Milken Institute Review
Iraq War Will Cost More-than-$2-Trillion
Two scholars, one a Nobel Prize winner, revisit their estimate of the true
cost of the Iraq war – and find that $2 trillion was too low. They consider
not only the current and future budgetary costs, but the economic impact
of lives lost, jobs interrupted and oil prices driven higher by political
uncertainty in the Middle East.
November 3, 2006
By Linda Bilmes
and Joseph E. Stiglitz
In January, we estimated that the true cost of the Iraq war could reach
$2 trillion, a figure that seemed shockingly high. But since that time, the
cost of the war – in both blood and money – has risen even faster than our
projections anticipated. More than 2,500 American troops have died and close
to 20,000 have been wounded since Operation Iraqi Freedom began. And the
$2 trillion number – the sum of the current and future budgetary costs along
with the economic impact of lives lost, jobs interrupted and oil prices driven
higher by political uncertainty in the Middle East – now seems low.
One source of difficulty in getting an accurate picture of the direct cost
of prosecuting the war is the way the government does its accounting. With
“cash accounting,” income and expenses are recorded when payments are actually
made – for example, what you pay off on your credit card today – not the
amount outstanding. By contrast, with “accrual accounting,” income and expenses
are recorded when the commitment is made. But, as Representative Jim Cooper,
Democrat of Tennessee, notes, “The budget of the United States uses cash
accounting, and only the tiniest businesses in America are even allowed to
use cash accounting. Why? Because it gives you a very distorted picture.”
The distortion is particularly acute in the case of the Iraq war. The cash
costs of feeding, housing, transporting and equipping U.S. troops, paying
for reconstruction costs, repairs and replacement parts and training Iraqi
forces are just the tip of an enormous iceberg. Costs incurred, but not yet
paid, dwarf what is being spent now – even when future anticipated outlays
are converted back into 2006 dollars.
Our Debt to Veterans
A major contributor to this long-term cost is the medical care and disability
benefits provided to veterans. More than one million U.S. troops have now
served in Iraq. And once they leave, each is entitled to a long list of benefits
for the remainder of his or her life. Veterans can apply for compensation
for any disabling injury or disease (physical or mental) that occurred on
active duty or any existing condition that was made worse by military service.
Benefits are based on the extent of the disability, ranging from 10 percent
to 100 percent. And, because some medical problems do not become apparent
right away, claims are likely to be filed for years after the war is over.
There are 2.6 million veterans currently receiving disability pay, including
a sobering 40 percent of the soldiers who served during the four-week-long
Gulf War in 1991. Accrued liabilities for U.S. federal employees’ and veterans’
benefits now total $4.5 trillion. Indeed, our debt for veterans’ health and
disability payments has risen by $228 billion in the past year alone.
These numbers are unlikely to fall. More than half of the troops in Iraq
have served two or three tours of duty under grueling conditions. Moreover,
depleted uranium, used in armor-piercing artillery shells because it is hard,
heavy and cheap, was implicated in many of the medical claims by soldiers
from the first Gulf War. And the same radioactive material was used in the
toppling of Saddam Hussein.
Note, too, that improvements in body armor mean that an unusually high number
of soldiers are surviving major injuries, but ending up disabled. About 20
percent of survivors have suffered major head or spinal injuries, 18 percent
incurred serious wounds and an additional 6 percent are amputees. The estimated
7,000 veterans with severe brain, spinal, amputation and other serious injuries
will require a lifetime of round-the-clock care.
Government medical facilities are currently overwhelmed by the needs of soldiers
injured in Iraq. Some 144,000 of them sought care from the VA in the first
quarter of 2006 – 23 percent more than the Bush administration had estimated
for the entire year! Similarly, the government projected that 18,000 returning
soldiers would seek treatment for posttraumatic stress disorder in 2006 –
but the VA treated 20,638 Iraqi war veterans for PTSD in the first quarter
alone. All told, in the past year, the VA has added 250,000 new beneficiaries
and still has a backlog of more than 400,000 pending claims.
