Photo: Andrew Renneisen/Getty Images
The United States has now been continuously at war for more than seventeen years. It is still fighting an active war in Afghanistan, has yet to fully defeat ISIS in Syria and Iraq – much less establish a state of lasting security in either country – and is playing a role in low level conflicts against extremist and terrorists in many other parts of the world.
The U.S. government, however, has never developed a convincing method of reporting on the cost of the wars, and its estimates are a confusing morass of different and conflicting Departmental, Agency, and other government reporting that leave major gaps in key areas during FY2001-FY2019.
It has never provided useful forecasts of future cost, instead providing empty "placeholder" numbers or none. It has failed to find any useful way to tie the cost estimates it does release to its level of military and civil activity in each conflict or found any way to measure the effectiveness of its expenditures or tie them to a credible strategy to achieve some form of victory.
The result is a national embarrassment and a fundamental failure by the Executive Branch and Congress to produce the transparency and public debate and review that are key elements of a responsible government and democracy.
The Burke Chair at CSIS is circulating a working draft of a report analyzing official reporting on the Cost of War since FY2001. It examines these issue in detail for the period from FY2001-FY2023, and is entitled America’s Military Spending and the Uncertain Costs of its Wars:
The Need for Transparent Reporting . It is available on the CSIS web site at https://csis-prod.s3.amazonaws.com/s3fs-public/publication/190123_Cost_of_War.pdf.
This report presents a range of total costs that range from lows of $2,002.4 billion to $2,106.2 billion for government reported war costs between FY2001 and FY2019. However, there are a wide range of outside estimates that include other major costs. One well-structured such estimate reaches a total of $5.9 trillion spent and obligated over the same period. As is examined later, the Department of Defense and other Departments and agencies have never provided credible estimates of futures pending, but the same outside estimate of war costs projects a total of $6,741 billion through end-FY2023.
It is clear that the building blocks dealing with defense costs have major problems and are not tied to useful descriptions of the activities they cost, the strategy involved, or the effectiveness of such spending. The other building blocks dealing with State Department/USAID spending, Veterans administration spending, spending by the CIA and other agencies – and that can be used to address issues like the impact of war on the national debate and interest – either have serious gaps and problems or are not addressed in an official form.
If anything, the United States has steadily cut back on the level of transparency it provides on its military and civil activity and has not developed any meaningful reporting on how the total cost of its wars are tied to well defined military and civil efforts, their effectiveness, and some clear strategy to terminate and "win" America's current conflicts. The U.S. government should not repeat these failures again in the course of debating the FY2020 budget, its future strategy, and the levels of force and civil aid it should continue to deploy.
There is a clear need for better reporting on the cost of America's wars. Such reporting should be unclassified, fully defined, and explain what costs are and are not included. It should tie such spending to its effectiveness, and to an estimate of whether U.S. strategy, plans, and budgets can end in some form of meaningful victory. The executive branch should provide regular reporting on the cost and level of success for each war, and the Congress should openly debate both the rising cost of America's wars and the adequacy of U.S. strategy and operations
The Department of Defense provides reporting on Overseas Contingency Operations (OCO), but these cost and budget data do not equal its own estimates of the total cost of its wars in its quarterly Cost of War report – a document that is unclassified but not even listed as a publication on its website. DoD budget requests have not matched its real spending in recent years, and its outyear estimates for the four future years in its annual Future Year Defense Programs (FYDPs) have long been empty nominal "placeholders" with no substantive meaning.
The Department of Defense has also steadily cut back on the supporting data it provides on U.S. operations and strategy. It has removed large amounts of reporting from its publications web page. It no longer provides reliable data on the number of U.S. military actually deployed on a permanent and temporary basis, and Inspector General reports show that it has classified much of the key data on its efforts to train and equip its strategic partners in Afghanistan, Iraq, and Syria.
As noted earlier, the report shows that U.S. State Department has provided virtually no meaningful recent unclassified reporting on its total level of war-related cost and activity –although the Lead Inspector General for Overseas Contingency Operations has for at least FY2019. The State Department budget justification document does not provide any basis for know what plans exist to deal with the aid and other efforts necessary to deal with the civil side of U.S. wars.
This is why credible estimates of total civil-military cost of its wars from FY2001-FY2019 can range from total OCO spending of some $1.8 billion to 1.9 billion to outside estimates of $5.9 billion or more. The methodology used in some earlier outside estimates would also bring this total cost to much higher levels, reaching well over $8 trillion by 2050.
