Concerns about the adequacy of the U.S. military frequently focus on two issues: whether its force structure has gotten too small, and whether U.S. forces are being modernized too slowly and are too dependent on an aging weapons inventory. While this report offers some partial answers to these questions, its goal is less to provide answers than to raise the level of discussion. The report is organized into four sections, focused respectively on:
- Historical trends in the size of the U.S. military’s force structure and the pace of its modernization efforts.
- The main drivers of these trends.
- The impact of these trends on U.S. military capabilities.
- Shortcomings in the U.S. military’s traditional approach to sizing, shaping, and modernizing its forces that may have left it with smaller and older forces than could have been sustained within historical funding levels.
More than anything else, the goal of this report is to make clearer the degree to which historical trends in the U.S. military’s force structure and modernization plans are largely the result of policy and programmatic choices made by Department of Defense (DoD) and service leadership. Contrary to a widely held belief, notwithstanding the influential role played by Congress and some other key actors, the size and shape of today’s forces are not simply a byproduct of budgetary or other pressures beyond DoD’s control. For good or ill, the trends described in this report largely reflect tradeoffs made by senior U.S. defense leadership. Among other things, this conclusion suggests that to the extent there are concerns about the current U.S. approach, more than anything else, charting a different course in the future will require a shift in the decisionmaking of that leadership.
Overview of Trends in Force Structure and Modernization
Measured in terms of personnel and major weapons platforms, the size of the U.S. military has been on a generally downward trajectory for decades. The path has not been simple, uniform, or smooth, and there have been important exceptions in certain areas. Nevertheless, the overall trend is unmistakable and, viewed from a long-term perspective, quite consistent. In the mid-1950s, after the Korean War drawdown the U.S. military consisted of some 2.9 million active-duty troops. By 1975, after the Vietnam War drawdown it stood at about 2.1 million. After the end of the Cold War, it fell to some 1.4 million troops. And today, the U.S. military is manned by some 1.3 million active-duty service members (see Figure 1).
Figure 1: Active-Duty End Strength, 1955-2016
Trends in weapons inventories paint a more complex picture, depending upon the particular force structure elements considered. For example, the overall number of battle force ships in the U.S. Navy has declined dramatically over time – from over 1,000 ships in 1955 to some 560 ships in 1975 and about 270 ships today. However, the size of the carrier force – still very much the core of the U.S. Navy – has been reduced at a more measured pace, falling from 15 in 1975 to 10 today.1 Similarly, while the number of Air Force fighter and attack aircraft has been cut significantly since the mid-1980s, falling from some 4,400 in 1985 to 2,500 by 2000, and to 2,000 today, the Air Force’s fleets of transport and intelligence, surveillance, and reconnaissance (ISR) and related aircraft have been cut much more modestly – respectively, by about one-third and one-quarter since the 1980s. Despite this more complex pattern, however, the overall trend in numbers has clearly been consistently toward a smaller force, whether measured in terms of military personnel or major weapons platforms.
Paralleling this downward trend have been trends in both the frequency with which new weapons are introduced and in the aging of the military’s inventory of major weapons platforms, reflecting a significant decline in weapons production rates. There has been even more variability across services in the aging of the weapons inventory than in cuts to force structure. Notably, the Air Force has witnessed a far more severe aging of its weapons inventory over the past several decades than have the other services. The impact on the Air Force can be seen especially in its combat aircraft and tanker fleets. In 1980, the average age of the Air Force’s fighter inventory was under 10 years;3 today the average age stands at 24 years.4 Likewise, over this same period the average age of the Air Force’s bomber force has increased from under 20 years to 39 years, and its tanker fleet from about 20 years to 38 years.5 But among the other services and in other areas, the trends in weapon systems aging have generally been substantially less significant. The average age of the Navy’s inventory of battle force ships, for example, is now about 17 years, up from 13 years in 1980, but roughly the same age as it was in the late 1980s, when it approached the Reagan-era goal of a 600-ship Navy. In the case of Navy fighter aircraft, as well as Army ground combat vehicles and helicopters and Air Force airlift aircraft, the trends in aging have been similarly more modest than for Air Force combat aircraft and tankers. Nevertheless, the overall trend has clearly been toward an older weapons inventory for the U.S. military.