Rebuilding the Post-Iraq Military
Another big future obligation is the cost to “reset” the military – that
is, to restore U.S. forces to their strength and preparedness prior to Iraq.
This will require a major capital investment to replace military equipment
depleted or destroyed by the war. The capital cost is in addition to the
operating costs for repairs, ammunition, spare parts and fuel. For example,
the United States now has 37,000 light military trucks in Iraq accumulating
mileage at up to six times the peacetime rate. And while there may be no
good time to replace the weapons, vehicles, medical equipment and the like
that will be used up, it’s clear the bill will come due at a particularly
bad time – that is, in the decades during which Americans will be wrestling
with the question of how to pay for the pensions and medical care of retired
Budgetary Cost of the War
Congress has already appropriated approximately $430 billion for military
operations, reconstruction and related programs in Iraq and Afghanistan.
And these cash outlays have been rising as the war has progressed. In fiscal
year 2003, the average monthly cost of operations was $4.4 billion, while
today operations are running about $10 billion a month.
Of the million troops who have served in Operation Iraqi Freedom, some 400,000
are reservists or members of the National Guard – which adds an additional
layer of costs. Reservists are expensive to activate because the military
needs to start paying them full-time salaries (instead of paying for one
weekend a month). By contrast, regular forces receive full-time salary in
war or peace. Most reservists are older and have families, so they are paid
additional compensation while on active duty. Moreover, if they are killed,
their dependents are entitled to compensation and benefits including housing,
education loans and job training.
The escalating costs also reflect the vast sums that the Defense Department
has been spending to recruit soldiers. In the past two years, the armed forces
have nearly doubled the number of recruiters, increased bonuses to as much
as $40,000 for new enlistees, and paid special bonuses and other benefits
worth as much as $150,000 for members of the Special Forces who re-enlist.
The Defense Department has also relied on contractors to support the war
effort, which has proved to be a very expensive way to keep the troop count
down. In many contracts, security costs represent 25 to 30 percent of the
total outlay. The Pentagon has managed some savings – such as no longer needing
to police the “no-fly” zone that protected the Kurds before Saddam was ousted.
But on balance, the Defense Department has increased spending by several
billion dollars annually for war-related expenses that are over and above
the sums going directly to combat operations.
While economists don’t generally include interest on extra budget deficits
as a cost of the war – interest payments can be viewed as transfer payments
to creditors – the budgetary reality is very different, and thus interest
costs are worth considering here. With rising interest rates (themselves
partly due to the war, as central banks around the world work to combat the
inflation brought on by high oil prices), these costs are soaring. The Congressional
Budget Office estimates that the interest payments on the money borrowed
to finance the Iraq war will total $264 billion to $308 billion.
We have used the CBO’s two scenarios for expected troop deployment to make
a reasonable projection of the likely underlying costs of operations, and
then adjusted these numbers to an accrual basis in order to reflect future
costs outlined above. Looking purely at direct costs to taxpayers, we estimate
that the total cost of the Iraq war will be in the $1 billion to $1.4 billion
range under the CBO’s core assumption that the U.S. maintains a small presence
in Iraq through 2016. Even under a more optimistic scenario – that all U.S.
troops are home by 2010, the budgetary cost of the Iraq operation will reach
nearly $1 trillion.
Economic Costs of the War
Economic costs differ from budgetary costs in three ways. First, some costs
are borne by individuals and families or by non-federal-government agencies,
and thus do not show up in federal accounts. Second, the prices paid by the
government do not reflect the market value of the services purchased. Third,
economic costs do not include interest payments (which from an economic perspective
can be viewed as transfer payments), but do include long-run impacts on the
growth of the economy. Here, we have focused only on a few of these additional
costs: the loss of productive capacity of the young Americans killed or seriously
wounded in Iraq, the loss of civilian wages that would have been earned by
those called back to duty in the Reserves, and the macroeconomic effects
that reduce output.