There seems to be little meaningful open debate over such cost estimates, and what to cost, and how to tie such spending to its effectiveness costs by the key Congressional committees with jurisdiction over such spending. As the report shows, there are a variety of conflicting OMB, CBO, or GAO reports on given aspects of such costs but each has critical limitations, and all fail to report on the effectiveness of such expenditures. There are no official reports or studies outside of the Departments of Defense and State that link the level of spending on a given war to a clear strategy or estimate of such spending to the strategy employed, the force levels and military operations taking place, and the cost-effectiveness of U.S. spending.
The closest the U.S. government comes to meaningful reporting and transparency by war, and by major level of activity, are the reports of the Special Inspector General for Afghan Reconstruction (SIGAR) and the Lead Inspector General for Overseas Contingency Operations. These reports, however, focus largely on civil and military aid to strategic partners, and these inspector generals are not charged with assessing how the cost of wars impact the overall level of U.S. federal and national security spending, the national debt and interest payments, the cost-effectiveness of U.S. military and civil activity, or how the costs link to a credible strategy.
The Burke Chair has reported previously on these problems in a report entitled U.S. Military Spending: The Cost of Wars. This report is still available on the CSIS web site at https://www.csis.org/analysis/us-military-spending-cost-wars. This new report updates that earlier report and provides substantial additional data. These new data provide substantial further evidence that the U.S. government is failing to meet its responsibility to report meaningfully on America's wars.
At the same time, this report also indicates that some outside efforts to estimate the total cost of U.S. conflicts may sharply overestimate total costs, and the impact of current war costs on the federal budget and U.S. economy. Taking cost data out of context may serve given political goals – either to call for more government funding or to oppose any form of military action and/or defense spending. They end result, however, far more s misleading than the morass of different estimates the U.S. government does issue.
Whatever the merits of U.S. warfighting expenditures since the U.S. first intervened in Afghanistan in 2001 may be, these costs have not dominated U.S. national security spending, sharply altered the overall cost of the federal budget, or had a major impact on the burden that U.S. defense spending puts on the U.S. economy.
U.S. Wartime Spending Since FY200 has been a Small Share of Total U.S. Defense Spending
The sections that follow raise many serious questions about the way the Department of Defense costs the military side of U.S. wars. They a make it clear that there are no convincing estimates of spending on the civil side of current conflicts and the impact of U.S. wars on future military and civil pensions and conflict-related medical costs.
It is still clear from these analyses, however, that the cost of U.S. wars has not dominated U.S. defense spending, or the burden that U.S. national security spending places on the U.S. federal budget and economy.
Total U.S. Defense Spending Dominates World Spending, But U.S. War Costs Are Only a Limited Part of that Spending
The relative size of U.S. national security spending can only be "guesstimated." There are no clear comparisons of total U.S. military spending and the spending of other states. U.S. estimates are particularly difficult to compare because the U.S. does not use a calendar year for its fiscal year. The U.S. fiscal year is the accounting period for the federal government which begins on October 1 and ends on September 30. The fiscal year is designated by the calendar year in which it ends; for example, fiscal year 2020 begins on October 1, 2019 and ends on September 30, 2020.
As Figure 2 has already illustrated, the U.S. also estimates its budget using two different levels of spending. Most U.S. budget analysis focuses on estimates of BA or Budget Authority – which is how much money Congress allows a federal agency to commit to spend in a given year, although the actual spending may occur years later as a project, weapon, or construction effort is completed.
At the same time, the U.S. reports a substantially lower spending figure called BO or Budget Outlays. BO estimates how much money will actually flow out of the federal Treasury in a given year. While BA shows total cost of a given fiscal year’s authorization by Congress, BO estimates actual cash flow in that fiscal year, and is used to determine the size of the overall deficit or surplus, and whether budget is balanced. Neither BA or BO report actual U.S. government spending in a given year since actual spending can lag obligation by a significant period, but both are more useful in that they allow the government to act on the estimate before the money is actually spent.
These issues in U.S. budget reporting, however, are only one of many factors that make it difficult to do more than put U.S. war costs in the broader context of how U.S. pending compares to the spending of other states. Virtually every country has somewhat different ways of reporting military and national security spending.
