Mobile Menu - OpenMobile Menu - Closed

Linda J. Bilmes

Linda J. Bilmes, Harvard University, Cambridge, MA, Daniel Patrick Moynihan Senior Lecturer in Public Policy, John F. Kennedy School of Government, and Joseph E. Stiglitz, Ph.D., University Professor, Columbia University, New York, NY (Nobel Laureate)

Chairman Filner, Congressman Buyer, and Members of the House Veterans Committee:

Thank you for convening this hearing today and for inviting us to testify on the true costs of war. 

There is no such thing as a “war for free.”  The history of warfare is a tragic cycle of people fighting, killing, wounding, exhausting armies and depleting treasuries followed by burying, taking care of the wounded, reconstructing, repaying war debts, and recruiting fresh troops.  The repercussions of war, and the costs of war, persist for decades after the last shot is fired.

Despite this well-worn path, the inevitable costs, the economic consequences and the long-term welfare of the troops are seldom mentioned at the start of a conflict.  Even when they are mentioned, the costs and risks are systematically understated. The result is that the burden of financing the war, the social cost of lives lost, quality of life impaired, families damaged and the expense of caring for veterans are typically not provided for in the run-up to conflict.

All wars, whether long or short, have continuing costs associated with the care of those who have fought in them.  It is a sobering thought that the peak year for paying out disability claims to World War I veterans did not occur until 1969 – more than 50 years after the armistice. The peak for paying out World War II benefits was in the 1980s – and we have not yet reached the peak cost for Vietnam veterans.  Even the Gulf War of 1991, which lasted just six weeks, costs more than $4 billion a year in disability compensation alone.

It is obvious now that the wars in Iraq and Afghanistan have been far more costly (in terms of both blood and treasure) than its advocates suggested at the outset.  Even with more realistic estimates, we might have come to the same decision about going to war. But the absence of reliable estimates meant there was no opportunity for a meaningful debate.  It has also prevented us from planning ahead for future costs.

The United States has already spent more than a trillion dollars in Iraq and Afghanistan for incremental war costs; in other words, costs that are in addition to regular military salaries, training and support activities, weapons procurement and so on.  There are other substantial incremental war-related expenditures across government for items including military medicine, military recruiting, contractors’ life insurance, Social Security disability benefits and paying interest on money borrowed to finance the war. 

But these figures do not include the long-term budgetary costs of veterans care, or any estimate of the economic and social costs of the wars.  

It may be hard to believe, but we still do not know the true cost of the Iraq war, much less the current war in Afghanistan. The U.S. Government budget is based on cash, rather than accrual accounting.  Government financial accounts track inflows and outflows of funds within a fiscal year, ignoring the long-term costs of depreciating equipment, purchasing complex weapons systems and caring for disabled veterans.  Basic information about outlays – what has actually been spent – is not readily available. The accounting systems at the Pentagon are notoriously poor at tracking expenditures; the Department has failed its annual financial audit for the past decade.  The Congressional Budget Office, the Congressional Research Service, the General Accounting Office, the Iraq Study Group and the Department’s own auditors and Inspector General, have all found numerous discrepancies in the Pentagon’s figures. Expenditures that relate directly or indirectly to the war are fragmented among many different departmental budgets and programs, making it laborious to piece together a complete picture.  Additional war funds are appropriated little by little, through supplementary budgets, making it all the more difficult to tally up the total costs.

The most detailed analysis of war costs has been conducted by the Congressional Research Service (CRS).  The CRS has noted that none of the known factors in the increasing war costs, including the operating tempo of the war, the size of the force, and the use of equipment, training, weapons upgrades and so forth, “appear to be enough to explain the size of and continuation of increases in cost.” We believe this discrepancy relates to the way the war has been fought, with excessive reliance on expensive contractors and funding for core defense activities getting mixed in with war funding due to poor budgeting and accounting. 

The U.S. Government also makes no attempt to capture the economic costs (including those associated with deaths or quality of life impairment of those injured), much less any tracking of how the economy might have fared in the absence of any conflict. 

These full costs are not transparent anywhere in the system.  Throughout the nine years of conflict in Iraq and Afghanistan, the Congressional Budget Office (CBO) has continued to use accounting frameworks that focus at best on the budgetary costs of war for 10 years, even as the long-term accrued costs of the wars and their impact on the economy have grown more apparent.  The only hint of the full costs of providing for military veterans is in the US Treasury’s financial statements for 2009, in the little-read “statement of net costs” which uses accrual methods.  According to this document, the US liability for burial and disability benefits for military veterans exceeds $1.3 trillion dollars.  (Even this figure – although large – does not reflect the full liability, because it excludes medical care and other benefits). There is no provision anywhere in the budget for how this liability will be paid.