Military Fatalities, Serious Casualties and Reserves Wage Differential
Although it is problematic to translate the value of a life into monetary
terms, economists and private insurance firms commonly determine the “value
of a statistical life” (VSL) by inferring how much workers demand to perform
hazardous jobs (think mining or firefighting) or how much consumers are willing
to pay to reduce risk (think mammograms or smoke alarms). In non-military
areas, such as safety and environmental regulation, the federal government
values the life of a young adult male at around $6.5 million.
One could argue that the true cost of death and disability for an all-volunteer
army is already reflected in military pay premiums for hazardous duty. But
we think this greatly underestimates the real cost. First, recruits, many
of whom are too young to buy a beer legally, have little information about
the likelihood of being killed or injured, or how much they will come to
value their own safety later in their lives. Second, many of the soldiers
in Iraq are not really volunteers. The majority serving there are either
reservists or Guard members who never expected to go to war, or regular army
personnel ordered by the Pentagon to serve far beyond their scheduled length
Hence, we would argue that very little of the true cost of the deaths of
American soldiers is reflected in the budget. Using a VSL estimate of $6.5
million, the economic cost of the American soldiers and contractors who have
already lost their lives adds up to $16.9 billion. (We have not included
the cost of the estimated 40,000 to 100,000 Iraqis killed in the conflict.)
By the same reasoning, the budgetary expenditures also underestimate the
true economic costs to the soldiers wounded because the outlays do not include
adequate compensation for what tort law calls pain and suffering, or additional
health care expenditures by the soldiers’ families and non-federal-government
agencies. We believe veterans, and their families, receiving full disability
payments bear costs equal to those who die in combat, and therefore we should
assign each case a non-budgetary cost of $6.5 million (the value of a statistical
life). We assign a modest 20 percent of that figure to those who are wounded
There is also an economic cost in the difference between civilian and military
wages for reservists. This difference is a cost borne by the economy and
shows up as lower productivity. In their study of the economic costs of the
war published by the AEI/Brookings Joint Center in 2005, Scott Wallsten and
Katrina Kosec calculated that the “opportunity cost” of using Reserve troops
at current levels is $3.9 billion to date.
Note, moreover, that a disproportionate number of these reservists work in
critical “first-responder” jobs back home – as fire-fighters, police and
emergency medical personnel. Nearly half the police forces in the United
States now have some of their ranks deployed in Iraq, and the average length
of Guard mobilization is 480 days. It is difficult to measure the cost of
this deployment in purely economic terms because there is a large unquantifiable
“insurance” value of having trained first responders available for domestic
emergencies. Consider, for example, the losses associated with Hurricane
Katrina that might have been avoided if the 7,000 Louisiana and Mississippi
Guardsmen in Iraq had been home to help.
Macroeconomic Effects of the War
As large as the direct costs are, the indirect impact on total economic output
may be several times larger. Consider just two sources of macroeconomic cost.
The price of oil is significantly higher today than it was before the war
in Iraq. But to even begin to assign a macroeconomic cost to this, we need
to know what the price would have been if there had been no war.
Commodity futures markets provide some insight. Before the war, they were
implicitly forecasting that oil prices would remain in the range that they
had been – $20 to $30 a barrel – in spite of other, more predictable factors
affecting prices, such as strong economic growth in China and India. Today,
by contrast, the oil futures markets predict prices will be in the mid-$60-per-barrel
range during 2006 and 2007, and fall no earlier than the year 2008.
One explanation is that the instability in the Middle East brought about
by the Iraq war has increased the risk of investing in the region. But because
costs of extraction are so much lower in the Middle East, high oil prices
have not stimulated a commensurate supply response elsewhere. If political
stability is restored, the reasoning goes, prices will fall and investments
in high-cost liquid fuels elsewhere in the world – think heavy oil in Venezuela
or tar sands in Canada – will prove to be losing ventures.