UN and international agency reporting reflect many different ways of defining such expenditures, and has many gaps. The annual reports by the Secretary General of NATO make it clear that even the relatively cohesive mix of countries in the NATO alliance does not allow the NATO international staff to ensure that its reporting produces directly comparable data for its member countries.
The various international research centers that attempt such comparisons – the International Institute of Strategic Studies (IISS) and the Stockholm International Peace Research Institute (SIPRI) make it clear that individual countries estimate their military and national security costs in very different ways, may or may not include paramilitary and other related security costs, treat pension and medical costs differently, and report differently on arms imports, modernization, and nuclear forces.
Countries also have very different force structures and force costs. For example, conscription radically lowers the personnel costs for some countries while power projection outside national boundaries sharply raises the operating costs for others. Waste and corruption consume large amounts of the national security expenditures of some states, as does spending on prestige weapons, systems, and forces that do not have a high priority in terms of deterrent and warfighting value. In practice, the market forces that drive national security spending do not produce comparable levels of efficiency or strategic focus, and there are no credible net assessments of how well given countries allocate their spending.
The key point from a policy, planning, and budgeting perspective is that a focus on U.S. war costs – while important for many other reasons – is largely irrelevant in comparing total national defense efforts, strategic needs and trends, and force size. These are the key priorities that shape U.S. national security planning. Moreover, it is total national security spending – not spending on limited wars – that is the key measure of the resources that shape a nation's military power and the burden it bears in funding such efforts.
The Declining Military Burden on the U.S. Economy in Spite of Spending on Current Wars
It is equally important to put the impact of U.S total defense spending and wartime spending on the U.S. economy and total federal budget into proper perspective. Focusing on either total defense spending or wartime costs without looking at relative impact of either set of costs can create a highly misleading picture of the burden they impose.
U.S. spending war costs, defense, and international relations does involve major sacrifices in terms of what the U.S can spend in other areas. However, some reporting that focuses on worst-case estimates of defense spending and war costs implies America's current wars drive its national debt and deficits, cripple its economy and have created major new highs in federal spending. There are legitimate cost models that warn that the conventional methods of budget reporting may sharply undercost America's current wars.
The charts and graphs in this section show, however, that total U.S. defense spending is now at a near record low in terms of the burden it imposes on the economy. It also shows that the CBO estimates that the burden of U.S. national security efforts will drop steadily through 2028, while mandatory spending on healthcare and Social Security will drive the increase in federal spending, the deficit, and the cost of the interest that must be paid on the national debt.
Even if one throws in the total cost of Veteran's Administration activities (rising from 2.6% to 4.5% – and all civil expenditures by DoD, like the efforts of the Corps of Engineers (1.6% to 2.0%) – total national security spending including all foreign affairs spending would go from 21.4% of the total federal budget in FY2000 to a wartime peak of 26.4% in FY2008, and drop back to 23% in FY2019.
The total direct cost of America's wars since FY2000 has also imposed a far lower burden than total national security spending – since they totaled only 10%-24% of U.S. total defense spending during FY2001-FY2019. Whatever the real cost of America's wars may be, their cost per se does not impose a major burden on either the budget or the economy. The key issue is rather whether such spending is justified in terms of the benefits these war offer in terms of added U.S. national security, and alternative uses of a comparatively limited amount of annual federal expenditures.
As is mentioned throughout this analysis, there are no credible U.S. government projections of future war costs, military budgets, and total national security costs. The DoD War Cost report to Congress only covers past spending. The future OCO costs reported in the DoD budget request for FY2020- FY2023 – a document issued in February 2018 – does not reflect any real-world form of Future Year Defense Program (FYDP). It only shows OMB approved "placeholder" costs that have no relation to actual plans.
A review of past estimates of future or "outyear" data indicate that the Greenbook data that provides an alternative to "placeholder" data have never been useful in previous years because of shifts in U.S. strategy, plans, and operations. As was the case in Vietnam, every Fiscal Year leads to changes that make it the "first year" in a new mix of warfighting efforts.
This, however, is only part of the problem. At least since the FY2000, the U.S. has made serious cuts in readiness and in force size and modernization affecting Baseline forces and missions. The U.S. does not, however, issue estimates of the cost of bringing Baseline forces back to full readiness – although the Secretary of Defense, the military Chiefs of Staff, and the House and Senate Armed Services Committees have all noted the growing seriousness of the problem between FY2001and FY2019. The November 2018 assessment of the National Defense Strategy Commission in its report on Providing for the Common Defense also notes the scope of the problem.
