Troop Levels in the Afghan and Iraq Wars, FY2001-FY2012: Cost and Other Potential Issues

Contents

• Introduction: Obama Administration Plans for Afghanistan and Iraq
• Wide Range in Deployed Troop Strengths in DOD Sources
• In-Country Troop Strength: FY2002-FY2012
• Overall Changes in Troop Strength
• Overall Troop Strength Rises from FY2002-FY2008
• Likely Overall Declines: FY2008-FY2012
• Cost Implications of Changes in Troop Strength: FY2009-FY2012
• The Effects of Timing
• Changes in Afghanistan and Iraq
• Estimated Troop Levels in Afghanistan in FY2009
• Estimated Troop Levels in Iraq in FY2009
• The FY2009 Supplemental and FY2010 War Request
• Adjustments Affecting FY2009
• Potential Changes to DOD's FY2010-FY2012 Requests
• Potential Costs of Withdrawal
• Potential Questions About Troop Strength and Cost
• Ways to Measure Troop Levels for the Afghan and Iraq Wars
• Policy Issues Raised by Differences in Troop Counts
• Trends in Troop Levels From FY2002-FY2008 Using Five Alternative Sources
• Trends for both the Afghan and Iraq Wars
• Allocating Troops Between the Afghan and Iraq Wars
• Troop Levels for the Afghan War Since October 2001
• Initial Combat Operations
• Troop Levels Grow Steadily From FY2003-FY2006
• Further Increases As Violence Rises After 2006
• Seasonal Variations
• Troops from Other Countries
• Troop Levels for the Iraq War Since the March 2003 Invasion
• Initial Combat Operations
• Troop Levels Plateau From FY2004-FY2006
• Troop Surge in FY2007 and FY2008
• Changes from Month-to-Month
• Using Different Troop Strength Figures
• Service Roles and Readiness Concerns
• Measuring the Burden of Deployment

Summary

In February and March 2009, the Obama Administration announced its plans to increase troop levels in Afghanistan and decrease troop levels in Iraq. In Afghanistan, 30,000 more troops are deploying this year while in Iraq, troops will gradually decline to 35,000 to 50,000 by August 31, 2011 with all troops to be out of Iraq by December 31, 2011. The most commonly cited measure of troop strength is "Boots on the Ground" or the number of troops located in Afghanistan and in Iraq. Based on average monthly Boots on the Ground figures, the number of troops in Afghanistan and Iraq increased from 5,200 in FY2002 to a peak of 187,900 in FY2008 primarily because of increases in Iraq beginning with the invasion in March 2003. In FY2009, total troop strength is expected to remain the same as planned increases in Afghanistan offset declines in Iraq. By FY2012, overall troop strength for the two wars is likely to decline to 67,500 when the withdrawal from Iraq is expected to be complete.

For Afghanistan, troops in-country grew gradually from 5,200 in FY2002 to 20,400 in FY2006. Between FY2006 and FY2008, average strength there jumped by another 10,000 to 30,100. Under the Administration's plans, CRS estimates that average monthly Boots on the Ground in Afghanistan may increase to 50,700 in FY2009 with a further increase to 63,500 the following year once all new units are in place. Currently, additional increases have not been approved.

For Iraq, troops in-country nearly doubled between FY2003 and FY2004 reaching 130,600. By the following year, average strength grew by another 13,000 to 143,800, with that level maintained in FY2006. During the surge in troops initiated by President Bush, average troop strength in Iraq grew by 7,000 or 6% in FY2007 and another 9,500 or 9% in FY2008, reaching a peak of 157,800. CRS estimates that average troop strength in Iraq will decline to 135,600 in FY2009, 88,300 in FY2010, 42,800 in FY2011, and 4,100 in FY2012. While it is not clear whether war costs will change precisely in tandem with troop levels, these changes can provide a benchmark to assess requests. Based on changes in troop levels and other adjustments, CRS estimates that war costs could be about $8 billion less than the Department of Defense (DOD) $141 billion request for FY2009, and about $13 billion below its $130 billion request for FY2010. For the next year, FY2011, CRS estimates that DOD's requests could be $42 billion more than the current planning figure of $50 billion. And in FY2012, CRS estimates war costs could be $20 billion higher than the Administration's estimate of $50 billion.

