007: Army Misses Benchmarks by Greater Margin,” National Priorities Project, January 22, 2008, available at www.nationalpriorities.org/militaryrecruiting2007 [accessed April 2, 2008].
41. Lt. Col. Bryan Hilferty, “Information Paper: West Point Graduate Retention After 5-Year Active Duty Service Obligation,” West Point, New York, December 5, 2007.
42. U.S. Army Fact Sheet, “U.S. Army Officer Retention Fact Sheet as of May 25, 2007,” U.S. Army, Washington, D.C., May 25, 2007, available at www.armyg1.army.mil/docs/public%20affairs/officer%20retention%20fact%20sheet%2025mayo7.pdf [accessed March 25, 2008].
43. Stephen J. Lofgren, “Retention During the Vietnam War and Today,” U.S. Army Center of Military History Information Paper, U.S. Army, Washington, D.C., February 1, 2008.
44. Michele A. Flournoy, “Strengthening the Readiness of the U.S. Military,” Testimony before the House Armed Services Committee, February 14, 2008, p. 3.
45. “U.S. Army Officer Retention Fact Sheet as of May 25, 2007,” available at www.armyg1.army.mil/docs/public%20affairs/Officer%20Retention%20Fact%2oSheet%2025Mayo7.pdf [accessed April 2, 2008].
46. Heidi Golding and Adebayo Adedeji, The All-Volunteer Military: Issues and Performance (Washington, D.C.: Congressional Budget Office, July 2007), pp. 14–17.
47. Ann Scott Tyson, “Military Waivers for Ex-Convicts Increase,” The Washington Post, April 22, 2008, p. A1.
48. Leslie Kaufman, “After War, Love Can Be a Battlefield,” The New York Times, April 6, 2008, p. ST1; and Pauline Jelinek, “Military Divorce Rate Holding Steady,” WTOPNEWS.com, March 1, 2008, available at www.wtopnews.com [accessed April 1, 2008].
49. Pauline Jelinek, “Army Suicides Highest in 26 Years,” Washingtonpost.com, August 15, 2007, available at www.washingtonpost.com/wp-dyn/content/article/20o7/o8/i5/AR20o7o8i502027_pf.htm [accessed April 1, 2008]; “Suicide Statistics,” available at www.suicide.org/suicide-statistics.htm/#death-rates [accessed April 15, 2008]; and Associated Press, “U.S. Army: Soldier Suicide Rate May Set Record Again in 2008,” The Examiner, September 5, 2008, p. 12.
50. Thom Shanker, “Army Is Worried by Rising Stress of Return Tours,” The New York Times, April 6, 2008, p. A1.
51. Laurinda Zeman, Making Peace While Staying Ready for War: The Challenges of U.S. Military Participation in Peace Operations (Washington, D.C.: Congressional Budget Office, 1999), p. xiii.
52. Department of Defense, Quarterly Readiness Report to the Congress, April–June 2000 (Washington, D.C.: Department of Defense, August 2000).
53. Carla Tighe Murray, Evaluating Military Compensation (Washington, D.C.: Congressional Budget Office, 2007), pp. 1–20.
54. Steven M. Kosiak, Military Compensation: Requirements, Trends and Options (Washington, D.C.: Center for Strategic and Budgetary Assessments, 2005), pp. 20–21; Cindy Williams, “Introduction,” in Cindy Williams, ed., Filling the Ranks: Transforming the U.S. Military Personnel System (Cambridge, Mass.: MIT Press, 2004), pp. 16–20; and Paul F. Hogan, “Overview of the Current Personnel and Compensation System,” in Williams, ed., Filling the Ranks, pp. 49–51.
55. For myself, in terms of policy recommendations, the above leads to three judgments: 1) we waited too long as a nation to increase the size of the Army and Marines after 2003, and we should keep increasing the size of the standing active ground forces today although we are belated in doing so; 2) we do not have the luxury of increasing overall deployed force levels abroad now, relative to where they were in late 2008, and we should be trying to draw. down, though the situation does not require this immediately as a matter of extreme urgency; and 3) we owe it to our men and women in uniform to further improve compensation and other help for them and their families, even if benefits are already relatively robust overall.
56. In 1990, the Air Force budgeted for 19.5 hours per air crew per month of flying, and the Navy for 24. In 2008/2009, the respective figures were down to roughly fourteen and eighteen a month. The decline was gradual; about half occurred during the Clinton years, the other half during the George W. Bush years. See Tamar A. Mehuron and Heather Lewis, “Defense Budget at a Glance,” Air Force Magazine (April 2008), p. 61.