Even more importantly, the U.S. no longer attempts to define its strategies, force plans, or programs in terms of detailed cost estimates and budget projections. Its Quadrennial Defense Reviews define concepts, not plans, programs and budgets. The same is true of the newNational Security Strategy and the new National Defense Strategy that the Trump Administration issued in 2017 and 2018. Both called for a major new effort to focus on the potential threat posed by China, Russia, North Korea, and Iran but did not provide any projections of future costs.
Further, the CBO and GAO do not project the cost of bringing Baseline forces to proper readiness, strength, and modernization levels. Equally important, the CBO does not provide any contingency projections of future war costs for new contingencies, and focuses almost exclusively on options for cutting defense and international affairs spending to reduce the deficit and debt– rather than the planned and potential cost of meeting national security needs.
The end result is that a substantial part of the projected decline in national security spending relative to mandatory spending ignores how suddenly and drastically defense and war costs can change. Similarly, Mandatory spending may be somewhat predictable, but the U.S. scarcely lives in a risk free world.
The fact that America's wars do not heavily impact its economy, dominate its federal budget, or act as the key forces driving its debt and the resulting interest payments does not mean, however, that these costs do not deserve close public examination and debate. It does not mean that major uncertainties do not arise in the current reporting of these costs, over how they should be defined, and assessed to determine their longer-term impact.
Official U.S. government reporting on its war costs since FY2001 is questionable and non-existent in all three areas. The most critical such issue that the Executive Branch and the Congress have failed to address is what to count in providing a consistent way totaling the total cost of America's wars overtime, and what to include in such costs.
The preceding analysis has focused on one aspect of such costs: how Department of Defense (DoD) war costs impact total defense expenditures and the broader impact on the federal budget and on the economy. It has also noted, however, that any honest estimate of the direct cost of U.S. wars must include the cost of Department of State (DoS) and USAID civil programs, and the directly related cost of combat-related Veterans Administration spending. It must also include all military and civil intelligence costs, the costs of other agencies, and address whether debt and interest costs should be included.
There is no consistent official unclassified estimate of such costs, and particularly estimates that address the costs to Departments and Agencies outside the Department of Defense. The closest thing to official U.S. government estimates that have attempted to provide such cost estimates over time have been the work of the Congressional Research Service, which uses publicly available data. This CRS reporting has been highly useful in spotlighting the costs of war, but it has not been updated to cover the costs of U.S. wars since FY2016/FY2017, and only the FY2001-FY2014/FY2015 estimate seems to have attempted to fully address Veterans costs.
This CRS reporting also only covered direct military and civil Overseas Contingency Operations (OCO) expenditures – which CBO, Department of Defense, and Lead Inspector General reporting all show do not include the full costs of war. As for the State Department and USAID, it is far from clear what costs should be included. U.S. aid to Pakistan, for example, probably should be because it was directly linked to Pakistani support for the Afghan war. The same is true of other U.S. aid efforts in Central Asia, and to some of the nations that contributed force to the military efforts in Afghanistan. The costs of expansions to U.S. embassies and consulates, and added personnel, security, and transport should also be included along with a careful study of other support and contractor costs and transfers to outside aid and NGO efforts. The present open source cost data are far too vague to be trustworthy.
Even the CRS reporting through FY2014/FY2015 only seems to have included part of the related Veterans Administration costs, did not cover full intelligence agency costs, and did not seem smaller expenditures by other federal departments and agencies.
As is touched upon in more depth later in this report in analyzing the DoD Total Cost of War report, and OSD Comptroller reports on OCO costs, additional issues arise in the CRS estimates because some of the expenditures reported as OCO costs seem to actually be the cost DoD Baseline activities that were put into the OCO account to avoid the budget caps imposed by the Budget Control Act (OCO costs were immune to these caps). The DoD reporting of Afghan war costs seem particularly suspect in this regard, but only an independent audit by an Inspector General, GAO, or CBO could resolve this uncertainty and no such audit has taken place.
More broadly, a variety of outside studies have suggested that reporting on the total cost of U.S. wars should also include the full medical costs of those wounded and injured in the wars, pension costs, all homeland defense spending, and the impact of wartime spending on the national debt. These arguments deserve official attention, and one of the best such estimates, done by Neta C. Crawford of the Watson Institute at Brown University, is analyzed in detail in this report to illustrate the issues involved, and the need for a full official investigation of such costs and how they should be assigned in costing America's wars.