Although Boots on the Ground is the most commonly cited measure of troop strength, that measure does not include over 100,000 other troops deployed in the region providing theater-wide support for Operation Enduring Freedom (OEF), the Afghan War, and Operation Iraqi Freedom (OIF), the Iraq War. Before the 9/11 attacks, the United States had deployed about 26,000 troops in the Central Command region, which includes Afghanistan and Iraq. Based on the most comprehensive DOD measure of troop strength, 294,000 troops were deployed for OEF and OIF as of December 2008, a tenfold increase since 2001.This more inclusive measure may more accurately capture the overall demand for troops. The Administration has not indicated how its plans would affect troops providing support in the region. Using five DOD sources, this report describes, analyzes, and estimates the number of troops deployed for each war from the 9/11 attacks to FY2012 to help Congress assess upcoming DOD war funding requests as well as the implications for the long-term U.S. presence in the region.

---

Troop Levels in the Afghan and Iraq Wars, FY2001-FY2012: Cost and Other Potential Issues

In February and March 2009, the Obama Administration announced its overall plans to increase troop levels in Afghanistan and decrease troop levels in Iraq for 2009 through 2011. Using several Department of Defense (DOD) data reports, this report describes, analyzes, and estimates deployed troop strength from the 9/11 attacks to FY2012 to provide Congress with a tool to assess

• current and future DOD war funding requests;
• implications for the U.S. military presence in the region; and
• deployment burdens on individual service members and each of the services.

Introduction: Obama Administration Plans for Afghanistan and Iraq

In February and March 2009, President Obama approved the deployment of an additional 21,000 service members to Afghanistan in the spring and summer of 2009, meeting most of the request from Secretary of Defense Robert Gates, and General David McKiernan, former Commanding General of the International Security Assistance Force (ISAF) in charge of the Afghan theater of operations. According to the President, these additional troops are intended to "stabilize a deteriorating situation in Afghanistan."¹ In the winter of 2008 before leaving office, President Bush approved deploying an additional Brigade Combat Team (BCT) of about 9,000 to Afghanistan in January 2009. Still pending is a DOD request for an additional 9,000 troops that could be deployed in Afghanistan sometime in the winter of 2009.

Last October, General David McKiernan requested about 35,000 more combat troops for Afghanistan composed of four Brigade Combat Teams with support.² Later, 4,000 trainers were added, bringing the total request to 39,000. About 30,000 of that request has been approved by either former President Bush or President Obama. With these increases, CRS estimates the number of U.S. troops deployed in Afghanistan is expected to reach about 63,450 by the end of FY2009, double that of the prior year; this estimate is somewhat lower than the Administration's estimate of 68,000.

If some 31,000 coalition troops from NATO nations serving in the International Security Assistance Force are included, average monthly foreign troop strength in Afghanistan would be about 81,000 in FY2009, and 93,450 in FY2010.³ These figures do not include from 10,000 to 17,000 more U.S. troops dedicated to Operation Enduring Freedom (OEF), most of whom provide support in the region. Because all but the 2,000 troops in the Philippines are linked to the Afghan War, CRS uses OEF and the Afghan War interchangeably in this report.

In a speech to U.S. Marines at Camp Lejeune on February 27, 2009, President Obama announced that the Administration's review of U.S. strategy in Iraq was complete and that the U.S. mission in Iraq would shift from combat to supporting and training Iraqi security forces, and that U.S. troops would decline from the February 2009 level of about 140,000 in-country to 35,000 to 50,000 troops by August 31, 2010. In addition, the policy would meet the deadlines set in the January 1, 2009 Security Agreement with Iraq that requires all U.S. combat troops to move outside of cities by the end of June 2009, as recently took place, and that all U.S. troops leave Iraq by December 31, 2011.⁴

The Administration has not explained the effect of these withdrawals on other troops deployed for OIF, or the Iraq War, which includes not only U.S. troops located in Iraq but also some 80,000 to 90,000 troops providing theater-wide support in neighboring areas. CRS uses OIF and the Iraq War interchangeably in this report.