Similarly, the pace at which successive generations of new weapons platforms are developed and deployed has, over time, slowed significantly. This trend has, again, probably been most significant in the Air Force, particularly among combat aircraft. Between 1946 and 1965, the Air Force deployed 15 different types of fighter and attack aircraft. By comparison, between 1966 and 1985 it introduced only five new aircraft of these types. And in the roughly 30 years since, it has introduced only two new designs – the F-22 and the F-35.
A Smaller and Older Force: What Are the Main Drivers of These Trends?
What have been the major drivers of the trends toward a smaller and older force? At the broadest level, it is possible to identify four major contributors. These are related to spending per troop on both military personnel and operations and maintenance (O&M) activities, and spending on both the development and procurement (i.e., production) of new weapon systems. Though easy to identify, as discussed in the next section of this report, understanding the policy choices and pressures that underlie trends in these areas is a substantially more difficult and less straightforward task. It is possible, however, to dismiss the two factors that are typically thought to be the most significant contributors. Despite widely held misconceptions to the contrary, neither a declining defense budget top line nor declining resources for weapons acquisition explain the trends in the size of the U.S. military or the aging of the weapons inventory. This is for a simple reason: Viewed from a long-term perspective, neither the defense budget top line nor overall acquisition funding has declined.6
If anything, over time, the defense budget top line has grown modestly in real (inflation-adjusted) terms. To be sure, the growth has been uneven and somewhat cyclical, with significant peaks and troughs. But a modest upward trend, with a significant cumulative effect, is nevertheless identifiable (see Figure 2). And this is true even when the cost of military operations – currently funded though the Overseas Contingency Operations (OCO) account – is removed from the data. The “base” defense budget (i.e., the defense budget exclusive of war costs) has averaged about $497 billion (Fiscal Year 2017 dollars7) annually over the past 20 years. This is actually about 10 percent more in real (inflation-adjusted) terms than provided in defense budgets over the preceding 20 years (1977–1996) and 76 percent more than provided in the 20-year period before that (1957–1976).8 So it would be difficult to blame the trends in force size or weapons age on changes in the defense budget top line.
Figure 2: Department of Defense Funding9
In the case of acquisition funding – including both research and development (R&D) and procurement spending – the long-term trend has been essentially flat. Even more so than the defense budget top line, acquisition funding has witnessed some fairly dramatic swings over time. But viewed from a long-term perspective (and given the length of time weapons are typically kept in the inventory, it is especially critical to take a long-term perspective in this case), acquisition funding has remained relatively stable. Exclusive of war-related costs, acquisition funding has averaged about $168 billion annually over the past two decades; this is only 3 percent below the average of the preceding 20-year period.10 And it is more than one-third higher than funding during the 1957–76 period – even including Vietnam War-related acquisition funding. As such, it would be difficult to blame the trends toward the less frequent introduction of new weapon systems and lower weapons production rates on changes in acquisition funding.
Rather than changes in either the defense budget top line or overall acquisition funding, what explains the trends in the size of the U.S. military, as well as the slower introduction of new weapon systems and the age of its weapons inventory, are funding trends within the operations and support (O&S) and acquisition portions of the defense budget. Specifically, as discussed in more detail below, these trends have been driven primarily by relatively consistent and significant growth in spending per troop on O&S activities – both military personnel and O&M – and even more dramatic growth in both spending on the development of individual weapon systems and the unit procurement costs of those weapons.
Cost Growth in Operations and Support Activities
Adjusted for changes in the size of the U.S. military, spending on O&S activities – funded primarily through the military personnel and O&M accounts – has been on a consistently upward path since at least the 1950s. An imperfect but useful way to measure this trend is to consider changes in O&S funding – exclusive of war-related costs – per active-duty service member over time. By that measure, funding grew from about $50,000 per troop in 1955 to $111,000 in 1980 and $184,000 by 2000. Today O&S spending per troop stands at about $266,000. In other words, O&S spending has over time typically grown at an average annual rate of some 2 percent to 3 percent per year (see Figure 3).