Consequently, the estimate of budgetary costs that is presented to the public and the press is a partial snapshot, based on faulty accounting and incomplete data. 

Our work, which is based entirely on government data, was intended to fill this void.

Two years ago we published The Three Trillion Dollar War: The True Cost of the Iraq Conflict, in which we estimated that the total cost to the United States – including military expenditures through 2017, and lifetime health care and disability costs for returning troops, as well as economic impacts to the United States – would be $3 trillion[i].  This price tag dwarfed previous estimates, but subsequent investigations by both the Congressional Budget Office and the Joint Economic Committee of Congress found our estimate to be broadly correct. To ensure the credibility of our analysis, we deliberately used conservative assumptions.  As we will explain today, the empirical data that has come to light since the publication of The Three Trillion Dollar War demonstrates that our cost projections were excessively conservative, and that the war has had far-reaching economic consequences.  In particular, the costs of diagnosing, treating and paying disability benefits for veterans of the Iraq and Afghanistan conflicts are proving to be much higher than our earlier estimates.

This morning we will focus on three issues.

First, we will discuss some of the costs that the war has imposed on the US economy. 

Second, we will provide an updated estimate for the single biggest long-term budgetary cost of the current war, which is the cost of providing medical care, disability compensation and other benefits to veterans of the Iraq and Afghanistan conflicts.

Third, we will argue that such costs are inevitable and can be estimated to some extent in advance; therefore, the United States should make provisions for its war veterans at the time we appropriate money for going to war.  We will recommend steps that can be taken to address this unfunded financial liability. 

I.  The Cost of War and Its Impact on the US economy

The United States went to war without a clear understanding of the costs to the budget or to the economy.   Today we have a better view of both the benefits and the costs.

The benefits of war center on the value of additional security obtained by the war. This is a subject on which reasonable people may disagree, since it requires assumptions (typically unverifiable) about what would have happened in the absence of the conflict.  But even in this area, basic analytic principles can be of help, especially as we confront the challenge of the global war on terrorism, a security threat that is markedly different from earlier wars such as World War I and II, where our main objective was the defeat of a particular government.  The wars in Iraq and Afghanistan are different.  For instance, securing a particular piece of territory—ensuring that it cannot be used, for instance, for training of terrorists—may have little value, since training and terrorist activity can easily shift.  We have to have a global perspective.  We have seen this as Al Qaeda has shifted from Afghanistan, to Iraq, to Pakistan, and to Yemen.  Secondly, victory in this war, like all such insurgencies, entails winning hearts and minds—killing innocent victims, even if only as collateral damage, is a sure way to lose this battle.  The supply of insurgents can increase even as we succeed in killing thousands of the enemy.  (Economists say that the supply of insurgents and more broadly the strength of the opposition are endogenous.)  Thirdly, mistakes made at one point can have long lasting consequences, some more so than others. 

Economists and physicists refer to this under the name hysteresis; historians by the term path dependence.  We cannot go back to the world as it was, or as it would have been, if we had conducted the war in Afghanistan differently, and had not become embroiled in the war in Iraq.  But the consequences of some actions are more irreversible than others, and it is in those areas that we have to be particularly careful not to make mistakes.

Estimating the cost of the war is more straightforward. There is no doubt that wars use up resources.  The question is how to estimate the full magnitude of those resources used and assign values to them. 

The taxonomy of costs centers on (i) resources spent to date; (ii) resources expected to be spent in the future; (iii) budgetary costs to the government; and (iv) costs borne by the rest of the economy.  These latter costs are very real, even if the government does not pay them, and are referred to as the economic as opposed to the budgetary costs of the conflict.   In terms of the economic costs, there are microeconomic costs—costs borne by particular individual people or firms—and macroeconomic costs—impacts on the total economy over and above the sum of the micro costs.

What makes this analysis challenging is that government accounting systems do not document most items in a way that would enable an easy assessment of the resources directly used, or the full budgetary impact.  Accurate accounting is important because it provides information on the use of resources that is essential for good governance.   Transparency—clear, accurate financial information that is made available in a useable and timely format—is an essential part of democratic governance and accountability. 

The way we account for our troops matters.  For example, from the sole perspective of military accounting, the cost of a soldier’s life is valued at $500,000, ($400,000 in life insurance and $100,000 in “death gratuity” payment).   This number does not reflect either the true budgetary cost to government or the economic cost to society.  It does not include, for instance, the cost to the military of recruiting and training a new troop to replace the one who is lost, and the impact on morale and mental health on the rest of the unit, which may result in higher medical costs.  It also does not reflect the economic loss of a young person.  By contrast, when civilian agencies such as the EPA and FDA are evaluating a proposed regulation – when they compare the cost of imposing a regulation to the potential lives saved – they estimate the value of a life at between $6 million and $8 million.