The President reportedly considered options ranging from the 16-month withdrawal proposed during the campaign to a 23-month alternative reportedly favored by some in the military. Ultimately, the President adopted a 19-month plan.⁵

DOD spokesmen have suggested that increasing troop levels in Afghanistan depends, at least in part, on decreasing troop levels in Iraq if the services are to continue current 12-month tours. During the surge in troops in 2007 and 2009, tours grew to 15 months, a move that proved very unpopular.⁶

These two decisions about troop levels are reflected in the FY2009 Supplemental request sent to Congress in early April and the FY2010 war request submitted in early May 2009 with the regular budget. The FY2009 Supplemental covers DOD's war costs for the rest of this fiscal year that ends September 30, 2009 because DOD received only part of its war funding in the FY2009 bridge fund enacted last summer.⁷ According to statute, the Administration is to include a full year of war costs with the regular budget in FY2010, and provide separate budget displays for each war, and troop strength levels as well as the underlying assumptions. DOD provided general information about troop levels but not the underlying assumptions or separate budget displays for each operation in its FY2010 request.⁸

The Administration's decisions to reduce the number of troops in Iraq reflects declining levels of violence and peacefully conducted provincial elections that took place in January 31, 2009, while its decision to increase troop levels in Afghanistan reflects rising levels of violence and elections slated for August of 2009. Another important element in decisions about troop levels is the effectiveness of Afghan and Iraqi security forces, where the U.S. has invested $38.9 billion thus far to train and equip these forces.⁹

To identify the implications of troop levels for these and other policy issues, this report

• explains how and why measures of troop levels in five different DOD data sources differ;
• estimates future troop strength in Afghanistan and Iraq for FY2009-FY2012 under the Obama Administration plan using "Boots on the Ground," the most commonly cited measure;
• discusses the potential cost implications of these changes in troop levels along with other adjustments;
• analyzes and explains past trends in troop levels for the Afghan and Iraq Wars under the different measures; and
• measures contributions and burdens of deployment for the four services.
• For detailed description and analysis of the political and military context for changes in troop levels in Afghanistan and Iraq, see the following CRS Reports: CRS Report RL30588, Afghanistan: Post-Taliban Governance, Security, and U.S. Policy and CRS Report RL31339, Iraq: Post-Saddam Governance and Security, by [author name scrubbed]; and CRS Report R40156, War in Afghanistan: Strategy, Military Operations, and Issues for Congress and CRS Report RL34387, Operation Iraqi Freedom: Strategies, Approaches, Results, and Issues for Congress, by [author name scrubbed].

Wide Range in Deployed Troop Strengths in DOD Sources

When discussing deployed troop levels, it is important to be clear about which troops are included and which troops are not included. CRS analyzed five different sources for DOD troop strength for the Afghan and Iraq Wars ranging from including only those troops deployed in-country to including not only troops deployed in-country but also those providing theater-wide support in the region.

In the June preceding the September 11, 2001 attacks, the United States had about 26,000 troops stationed in the U.S. Central Command region, which encompasses Afghanistan and Iraq and neighboring areas. Most troops were located in Kuwait, Saudi Arabia and on ships in the region, including some conducting over flight operations for Northern and Southern Watch to contain Saddam Hussein after the first Gulf War.¹⁰

Based on the most comprehensive measure of troop strength, DOD's Defense Manpower Data Center's Location Report, capturing all troops deployed for both wars, there were 294,000 U.S. troops in the region for Operation Enduring Freedom and Operation Iraqi Freedom in December 2008, more than ten times the level seven years ago (see Figure 1 and Box 1).¹¹ Figure 1 shows the number and location of these U.S. troops in December 2008, with an allocation between OEF or OIF based on where personnel are located that generally reflects guidelines developed by the Joint Staff, Joint Chiefs of Staff and Central Command. These allocations may or may not match distributions of war costs, which are collected separately by each service.¹²

This figure of 294,000 troops is over 100,000 higher than the December 2008 total of 181,000 reported in DOD's oft-cited Boots on the Ground," (BOG) report.¹³ Military leaders, DOD press spokesmen and Members of Congress use the monthly BOG figures to give a snapshot of the number of troops deployed in-country in Afghanistan and in Iraq. For example, Admiral Mullen, Chair of the Joint Chiefs of Staff, said there were 150,000 troops in Iraq "right now" on November 17, 2008, and Pentagon press secretary, Geoff Morrell said in January 2009 that the proposal to add 30,000 troops in Afghanistan would double the number there.¹⁴