Figure 3: Operations & Support (O&S) Funding Per Service Member11
O&S programs and activities include everything from military and civilian pay to funding for health care and other benefits, food, fuel, utilities, most spare parts, and other supplies. Increases in compensation both for military and civilian DoD personnel have played a significant part in the growth in O&S spending. Although the specific rate of growth has varied during different periods, viewed from a long-term perspective, both groups have received substantial increases, with the growth significantly exceeding the overall inflation rate. Funding for military personnel, which covers the cost of military pay, housing and subsistence allowances, retirement pensions, and most (but not all) other forms of military compensation outside of health care, has grown from about $52,000 per active-duty service member in 1980 to about $111,000 in 2016. That works out to an average annual growth rate of some 2.5 percent. Pay for civilian DoD personnel has increased more slowly. However, here too the long-term trend has involved significant growth. Between 1980 and 2016, for example, spending per civilian DoD employee grew from an average of some $59,000 to $93,000 per year.12 Because compensation (exclusive of military health care) accounts for well over half of O&S spending, this growth has been the biggest driver of the overall increase in per capita O&S spending.
Another significant contributor to the growth in O&S spending has been the rise in health care expenditures. Although representing a far smaller part of overall O&S spending than pay, health care spending – an important component of military compensation – has, on a per capita basis, grown much faster than either military or civilian pay. Since 1980, spending on military health care has nearly quadrupled, reaching some $48 billion in 2016.13 This equates to an annual growth of 5 percent to 6 percent per active-duty service member.14 To be sure, if the total number of beneficiaries, rather than the number of active-duty personnel, is used as the basis for the calculation, the growth rate has been lower – reflecting the fact that while the number of active-duty service members has declined by about one-third during this period, the total number of beneficiaries (which also includes dependents and retirees) has actually increased modestly, from about 9 million to 9.4 million. But that still implies a robust annual growth rate of some 4 percent per capita.
Part of this growth is the result of the same factors that have caused per capita health care costs to generally rise faster than the overall inflation rate in the civilian economy (e.g., higher costs associated with the introduction of new medical technologies). But various benefit expansions have also accounted for a significant share of this growth. Particularly costly was the establishment in FY 2002 of the TRICARE for Life program, which made TRICARE a “second payer,” covering any costs not covered by Medicare, for military retirees 65 and older. Taken together, growth in military and civilian pay and the rise in military health care costs have accounted for about two-thirds of the increase in O&S spending since 1980.
The remaining roughly one-third of the O&S budget is allocated to a broad range of other programs and activities supported through purchases of goods and services from the private sector. This includes everything from weapons and facilities maintenance and repair to the provision of fuel, some spare parts, and other consumables, as well as spending on utilities, information technology, and other support. As with military health care, spending in this category has grown especially rapidly and has substantially exceeded the growth rate in military and civilian pay.
The increased O&S spending per active-duty troop described above was not inevitable or preordained. To be sure, pressures caused by health care inflation, the need to compete with growth in private-sector compensation, and other factors to some extent outside of DoD’s control created powerful incentives for DoD to increase spending per troop. And in the case of pay and other benefits, in particular, Congress and outside interest groups have frequently played a significant part in pushing spending upward. But ultimately, as much as anything else, the increased spending on O&S activities has reflected a policy choice. It is a choice that rests on the assumption that increasing pay and other O&S spending per troop is the best way to maintain or improve the effectiveness of the military’s overall combat capabilities – even if that means reductions in the size of the force must be made to offset those higher per capita costs.