Once a government embarks on a war, it has a myriad of decisions to make.  Not the least of these is the decision about when to exit.   An accurate assessment of the full costs of war—including, for instance, the full incremental cost of a surge of, say, 30,000 troops for one year—is an essential ingredient in making good decisions. The budgeting and accounting systems should be able to accurately track what has been spent as well as to anticipate the order of magnitude of future costs. For example, if 50,000 troops have already been wounded, it is feasible to estimate the approximate minimum future liability that the government will incur to provide these veterans with medical care and disability compensation (if a business incurs a liability to pay for injuries to some of its employees, it is required to make a provision for this liability). For an ongoing war, an accurate accounting of costs incurred is important information in assessing likely costs going forward.  Any business would want this kind of information as it made decisions; any publicly owned business would be required to keep its books in ways that investors could see the future consequences; and good business practice requires that the firm set aside money today for future obligations, like retirement benefits, accrued today.  We should expect no less of government.

It is important to realize why such information is so important.  It is partly a matter of accountability—how are our citizens to evaluate and judge a particular course of action if they do not know the costs?  But bad accounting leads to bad decisions.  If we do not take into account future disability and health costs, there is a temptation to scrimp on current expenditures, without regard to future costs.  Good accounting frameworks would show that such a course was penny wise but pound foolish.  Some actions entail cost shifting—say from government to others.  If VA hospitals are underfunded, some of our veterans who served their country so well—those who can afford it—may pay some of their own medical costs.  The total societal costs have not disappeared, even if budgetary costs are reduced.  This is one of the reasons that one needs an assessment of the overall economic costs.

The overall economic costs are typically much larger than the budgetary costs. However, there are instances where this is not the case. An example is where payments from the government to the private sector exceed the value of the resources procured—i.e., in war profiteering, which has been widely documented during the Iraq war.  The sheer size of the US military operations in Iraq and Afghanistan, (the biggest wartime mobilization since the all-volunteer force was created in 1973) placed a strain on the enlisted force, which led to an unprecedented reliance on paid private contractors.   This resulted in some cases, in payment of exorbitant sums for simple tasks such as painting walls and repairing trucks and gross over-payments to contractors such as Halliburton and Blackwater.  There have also been numerous cases of outright fraud where the U.S. Government has been found to have paid contractors for services that were never provided at all. Though such problems arise in all government procurement, there are normally safeguards in place that limit its scale.  During the Iraq War, many of these safeguards were suspended or relaxed. 

The best-run government organizations use cost accounting to estimate the direct and indirect costs of their activities. They also use accrual-based accounting to try to take future costs into account.  The focus on current-year cash budgeting leads to costly mistakes.  For example, the decisions not to buy more protective armor for troops or not to purchase mine-resistant vehicles certainly saved money on a cash basis.  But these decisions led, predictably, to much higher death and injury rates.   So too, the decision not to fund the Veterans Department adequately in 2005, 2006 and 2007 reduced current budgetary expenditures but at the expense of increasing the long-run (budgetary and economic) costs of providing care to returning veterans.   These and similar decisions were shaped by an accounting system that does not provide for the full long-term budgetary costs of current policies and by a budgetary system that does not estimate costs to the economy. 

In addition to the known costs of conducting current and future military operations and caring for war veterans (which we discuss below) the most sobering costs of the conflict are in the category of “might have beens”—what economists call opportunity costs. Specifically, in the absence of the Iraq invasion:  would we still be mired in Afghanistan? Would oil prices have risen so rapidly? Would the federal debt be so high? Would the economic crisis have been so severe?

Arguably the answer to all four of these questions is “no.” 

The first question concerns the “security opportunity costs” of the war.  The Iraq invasion diverted our attention from Afghanistan, a war that is now entering its tenth year and which threatens to destabilize nuclear-armed Pakistan.  By most accounts, the effort is encountering serious challenges, and even General Petraeus sees little prospect of an early exit.  While “success” in Afghanistan might always have been elusive, we would probably have asserted control over the Taliban, and suffered less expense and loss of life, if we had maintained our initial momentum and not been sidetracked in Iraq. Between 2003 and 2006, we spent five times as much money in Iraq as in Afghanistan.  It is likely we would have done far better if we had devoted those resources to Afghanistan, before the Taliban had re-established control.