DOD has reported these Boots on the Ground figures to Congress each month since 2008.¹⁵ As of the most recent data for April 1, 2009, DOD reported that there were 178,400 Boots on the Ground in Afghanistan and Iraq, slightly lower than in December 2008 reflecting the withdrawal of some troops from Iraq offset by the deployment of additional troops to Afghanistan.¹⁶

Part of the difference between DOD's Boots-on-the-Ground figures and the Location Report is definitional. Boots on the Ground reports only personnel in Afghanistan or in Iraq while the Location Report counts deployments of all military personnel tracked as part of OEF and OIF, including primarily personnel providing support in neighboring countries, as well as about 2,000 OEF personnel engaged in counter-terror operations in the Philippines.¹⁷ Relying primarily on guidelines developed by staff of the Joint Chiefs of Staff and Central Command, CRS allocated troops to each operation based on their location (see Figure 1 and Box 1).

CRS uses the Iraq War interchangeably with OIF and the Afghan War interchangeably with OEF to include not only troops in-country, but also those providing theater-wide support primarily in the region. (All but the 2,000 troops in the Philippines are part of OEF or the Afghan War.)

For example, in addition to the 38,000 troops in Afghanistan in December 2008, OEF includes:

• 2,300 troops in Kyrgyzstan supporting Afghan operations;
• 2,100 troops in Djibouti where there are groups connected to Osama Bin Laden;
• 2,200 in the Philippines conducting other counter-terror operations; and
• a scattering of other military personnel in the region (see Figure 1).

Figure 1. Location of U.S. Troops Deployed for Operation Enduring Freedom (OEF) and Operation Iraqi Freedom (OIF), December 2008

Similarly, while there are 161,000 troops for OIF in Iraq, the operation also includes:

• 46,000 troops in Kuwait providing a wide range of support;
• 15,000 Navy personnel on ships in the region; and
• several thousand troops deployed to regional headquarters (4,800 Navy personnel in Bahrain, 2,000 Air Force (AF) personnel in the United Arab Emirates, and 8,000 AF personnel in Qatar for the Air Force (see Figure 1 and Box 1).¹⁸

There are also some 15,000 troops whose location is unknown or was not provided to the Defense Manpower Data Center, probably reflecting either data collection problems in the services or some personnel whose location is classified (see Figure 1).²⁰

The gap between Boots on the Ground figures and the Location Report reflects differences in both who is counted and how personnel are counted. At one end of the spectrum, Boots on the Ground figures reflect once-a-month headcounts collected from the services by the Joint Staff that include only troops located in-country in Afghanistan and Iraq. At the other end of the spectrum, the Defense Manpower Data Center data captures all deployments during each month for those in the 28 countries, plus Navy members aboard ships in the region who are deemed part of OEF and OIF. These figures include short and long-term deployments.²¹

Another inclusive measure is "average strength," which captures "person years" for all deployed OEF and OIF personnel taking into account the length of deployments. For example, an Army member deployed for 15-months would count as 1.25 in average strength while an Air Force member flying missions for six months of the fiscal year would count as .5 in average strength.

The most appropriate figure to capture troop strength for the Afghan and Iraq wars may depend on the purpose. For example, Boots on the Ground counts may be the best measure for those troops most likely to face combat situations. On the other hand, Location Reports capture all troops deployed in the region for the two operations. Finally, Average Strength may be the best measure for looking at the total demand for troops as well as the cost of both operations.

In-Country Troop Strength: FY2002-FY2012

The Obama administration outlined its plans for troop strength in the next several years only in terms of the number of troops in-country, also referred to as "Boots on the Ground" in Afghanistan and Iraq. For this reason, CRS estimates of future troop levels are also expressed in terms of troops in-country. These figures do not include over 100,000 other deployed troops who provide theater-wide support in the region and are dedicated to Operation Enduring Freedom, the Afghan War, and Operation Iraqi Freedom (see Figure 1 and later discussion). Nevertheless, Boots on the Ground figures captures those troops most likely to face combat and can provide an overall sense of the direction and scope of change.