In the case of military and civilian personnel, DoD and service leadership have consistently taken the view that maintaining or improving quality is generally more important than sustaining numbers of personnel. This has been especially true since the establishment of the All-Volunteer Force (AVF) and the end of the draft in the 1970s. That the quality of the U.S. military’s workforce – both uniformed military and civilian – has, indeed, improved significantly over time is clear. Among other things, this can be seen through changes in education, aptitude, and level of experience. DoD defines recruits who have both graduated from high school and score above the median on the Armed Forces Qualification Test (AFQT) as “high-quality” recruits.15 The share of such recruits fell from 43 percent in 1973 (when the draft was ended) to only 28 percent in 1977. However, thanks in part to a series of significant increases in compensation and other benefits, the numbers improved from the late 1970s through the late 1980s and had reached 60 percent by 1990. Since then, with the exception of the Army during a few years around 2000 and the peak years of the wars in Iraq and Afghanistan, the services have generally been able to maintain the share of recruits in this category at between 55 percent and 65 percent. Over time, the U.S. military has also become significantly older and more experienced. In 1969, only 18 percent of Army enlisted personnel had more than four years of service. By 1977 that share had increased to 37 percent, and by 2000 it had reached 50 percent.16 This same general trend is also reflected in the experience of the other services, as well as among the officer corps.
And to the extent that significant improvements in personnel quality were achieved, increases in pay, as well as the sustainment – and in some cases expansion – of relatively generous medical and other benefits, clearly played a part in achieving that result. Such increases also certainly played a part in helping to sustain the force during wartime. The wars in Iraq and Afghanistan placed tremendous stress on the U.S. military, particularly the Army and Marine Corps. In part due to the significant pay raises provided over much of the past decade and a half, personnel quality was nevertheless maintained to a remarkable degree. Likewise, thanks in part to compensation increases, DoD has also been able to support a civilian workforce that has generally grown more professional and skilled over time.17
It is more difficult to correlate increases in O&S spending over time with improvements in areas other than personnel quality – such as unit readiness, equipment mission-capable rates, flying hours, steaming days, and other readiness indicators.18 However, certainly DoD and the services believe that the funding increases in O&S, and particularly O&M, provided over the decades have been critical to sustaining readiness as measured by such indicators.
Since, as noted earlier, the long-term trend in the defense budget top line has been upward, but only relatively modestly so, and acquisition funding has remained relatively flat, the only possible way to accommodate this relatively consistent and enduring long-term increase in per capita O&S spending – both for military personnel and O&M activities – has been to reduce the size of the U.S. military. And that is the long-term tradeoff the U.S. military has consistently made. Overall funding for O&S activities is about $341 billion today, 19 percent higher than it was in 1990, at the end of the Cold War, and more than double what it was in 1964 before the start of the Vietnam War. But the U.S. military today, measured in terms of active-duty service members, is 37 percent smaller than it was in 1990 and less than half as large as it was in 1964. Although in a world of less constrained resources many in DoD would certainly have preferred to fund both high readiness levels and a larger force structure, given the choice between a smaller and ready force and a larger but to some extent “hollow” force, few would have chosen the latter.
Cost Growth in Weapons Acquisition
Historically, successive generations of weapon systems have tended to cost far more to acquire than the systems they are intended to replace. This intergenerational cost growth has, if anything, been even more consistent and, in many instances, more dramatic than in the case of O&S cost growth. This trend is clearly illustrated with the F-35 fighter, which with total acquisition costs of about $350 billion is far and away the Pentagon’s largest and most costly acquisition program. The F-35 is projected to cost about $64 billion to develop and have a unit procurement cost averaging some $110 million among the three variants of the aircraft being produced – for the Air Force, Navy and Marine Corps.19 By comparison, the aircraft the F-35 is intended to replace had total development costs of about $21 billion and unit procurement costs ranging from about $16 million (for the A-10) to $64 million (for the F/A-18).20 This kind of cost growth is typical among major acquisition programs (see Figure 4).