The second cost is the higher price of oil, which has had a devastating effect on the economy. When we went to war in Iraq, the price of oil was under $25 a barrel, and future markets expected it to remain around that level.  With the war, prices started to soar, by 2008 reaching $140 a barrel.  The war and its impact on the Middle East, the largest supplier of oil in the world, clearly had something to do with the price rise.  We believe it was one of the major contributing factors—not only was Iraqi production interrupted, but the instability it brought to the Middle East dampened investment in this vital region from what it otherwise would have been.    In our conservative $3 trillion estimate, we attributed only $5-$10 of the increase to the war.  But, given our thirst for imported oil, even that small amount has a big impact—it translates into a much higher import bill for the United States.  We now believe that a more realistic estimate of the impact of the war on the oil price over a decade is at least $10-15 per barrel. That translates into a $250 billion increase in the cost of war.     

Third, the war added substantially to the federal debt.  It is the first time in America’s history where a government cut taxes as it went to war, even in the face of continued government deficits.  The US debt rose from $6.5 trillion to $10 trillion between 2003 and 2008, before the financial crisis.  At least one-fourth of that debt is directly attributable to the wars.  Of course, this doesn’t include unfunded future liabilities, for instance the more than half trillion dollars in future health care costs and disability payments for returning troops. 

The increased indebtedness meant that the US had far less room for maneuver in dealing with the global financial crisis.  Worries about the debt and deficit constrained the size of the stimulus. 

But the crisis itself was, in part, due to the war, and while, as we will explain below, the estimates that we provided in our book were overly conservative overall, e.g. in estimating future health care and disability costs, the most serious underestimate involved the macroeconomic consequences of the war.  The increase in oil prices reduced domestic aggregate demand—money spent buying oil abroad was money not available for spending at home.  The war spending itself provided less stimulus to the economy than other forms of spending—giving money to foreign contractors working in Iraq neither stimulated the economy in the short term (compared to investments in education, infrastructure, or technology) nor did Iraq spending provide a basis for long term growth.  Loose monetary policy and lax regulations kept the economy going—through a housing bubble, whose breaking brought on the global financial crisis.  We mentioned earlier that the deficits, to which the war contributed, reduced our room for maneuver.  But even today, three years into the crisis, as we struggle to deal with the aftermath—with more than one out of six Americans who would like a full time job unable to get one, with one quarter of Americans with mortgages owing more than the value of their house—it is increasingly clear that the size of the national debt—of which more than $1 trillion, or more than 7 percent, is attributable to the war—imposes important constraints our response. The result is that the recession will be longer, output lower, unemployment higher, deficits larger, than they otherwise would have been.

Counterfactuals—what might have happened if we had not gone to war—are always difficult, and especially so with complex phenomena like global financial crises with many contributing factors.  What we do know is that one of the true costs of war is its contribution to a worse economic recession, higher unemployment and larger deficits than might have otherwise occurred.  

I want to emphasize that there is a marked difference between deficit spending to finance investments—in infrastructure, technology, education—and to finance a war such as those in Iraq and Afghanistan.  Borrowing in the former case may make sense, especially when the economy has significant unemployment and interest rates are low.  Such expenditures improve the long-term debt, lower the long term debt to GDP ratio, and enhance growth—in short, they improve the country’s balance sheet.  That is not the case for debt-financed war expenditures, which worsen the country’s balance sheet.

The large disparity between budgetary and the full economic costs of war means there is a need for a comprehensive reckoning of the cost to the economy as a whole. The fact that we have been able to construct estimates of both underlines the fact that this exercise can be done once there is a will to do it. There are many skilled economists and plenty of data in various branches of government. Going forward, it is important that major decisions in the military arena, especially when they are decisions of choice, are subject to the same sort of rigorous analysis, both budgetary and economic. No estimate and no accounting system will be perfect. But the discipline that comes from applying these techniques routinely should increase the quality of debate and enable us as a country and a government to make more informed decisions in the future.

II. Updated estimates of long-term budgetary costs for returning Iraq and Afghanistan veterans

Over the past nine years more than 2.1 million Americans have served more than three million tours of duty.  More than 1.25 million veterans from these conflicts have returned home. The most significant long-term budgetary cost of war is providing medical care to those who have served, and paying disability compensation, pensions and other benefits to eligible veterans. As of this month, 5700 US servicemen and women have died and over 90,000 have been wounded in action or injured seriously enough to require medical evacuation.  A much larger number – nearly 600,000 – have already been treated in veterans’ medical facilities for issues ranging from brain injuries to hearing loss. The number returning home with serious mental problems has increased as troops were obliged to do repeated tours of duty, with shorter spans to recuperate. The medical community reports an “epidemic” of post-traumatic stress disorder (PTSD).