57. Frances M. Lussier, Replacing and Repairing Equipment Used in Iraq and Afghanistan: The Army’s Reset Program (Washington, D.C.: Congressional Budget Office, 2007), pp. 1–15.
58. International Institute for Strategic Studies, The Military Balance 2007 (Oxfordshire, England: Routledge, 2007), pp. 406–11.
59. Office of Management and Budget, Budget of the U.S. Government, Fiscal Year 2008: Historical Tables (Washington, D.C.: Government Printing Office, 2007), pp. 133–36.
60. International Institute for Strategic Studies, The Military Balance 2007, pp. 357, 406–11; and Office of the Under Secretary of Defense (Comptroller), National Defense Budget Estimates for FY 2008, p. 134.
61. 9-11 Commission report, Kathleen Ridolfo, “Iraq: Smuggling, Mismanagement Plaguing Oil Industry,” Radio Free Europe/Radio Liberty, Washington, D.C., November 13, 2007, available at www.rferl.org/featuresarticle/2007/11/38c235c1-6f71-46ac-9463-4119f5cb6fea.html [accessed March 2, 2008].
62. International Institute for Strategic Studies, The Military Balance 2007, pp. 406–11; and Office of the Under Secretary of Defense (Comptroller), National Defense Budget Estimates for FY 2008, pp. 214–15.
63. International Institute for Strategic Studies, The Military Balance 2007 (Oxfordshire, England: Routledge, 2007), pp. 340–41; International Institute for Strategic Studies, The Military Balance 2008 (Oxfordshire, England: Routledge, 2008), p. 376; and Keith Crane, Roger Cliff, Evan Medeiros, James Mulvenon, and William Overhalt, Modernizing China’s Military: Opportunities and Constraints (Santa Monica, Calif.: RAND Corporation, 2005), pp. 101–3.
64. Department of Defense, Military Power of the People’s Republic of China, 2007: Annual Report to Congress (Washington, D.C.: 2007), pp. 25–29, available at www.defenselink.mil/pubs/pdfs/070523-China-Military-Power-final.pdf [accessed June 10, 2007].
65. Department of Defense, Military Power of the People’s Republic of China, 2008: Annual Report to Congress (Washington, D.C.: 2008), p. 32, available at www.defenselink.mil/pubs/pdfs/China_Military_Report_08.pdf [accessed March 10, 2008].
66. See Eric J. Labs, Increasing the Mission Capability of the Attack Submarine Force (Washington, D.C.: Congressional Budget Office, March 2002), p. 41.
67. Ibid., pp. 10–13, 41.
KEY REFERENCES AND SUGGESTIONS
FOR FURTHER READING
Betts, Richard K., Military Readiness: Concepts, Choices, Consequences (Washington, D.C.: Brookings, 1995).
Crane, Keith, Roger Cliff, Evan Medeiros, James Mulvenon, and William
Overhalt, Modernizing China’s Military: Opportunities and Constraints (Santa Monica, Calif.: RAND Corporation, 2005).
Department of Defense, Conduct of the Persian Gulf War: Final Report to Congress (Washington, D.C.: Department of Defense, 1992).
———, Military Power of the People’s Republic of China, 2008: Annual Report to Congress (Washington, D.C.: 2008), available at www.defenselink.mil/pubs/pdfs/China_Military_Report_o8.pdf.
Golding, Heidi, and Adebayo Adedeji, Recruiting, Retention, and Future Levels of Military Personnel (Washington, D.C.: Congressional Budget Office, 2006).
International Institute for Strategic Studies, The Military Balance 2008 (Oxfordshire, England: Routledge, 2008).
Kaufmann, William W., Assessing the Base Force: How Much Is Too Much? (Washington, D.C.: Brookings, 1992).
Kosiak, Steven M., Military Compensation: Requirements, Trends and Options (Washington, D.C.: Center for Strategic and Budgetary Assessments, 2005).
Labs, Eric J., Transforming the Navy’s Surface Combatant Force (Washington, D.C.: Congressional Budget Office, 2003).
Lussier, Frances M., Replacing and Repairing Equipment Used in Iraq and Afghanistan: The Army’s Reset Program (Washington, D.C.: Congressional Budget Office, 2007).
Newman, David, and Jason Wheelock, “Analysis of the Growth in Funding for Operations in Iraq, Afghanistan, and Elsewhere in the War on Terrorism,” Congressional Budget Office, February 11, 2008.
Under Secretary of Defense (Comptroller), National Defense Budget Estimates for FY 2008 (Washington, D.C.: Department of Defense, March 2007).
Williams, Cindy, ed., Filling the Ranks: Transforming the U.S. Military Personnel System (Cambridge, Mass.: MIT Press, 2004).