Figure 4: Unit Procurement Costs of Fighter and Attack Aircraft21
Even more so than in the case of O&S spending, it is important to recognize that this growth in the cost of developing and procuring next-generation weapon systems is not something imposed on Pentagon planners. There is no iron rule of nature mandating that successive generations of weapon systems cost so much more to acquire than the systems they are intended to replace. The Pentagon could – and occasionally does – buy next-generation systems that do not manifest such significant cost growth. Alternatively, it sometimes acquires entirely new types of weapon systems (e.g., unmanned aircraft) to perform certain missions. But for the most part, military planners have shown a very enduring preference for the acquisition of next-generation systems that are both far more capable and far more costly than their predecessors, even though – in the context of an overall acquisition budget that has remained relatively flat – this clearly means that next-generation systems must be introduced less frequently and procured in smaller numbers. As noted earlier, the time separating the introduction of successive generations of weapon systems has grown substantially over the past several decades, while procurement rates for many weapon systems have likewise declined significantly.
Along with the growth in O&S costs discussed earlier, these lower production numbers have contributed to the decision by DoD to accept cuts in the size of the U.S. military over time. However, the impact of lower production numbers on force size has been mitigated to some extent by another long-term feature of the U.S. military’s approach to modernization: longevity. The flip side of its preference for pursuing the acquisition of very capable and costly next-generation weapon systems is a willingness to keep current-generation systems in the force longer. This willingness is at least implicitly based in part on a belief that, as platform technologies have matured over time, the pace at which they must be replaced has slackened and in part on the view that the effective service lives of older platforms can be extended through the exploitation of advances in precision strike, ISR and related technologies.
The U.S. Approach to Shaping, Sizing, and Modernizing its Forces
To summarize, the persistent trend toward a U.S. military that is smaller, generally equipped with older weapons platforms, and less frequently infused with next-generation weapon systems reflects several relative constants in DoD policymaking and planning:
- Support for the acquisition of increasingly costly and more capable next-generation weapon systems, as well as higher spending on personnel and other operations and support activities.
- A willingness, as these costs grow, to trade off quantity for quality in terms of both weapons platforms.
- And (partially mitigating the need to make the above tradeoffs) a belief that, as platform technologies have matured over time, the pace at which they must be replaced has slackened, and faith in the potential for extending the effective service lives of older platforms through the exploitation of advances in precision strike, ISR, and related technologies.
Absent these three policy choices and assumptions, the rapid growth in development and unit procurement costs, as well as O&S costs per troop, experienced since the 1950s could not have been reconciled with the significant but slower rate of growth that occurred in the defense budget top line over those same years. The next question we turn to concerns the effectiveness of this approach, both in principle and in execution. Put bluntly, should it matter that U.S. military forces have been getting smaller and older, and if so, what can and should the United States do about it?
The full report is available online.
- The decline is projected to be temporary. Under current plans, the number of aircraft carriers will increase to 11 once the USS Gerald Ford, the first of a new class of aircraft carrier, is deployed. ↩
- 258-260, Table 7-5, “Department of Defense Manpower.” ↩
- All estimates for weapon systems for 2009 and prior years in this report are from Congressional Budget Office, The Long-Term Implications of the Fiscal Year 2009 Future Years Defense Program: Detailed Update, January 6, 2009, 28. ↩
- “The Air Force in Facts and Figures,” Air Force Magazine (May 2016), 30. ↩
- Ibid. ↩
- For a discussion of trends in the defense budget, see Todd Harrison, “Analysis of the FY 2017 Defense Budget” (Center for Strategic and International Studies, April 2016). ↩
- Unless otherwise noted, all spending and cost figures cited in this report are expressed in Fiscal Year 2017 dollars. ↩
- Unless otherwise noted, all changes in spending or costs cited in this report are expressed in real terms. ↩