The evidence from previous wars shows that the cost of caring for war veterans continues typically rises for several decades and peaks in 30-40 years or more after a conflict.  The costs rise over time as veterans age and their medical needs grow. For example, the annual disability payment to veterans aged 34 and under is $6633. This rises to $8641 for veterans aged 35-54 and to $12,237 for those aged 55-74.  (In addition, the older veterans who are retired may now receive concurrent receipt of benefits from the Defense budget.  Those veterans who are not enrolled in the VA system are likely to be requiring significant costs from the Medicare system). 

However, for several reasons the long-term costs of the Iraq and Afghanistan conflicts can be expected to be even higher than in previous conflicts. This is due to a) higher survival rates; b) higher incidence of PTSD and other mental health ailments; c) a higher percentage of veterans claiming for benefits, especially those associated with mental health conditions; and d) more generous medical benefits, more presumptive conditions, and higher benefits in some categories. 

Let me briefly address each one of these factors.

First, the survival rate for severely injured troops has increased, relative to previous wars, as a result of improvements in battlefield medicine and other advances in health care.  In Iraq, the ratio of deaths to wounded-in-action was 1:7.3; compared with 1:2.6 in Vietnam, 1:2.8 in Korea, and 1:1.6 in World War II[ii].  This means that a large number of seriously wounded troops, some of whom have severe disabilities, will require lifetime care.  The wars have also had a high level of non-hostile injuries; our research shows that such injuries were more than 50 percent higher than during peacetime.

Second is the issue of mental health diagnosis and PTSD. There has been a considerable amount of medical research on this subject, including a number of recent studies on Iraq and Afghanistan veterans. The studies conducted at the University of California, San Francisco Medical School (UCSF) and elsewhere control for variables such as demographic factors, smoking, BMI, alcohol use, depression, and other factors, so they are an important way for us to understand what is attributable purely to war exposure.

There are three key findings in this literature.

First, the incidence of PTSD is closely correlated to the number of exposures to firefights that a soldier experiences.  That means that almost certainly, the long deployments, multiple deployments, and the lack of a clear “front line” for many of those serving has contributed to the extremely high levels of PTSD and other mental illness. There are now close to 900,000 troops who have served two or more tours of duty.

Second, PTSD is widespread, and has increased by 4-7 times since the invasion of Iraq.  The team at UCSF medical school, led by Dr. Karen Seal, studied all returning veterans who had been treated by the VA from 2002 through 2008[iii]. Her team found that 37 percent of returning troops received a mental health diagnosis.  Almost one in five of the troops were diagnosed with PTSD, with others diagnosed with depression. The majority of troops had concurrent diagnosis with other problems. Younger, lower-rank troops with the highest combat exposure were at the highest risk for PTSD.

Third, there is strong correlation demonstrated between PTSD and long-term physical health problems.  This includes heart disease, rheumatoid arthritis, heart failure, bronchitis, asthma, liver and peripheral arterial diseases[iv]. One recent study (Judith Andersen et al., 2010)[v] found that PTSD sufferers are 200 percent more likely to be diagnosed with a disease within 5 years of returning from deployment than the control group. Another new study (Beth Cohen, 2010) found that that veterans with PTSD utilized  non-mental health care services such as primary care, ancillary services, diagnostic tests and procedures, emergency services and hospitalizations 71-170 percent higher than those without PTSD. In addition, recent studies have shown that traumatic brain injury, which is estimated to affect some 20 percent of Iraq and Afghanistan veterans (often in conjunction with PTSD) places sufferers at higher risk for lifelong medical problems, such as seizures, decline in neurocognitive functioning, dementia and chronic diseases[vi].

Regarding the other reasons for higher costs:

Compared to previous conflicts, a higher percentage of Iraq-Afghanistan veterans are claiming for benefits, especially those associated with mental health conditions.  In large part, this is due to the outreach efforts that VA has undertaken, as well as the introduction of the post-deployment screen for mental health symptoms, and successful efforts by VA and many veterans groups and local organizations to make returning service members more aware of what they have earned and how to apply for it.  It is also likely that the Internet has made it easier to obtain information and to file disability applications. 

In addition, since our book was written, a number of recommendations that we and others urged have been adopted.  VA has expanded the Benefits Delivery at Discharge (BDD) program and Quick Start, increased the number of conditions that are presumptive in favor of the veteran, liberalized the PTSD stressor definition, increased some categories of benefits and outreach, provided five years of free health care instead of two, and is in the process of restoring medical care to 500,000 moderate income “Category 8” veterans. 

VA has also hired more medical and claims personnel, invested heavily in IT upgrades to the claims processing system, and is preparing to do much more.

All of these factors contribute to the rising cost estimates we will describe.  

Our model for projecting long-term budgetary costs is based entirely on government data.  We based our projections for troop levels on estimates by the CBO and CRS, and we used rates of average disability compensation, social security disability benefits and medical costs on information from the VBA, VHA, Social Security Administration and government economic indicators. 