An outside estimate by Neta Crawford – which is discussed in detail in the next section – puts directly comparable total DoD and State/USAID costs at $2.022 billon through FY2019, and the cost including Veterans at $2,375 billion.
Given the wide range of DoD and State costs provided in different reports, and the very real impact of Veterans costs, the resulting range of uncertainty from $2,002.4 billion to $2,375.0 billion from FY2001 through FY2019 represents a valid picture of the uncertainty in the currently available open source budget data. It also illustrates the need for a full independent audit of DoD, State/USAID, and Veterans Administration reporting by the GAO or CBO, and an audited official report that fully explores alternative ways of estimating the cost of war.
As becomes steadily clearer from the analysis in the following sections, the present U.S. government effort to cost America’s wars between FY2001 and FY2019 is a morass of contradictory, uncertain, and dysfunctional figures . Once again, this uncertainty reflects a serious failure by the executive branch and Congress to meet its responsibilities for accountability and transparency in a democratic society.
The most serious uncertainties in costing America's current wars do not lie in the uncertainties that have just been addressed. They lie in the choice of what costs to include or exclude from the estimate. There are a range of outside estimates of war costs that include a wide range of additional costs that are not reflected in the previous CRS, DoD, and State/USAID official estimates of the cost of war.
Some of these outside estimates are obvious political attempts to dramatize the analysts' opposition to the U.S. warfighting they are costing by including as many additional costs as possible to raise the total. Some such estimates, for example, create new cost categories like “opportunity cost” – which an analyst can use to raise the cost of any government expenditure he or she opposes.
Other analysts, however, have raised all too valid issues regarding what should or should not be included in estimating the true cost of war and such analyses have reached totals in excess of $6 trillion for FY2001-FY2019. The fact the Federal government does not include such costs should be a subject of public and Congressional debate, and some costs like the full wartime medical costs of Veterans should be made a formal and explicit part of funding U.S. combat operations. At the same time, some estimates of other costs seem questionable, and it is not clear that war costs deserve to be singled out for their potential impact on the national debt and interest payments,
It should be stressed that there is no" right" or "wrong" way to cost wars and military operations –any more than there is a " right" or "wrong" way to cost federal mandatory spending programs like Social Security and medical programs, or federal spending in any other area like education, homeland defense, law enforcement, or the environment and national parks. Every federal expenditure impacts on the debt and interest payments, on the economy, and involves an "opportunity cost" that means the money is not spent on other needs that may or may not have a higher priority.
As has already been demonstrated, the U.S. national budget is also often remarkably opaque. Outside estimate often can raise key issues, but lack the access and data to fully analyze a given cost. This analysis is no more immune to such problems than any other, and this why the proper Congressional and Executive Branch tasking of experts in the GAO, CBO, and offices of various inspector generals is so critical.
At the same time, it is worth pointing out that some added costs raise serious questions that can only be fully addressed by some form of federal audit. These include:
War-Related Additions to DoD Budget: It seems doubtful that DoD used major amounts of its FY2011-FY2018 Baseline Funding to pay for U.S. wars. It is far from clear, however, that this is the case. Other reporting indicates that DoD used the OCO account to pay for Baseline spending. Moreover, DoD was under intense pressure to cut Baseline spending for other reasons because of the budget ceilings or "caps" set by the Budget control Act. These broader pressures on Federal spending, the deficit, and the debt are summarized in Figure 23.
Other Pressures on the Baseline Budget: Total Baseline Department of Defense spending plans alone would have spent more on defense after FY2010 than the actual combined actual spending on both Baseline and OCO costs if it had not been for the Budget Control Act. Moreover, the budget caps on the DoD baseline forced serious cuts in manning, readiness, and modernization that were more serious than the diversion of funding to wartime expenditures in OCO spending. This, and the fact that the OCO account was exempt from the budget caps ion the Budget Control Act, gave both DoD and members of Congress a strong incentive to include some Baseline spending in the OCO account.
In short, these costs needs to be fully audited and assessed, and the inclusion of such costs in estimating war costs should be a critical and important subject of public and Congressional debate.
Anthony H. Cordesman holds the Arleigh A. Burke Chair in Strategy at the Center for Strategic and International Studies in Washington, D.C. He has served as a consultant on Afghanistan to the United States Department of Defense and the United States Department of State.