We believe, accordingly, that the best estimate of the impact of Iraq on
oil prices is a large proportion of the $45-a-barrel increase since the war
began. Nonetheless, we offer a conservative calculation based on the assumption
that only a small fraction of that amount – $5 to $10 – is due to Iraq. Given
U.S. imports of roughly five billion barrels a year, a $10-per-barrel increase
translates into an extra expenditure of approximately $50 billion. Americans
are poorer by that amount. If merely a $5 price increase persists for five
years, this generates a conservative estimate of $125 billion in costs. More
plausibly, if we base our estimates on a $10 price increase, and assume (as
futures markets believe) it extends for at least six years, the cost is $300
Most macroanalyses assume that one must reckon with more than just these
direct supply-side effects if the economy is prone to operating below full
capacity. The increase in oil prices means Americans have that much less
to spend on other goods – including goods made in the United States. This
in turn leads to a reduction in aggregate demand, and the reduction leads
to lower economic output. Standard macroeconomic models suggest an “oil multiplier”
of around 1.5 (achieved over two years). Thus, assuming that the economy
remains below its potential, our cost estimate rises to $450 billion.
The macroeconomic costs associated with the increased expenditure on the
war are more difficult to estimate. If we were not spending the money on
Iraq, would we be spending it on something else? Would we have had the same
deficit, but just more tax cuts? Would the Federal Reserve have stopped raising
interest rates sooner if it wasn’t worried about the inflationary effects
of higher oil prices – and thereby made recession in 2006 less likely?
Here, we offer a very conservative estimate of these macroeconomic effects
using an “expenditure-switching” model. Spending money to hire, say, Nepalese
workers in Iraq provides little indirect stimulation to the American economy
– far less than would have been provided if the money had been spent on investments
in schools or roads (or, for that matter, on houses and cars) in the United
States. In estimates presented last January, we put the cost of budgetary
impacts (including expenditure switching and the impact on future productivity)
at $450 billion.
$2 Trillion and Counting
The total costs of the war, including the budgetary, social and macroeconomic
costs, are likely to exceed $2 trillion. As large as these costs are, an
equally large set of costs have been omitted. We have not included the costs
borne by other countries, either directly (as a result of military expenditures)
or indirectly (as a result of the increase in the price of oil.) Then there
are the intangible costs – the cost of our reduced capability to respond
to national security threats elsewhere in the world, and the cost of rising
anti-American sentiment in Europe and the Middle East. Americans have long
taken pride in fighting for human rights. But our credentials have been badly
tarnished by the Iraq war, leading to a sharp decline in America’s “soft
power.” On issues from trade negotiations to global warming to the international
criminal justice system, this decline will have a continuing impact on the
United States’ ability to have its point of view prevail.
In responding to cost-based criticisms of the invasion and occupation of
Iraq, the Bush Administration argues that one does not go to war on the basis
of calculations by bean counters. After all, Franklin Roosevelt did not wait
to respond to Pearl Harbor until his budget analysts could assay the costs
and benefits. But, with Iraq, America had a choice of whether and when to
attack. If there ever was a “project” that should have been subject to careful
scrutiny from all perspectives – including the economics – this was it.
Just as going to war was a matter of choice, staying in Iraq is also a matter
of choice. There may be costs associated with leaving. But there will be
costs associated with staying. Every day we stay in Iraq we accrue costs
that will be reflected in budget outlays, lost productivity and individual
pain and suffering for decades to come. We need to ask: are they outweighed
by the benefits?
Linda Bilmes teaches public finance at the Kennedy School of Government at
Harvard. - E-mail: firstname.lastname@example.org
Joseph Stiglitz, a former chairman of the Council of Economic Advisors and
chief economist at the World Bank, teaches at Columbia University. He was
awarded the Nobel Prize in economics in 2001.
- E-mail: email@example.com
This article originally appeared in the December 2006 issue of the Milken