In our earlier work, we estimated that the long-term cost of providing medical care and paying disability compensation for veterans of the Iraq and Afghanistan wars would be between $400 billion and $700 billion, depending on the length of the conflict and future deployment levels. This estimate was based on assumptions derived from historical patterns of medical claims and disability claims experienced in previous wars.  Since then we have updated our analysis to reflect the actual data for veterans returning from Iraq and Afghanistan and it is clear that the costs will be much higher. 

Revised Disability Cost Projections

In 2008 we had projected that between 366,000 and 398,000 returning Iraq and Afghanistan veterans would have filed disability benefit claims by this point (given 1.2 million returned troops, which we had correctly projected).  In fact, more than 513,000 veterans have already applied for VA disability compensation.  In our projections, the VA would not have received this many claims until 2013 at the very earliest.  We had also underestimated the complexity of these claims, the number of disabling conditions being demonstrated, and the likely increases in disability ratings over time for veterans who have been diagnosed with PTSD.  We now estimate that the present value of these claims, over the next 40 years, will be from $355 billion to $534 depending on the duration and intensity of US military deployment to the region.   

In addition, veterans who can no longer work may apply for Social Security disability benefits. We estimate that the present value of the lifetime benefits for these veterans will range from $33 to $52 billion. 

Revised Medical Cost Projections

In our earlier analysis, we had estimated that 30-33 percent (which would be fewer than 400,000) of returning veterans would be treated in the VA health system by 2010. The actual number is running at more than more than 565,000 veterans, which is about 45 percent of discharged troops[vii].  In our earlier work, we projected that the VA would not reach this level until 2016.

We had also underestimated the long-term costs of treating and caring for these veterans.  We had projected that at worst 20 percent of veterans would be diagnosed with mental health issues, whereas we now know that 30-40 percent of returning veterans are receiving these diagnoses.  This increases both immediate and long-term costs, given the relationship between mental illness and other conditions.  We also did not account for the cost to VA of adding personnel and increasing the mental health infrastructure. 

Accordingly, we can project how disability claims, and medical costs of the Iraq and Afghanistan veterans are likely to continue to increase with age. In this respect, they are likely to follow the pattern of Vietnam veterans, where it is estimated that 30 percent suffered from PTSD.   For example, the disability compensation paid to Vietnam veterans is 60 percent higher than the amount paid to veterans who served in peacetime. 

We now estimate that the present value of medical care provided by the VA to veterans from Iraq and Afghanistan over the next 40 years will be between $201 billion and $348 billion, depending on the duration and intensity of military operations in the region. 

Table 1: Estimated PV of Iraq and Afghanistan Veterans Disability and Medical Costs

(US$ Billions)


Best Case




Disability (VA)



Disability (SSA)



Total Cost $ billion




Original Estimate (2/08)









Social Security



Total Cost $ billion



Other Budgetary Costs

These estimates do not include a range of additional costs that will be paid by departments across government, including veterans’ home loan guarantees, veterans’ job training, concurrent receipt of pensions, and higher costs to Medicare and TRICARE for Life for veterans who are not enrolled in the VA system.  For example, Pentagon spending on health care for active-duty military has increased by 167 percent since 2001[viii]. It also does not include costs paid by state and local governments, or billions of dollars in VA capital investments, such as the construction of mental health clinics and construction of new hospitals, that will serve all veterans but are primarily targeted toward those returning from Iraq and Afghanistan.

One of our core recommendations in the book was that Iraq and Afghanistan veterans should be able to receive full education benefits, on a par with those provided to World War II veterans in the GI Bill.  Congress and the Administration finally enacted a new GI bill in 2008. This is an investment that will yield significant economic benefits.  However it will also add to the budgetary cost of the war.

Taking these costs into account, the total budgetary costs associated with providing for America’s war veterans from Iraq and Afghanistan approaches $1 trillion.

Economic Costs of Veterans

Earlier, we explained how the true cost of war goes beyond the budgetary costs; there are much larger social and economic costs.  While this is true for the country, it is especially true for our veterans and their families.

The military has faced its biggest challenge since conscription ended in 1973.  In many respects, the “All-Volunteer Force” has come under enormous strain.  Suicide among veterans is at record levels. Women troops (who make up 11 percent of the force) have been especially hard-hit:  divorce rates are three times higher for female than for male troops,  and more than 30,000 single mothers have deployed to the war zone.  These social costs are far-reaching.  They include the loss of productive capacity of young Americans who have been killed or seriously wounded in Iraq and Afghanistan, lost productivity due to mental illness, the burden on caregivers who frequently have to sacrifice paid employment in order to take care of a veteran with a disability, as well as increases in divorce, domestic violence, substance abuse, and other social problems. Additionally, a substantial number of those who were deployed, particularly among Reservists and Guards, were self-employed and have lost their livelihood as a result of deployment. For many veterans there is simply a diminished quality of life, the costs of which are borne by the individuals and families.