- Department of Defense, National Defense Budget Estimates for FY 2017 (March 2016), Table 10-1, “Gross Domestic Product and Deflators Used in the Historical Tables, 1940-2021,” in Office of Management and Budget, Fiscal Year 2017 Budget of the United States, Historical Tables. ↩
- Adjusted for the substantial amount of procurement funding that has been provided through the OCO budget that is intended essentially to meet base (rather than war-related) budget requirements, it is likely that even this marginal reduction would disappear. ↩
- Derived from Table 7-5, “Department of Defense Manpower,” 258–260, Table 6-8, “Department of Defense Budget Authority by Public Law Title,” 133–139, and Table 2-1, “Base Budget, War Funding, and Supplementals by Military Department, By Public Law Title,” 22–31, in Department of Defense, National Defense Budget Estimates for FY 2017 (March 2016). Converted into FY 2017 dollars using Office of Management and Budget deflators from Table 10-1, “Gross Domestic Product and Deflators Used in the Historical Tables, 1940-2021,” in Office of Management and Budget, Fiscal Year 2017 Budget of the United States, Historical Tables. ↩
- The cost of pay and other benefits (e.g., health care) for civilian DoD employees is covered primarily through the O&M budget. ↩
- Congressional Budget Office, Long-Term Implications of the 2016 Future Years Defense Program, CBO 51050 (January 2016), 28. Excluding the military and civilian DoD pay components of military health care, spending has grown from about $6 billion to $32 billion. ↩
- The military health care system provides care not only to active-duty service members but also their family members, as well as military retirees and their family members. Altogether, more than 9 million individuals are eligible for health care through that system. Since, unlike the number of active-duty service members, the total number of eligible beneficiaries has actually increased modestly in recent years, on a per beneficiary basis the rate of cost growth, while still significant, would be substantially lower than the cost growth per active-duty service member cited above. ↩
- The AFQT assesses basic verbal and mathematical abilities. It is benchmarked against the 18 to 23-year-old civilian population. ↩
- Bernard Rostker, “I Want You!: The Evolution of the All-Volunteer Force” (Rand Corp., 2006), 8. ↩
- For a discussion of trends in the civilian workforce and DoD and other federal agencies, see Government Accountability Office, Federal Workforce: Recent Trends in Federal Civilian Employment and Compensation, GAO-14-215 (January 2014). ↩
- For an excellent discussion of trends and issues related to military readiness, see Todd Harrison, “Rethinking Readiness,” Strategic Studies Quarterly (Fall 2014), 38–68. ↩
- These estimates, converted into FY 2017 dollars, are taken from Jeremiah Gertler, “F-35 Joint Strike Fighter (JSF) Program,” RL30563 (Congressional Research Service, April 29, 2014), 14. ↩
- Author’s estimates based on DoD and Congressional Budget Office data. ↩
- Author’s estimates based on Department of Defense and Congressional Budget Office data from Congressional Budget Office, A Look at Tomorrow’s Tactical Air Forces (January 1997), 37, Figure 7, “Unit Procurement Cost for Fighter and Attack Aircraft”; Department of Defense, F-22, Selected Acquisition Report, December 31, 2010, 12-13; Department of Defense, F-35 Joint Strike Fighter Aircraft, Selected Acquisition Report, December 2015, 24–67; and Department of Defense, F/A-18E/F, Selected Acquisition Report, December 31, 2011, 13–19. Converted into FY 2017 dollars using Office of Management and Budget deflators from Table 10-1, “Gross Domestic Product and Deflators Used in the Historical Tables, 1940-2021,” in Office of Management and Budget, Fiscal Year 2017 Budget of the United States, Historical Tables. ↩
More from CNAS
Why There Are No Game-Changing Weapons for Ukraine
Germany has become the second-biggest contributor of military aid to Ukraine after the United States, but you wouldn’t know it by following the debate in Berlin. In a replay o...
By Franz-Stefan Gady
To Avoid AI Catastrophes, We Must Think Smaller
These incidents are not theoretical, nor are they projections of long-term dangers; rather, these AI tools are already presenting tangible threats to individual health and wel...
By Josh Wallin
For Replicator to Work, the Pentagon Needs to Directly Help with Production
Today’s innovation ecosystem alone cannot achieve the necessary production scale, especially for the less commercially viable classes of systems relevant in the Indo-Pacific....
By Andrew Metrick
Sharper: Campaigning and the National Defense Strategy
The United States faces the unprecedented challenge of simultaneously deterring large-scale conventional aggression by two nuclear-armed powers while also managing other threa...
By Philip Sheers, Molly Campbell & Anna Pederson