The military has also been forced to employ a shadow workforce of several hundred thousand contractors, who have proven to be indispensable to the war effort. These contractors have also suffered from high rates of casualties, injuries and mental health problems.  These impose both budgetary costs (through subsidies to worker compensation and insurance companies) and social costs in all the areas mentioned for troops. 

These substantial “social” costs are not captured in the federal government budget but nevertheless represent a real burden on society.    In a number of countries, this is actually recognized with quality of life impairment lump sum payments. In our book, we attempted to determine the monetary value of some of these costs, although many cannot be quantified.  At that time we estimated that the social costs would reach between $295 and $400 billion, in excess of the budgetary costs.  Given the high number of casualties in the war and the high incidence of illnesses, especially mental illness, it is certain that the true cost will be even higher.

III.  Funding War Veterans 

The scale of our financial commitment to providing for veterans is huge; both in terms of the payments we make today—mostly for previous wars -- and in the future.   We have predicted that the long-term cost of caring for the veterans of Iraq and Afghanistan will be at least $500 billion, and quite possibly much higher.  But at present, the U.S. has no provision for how it will pay for this growing long-term liability.   

The size of the current outgoings for veterans can be seen most clearly in the financial statements of the United States on the Statement of Net Cost, which lists the gross cost of US expenditures minus revenues. It shows that the net cost of providing for veterans is the fourth largest cost to the US Treasury.  For example, for the year ending September, 2008, the net costs were Defense: $740 billion; HHS: $713 billion; Social Security Administration: $663 billion; Veterans: $430 billion; Interest on the Debt: $241 billion; with all other costs far below.  In other words, the cost of providing for veterans equaled 12 percent of the cost of running the country. 

In terms of accrued long-term liability, the Balance Sheet of the United States lists $1.3 trillion in veterans’ compensation and burial benefits, and a liability for $220 billion in veterans housing loan guarantees.  But this does not take into account the accrued liability for providing medical care, or for veterans pensions, or for many of the other benefits we intend to provide.

Yet, while there are extensive debates and hundreds of studies on how to fund our obligations for Social Security and Medicare, there is little attention being paid to how best to fund veterans’ care.  In addition, both Social Security and Medicare are financed in part by taxes on non-recipients.  But there is no dedicated mechanism through which taxpayers who are not in military service contribute directly to caring for war veterans.  Funding must come from general revenues, competing with the myriad of other demands. 

The consequence of essentially ignoring the cost of caring for veterans is threefold.

First, it understates the true cost of going to war. We know that every war will have a long “tail” of costs, including the significant cost of providing for those who fight in the war, and their families and survivors.  However, in the appropriations process, we do not make any provision for this inevitable cost. This disguises and hides the true costs.

Second, from an economic perspective, it is poor financial management.  We should not be financing a 40-year long pension and benefit obligation from annual budget revenues. 

Third, it inevitably leads to the possibility that veterans’ needs will not be funded adequately.  There are always pressures to cut unfunded entitlements.  But veterans’ benefits are different from social security and Medicare.  They are more akin to “deferred compensation.”  They are payments for services rendered.  They are part of the implicit contract between our country and those that serve our country by fighting for and defending it.  The VA has the responsibility to determine the availability of VA care based on appropriations levels.  The financial statements explain that:  “In addition to health care benefits for civilian and military retirees and their dependents, the VA also provides medical care to veterans on an ‘as available’ basis, subject to the limits of annual appropriations. …VA’s Secretary makes an annual enrollment decision that defines the veterans, by priority, who will be treated for that fiscal year subject to change based on funds appropriated, estimated collections, usage, the severity index of enrolled veterans, and changes in cost.”[ix]

VA does not have the capacity to fully estimate its long-term obligations, and even with the best will in the world, this may result in insufficient funding.  It is well known that VA ran short of funds in 2005 and 2006 due to budget planning that was based on 2001 numbers, before the conflict began.  As recently as January 2009, GAO found that VA’s assumptions of the cost of long-term care were “unreliable” because the assumed cost increases were lower than VA’s recent actual spending experience[x].  VA is now facing the challenge of estimating demand for two years for the advance appropriations.  However, even this is proving very challenging since, using its current model, VA cannot determine its precise operating needs two-and-a-half years in advance; yet it is being asked by OMB and the appropriators to do this.  This places an impossible burden on the top VA officials.


We recommend a different funding model that would include the following:

  1. Establish a Veterans Trust Fund that would be funded as obligations occur.  Although we cannot estimate precisely the magnitude of long-term demands, it should be possible to develop a framework for setting aside some funding at the time war money is appropriated.
  2. Improve the actuarial capacity of the VA.  The VA should be directed to work with the Institute of Medicine to develop a better system of forecasting the amounts and types of resources needed to meet veteran’s needs in 30 years or more, when their needs are likely to peak. This should also include forecasting the regional impact and the infrastructure needs of the VA.
  3. The cost of any conflict that persists beyond one year should be funded by current taxpayers, through war surtaxes, war bond issues, or other means.


It is commonplace today for government to undertake extensive cost-benefit analyses of individual projects and regulations, to assess and, where possible, to quantify the benefits and costs.  Our analysis of true war costs follows in this tradition.  While expenditures on the military represent the single largest item for many countries, it has largely been immune from this kind of scrutiny.  Even if such an analysis does not change the decision to go to war, it can alter how the war is fought—and how we plan for the inevitable future costs of the war. 

We hope that the kind of analysis that we conducted for the Iraq and Afghanistan wars will become routine.  While the kind of economic calculus that we have conducted can only capture a fraction of the broader costs of war, we believe that even a greater awareness of these immense economic costs may have a salutary effect.  In particular, we hope that our work will contribute to a new way of thinking about long-term veterans costs, a way of thinking that would require us to budget for the lifetime needs of war veterans at the same time that we appropriate funds for the wars they will fight.

At the very least, we believe that democratic processes require an informed citizenry—and an informed citizenry must have a sense of the true costs that are likely to be encountered before it embarks on war.[xi] 

[i] Joseph E. Stiglitz and Linda J. Bilmes.  The Three Trillion Dollar War: The True Cost of the Iraq Conflict,  WW Norton, 2008.

[ii] Anne Leland and Mari-Jana Oboroceanu, “American War and Military Operations Casualties: Lists and Statistics,” February 26, 2010.

[iii] Karen Seal et al., 2009, “Trends and Risk Factors for Mental Health Diagnoses Among Iraq and Afghanistan Veterans Using Department of Veterans Affairs Health Care, 2002-2008,” American Journal of Public Health, 99(9): 1651-1658. (See also Karen Seal et al., 2007.)

[iv] Daniel Bertenthal, Beth Cohen, Charles Marmar, Li Ren and Karen Seal, 2009, “Association of cardiovascular risk factors with mental health diagnoses in Iraq and Afghanistan war veterans using VA health care,” JAMA 302 (5):489-492.;  and Boscarino JA, 2008, “A prospective study of PTSD and early-age heart disease mortality among Vietnam veterans: implications for surveillance and prevention,” Psychosomatic Medicine, July, 70(6):668-7;  Boscarino, JA, CW Forsberg andJ Goldberg, 2010, “A twin study of the association between PTSD symptoms and rheumatoid arthritis,” Psychosomatic Medicine, June 72(5):481-6 . In the latter, a study of twin pairs showed that the highest PTSD sufferers were 3.8 times likely to have rheumatoid arthritis compared with the lowest sufferers).  Spitzer has also shown increased incidence of angina, heart failure, bronchitis, asthma, liver and peripheral arterial diseases among PTSD sufferers).  

[v] Judith Andersen, et al., 2010, “Association Between Posttraumatic Stress Disorder and Primary Care Provider-Diagnosed Disease Among Iraq and Afghanistan Veterans,” Psychosomatic Medicine 72.

[vi] See Hoge, C.W. et al., “Mental disorders among US military personnel in the 1990s: Association with high levels of health care utilization and early military attrition,” American Journal of Psychiatry, 159(9):1576-1583; see also work from the Veterans Health Research Institute.

[vii] As of June 2010, 2.15 million US troops had served in the GWOT in Iraq and Afghanistan and there were 1.25 million veterans who were discharged. The number who had filed claims for compensation in connection with their service disabilities was 513,000 (Veterans for Common Sense, from DOD, previous number of 483,000 from Veterans Benefits Administration Office of Performance Analysis and Integrity, 11/18/09).  The number of GWOT veterans who had been treated at VA Hospitals and medical facilities was 565,000 (Veterans Health Administration).  

[viii] Statement by Rear Adm. Christine Hunter, Deputy Director of TRICARE, that Pentagon spending has increased from $19 billion in 2001 to projected $50.7 billion in 2011. (USA Today 4/25/10)

[ix] “Financial United States Government Notes to the Financial Statements for the Years Ended September 30, 2008,and September 30, 2007,” Notes to the Financial Statements, Page 73.

[x] GAO-09-664T.

[xi] For more information on the ongoing costs of war:  Visit