Overall Changes in Troop Strength

To better capture strength levels in each fiscal year, CRS uses monthly averages rather than the once-a-month snapshots commonly cited by DOD spokesman.²² These figures smooth out month-to-month variations and reflect overall demands for troops each year and are more closely related to the likely cost to deploy, conduct operations, and support troops in-theater.

Overall Troop Strength Rises from FY2002-FY2008

Between FY2002 and FY2008, the number of troops in Afghanistan and Iraq increased from 5,200 troops to a peak of 172,000. Troop levels first jumped in FY2003 with the invasion of Iraq when troop strength in-country reached 78,000 for both wars (see Table 1). This figure does not include troops on ships or deployed in the region.

Between FY2003 and FY2005, average monthly troop levels doubled from 78,100 to 162,900, reflecting increases in both Afghanistan and Iraq. There was little change in FY2006. Then in January 2007, President Bush announced his decision to send five more Brigade Combat Teams to Iraq for a temporary surge in troop levels to quell rising violence. Average strength rose in FY2007 and FY2008 because of these additional troops as well as additional increases in Afghanistan. Average strength for both wars peaked at 187,900 in FY2008.

During the surge, average troop strength for both wars grew from 161,500 in FY2006 to 172,000 in FY2007, an increase of 10,500 or 7%. The following year, in FY2008, when the peak was reached, troop levels were 187,900, an increase of another 15,900 or 9%. These increases reflect growth in troop levels in Afghanistan as well as the surge in Iraq.

Likely Overall Declines: FY2008-FY2012

Based on the plans announced by the Obama Administration described above, CRS estimates that average monthly troop strength in Afghanistan and Iraq will decline from 187,900 in FY2008 to 67,500 in FY2012, a drop of 64%. These estimates reflect announcements made this spring by the Obama Administration. If these plans change — with more rapid increases in Afghanistan or slower decreases in Iraq — these estimates would, of course, change.

Under the current plan, overall troop strength in Afghanistan and Iraq in FY2009 will remain essentially the same between FY2008 and FY2009 as declines in troop levels in Iraq are offset by increases in Afghanistan (see Table 1). In FY2010, assuming troop levels remain the same in Afghanistan and the pace of withdrawals in Iraq picks up, average troop strength for both wars would fall to 151,800 or by 19%. By that year, troop levels in Afghanistan would be about 25,000 less than those in Iraq.

In FY2011, the pace of withdrawal in Iraq would quicken to meet the deadline of withdrawing all but 35,000 to 50,000 troops in Iraq by August 31, 2009 announced by the President. CRS estimates this would reduce overall average troop levels to 106,200 in FY2011, a decline of 30%, assuming no change in Afghanistan. This would decrease overall deployed strength by another 30%, still above the FY2003 level when operations in Afghanistan and Iraq were both underway. By FY2011, the balance of troops between the two operations would shift with 20,000 more troops in Afghanistan than deployed in Iraq.

By FY2012, when the withdrawal from Iraq is slated to be complete to meet the Security Agreement with Iraq, overall troop levels would fall by another 36% to 67,500 as all but the last three Brigade Combat Teams leave Iraq, and some 7 BCTs would be in Afghanistan (see Table A-1). DOD did not include troop estimates for these years in its budget submissions.

By showing average troop strength both in the past and estimating troop strength in the future, CRS hopes to provide a tool that can help Congress assess the pacing cost implications in current plans to increase troop levels in Afghanistan and decrease troop levels in Iraq including:

• Are the planned increases in Afghanistan too much, too little or about right and how long are higher levels likely to be maintained?
• Is the pace of withdrawal in Iraq too fast, too slow, or about right and what are the implications if current plans change?
• What are the likely implications for war costs in FY2009-FY2012?

Cost Implications of Changes in Troop Strength: FY2009-FY2012

The Administration's FY2009 supplemental, submitted on April 9, 2009, and the FY2010 war budget, submitted on May 7, 2009, presumably incorporate the recent decisions to increase troop levels in Afghanistan by 30,000 by the summer of 2009 and to reduce troop levels in Iraq gradually in FY2009 and then more rapidly in FY2010 and FY2011. Thus far, the Administration has not indicated how these decisions will affect the other 100,000 troops in the region.

These planned adjustments in troop levels are likely to be a key variable in how DOD's war costs change. War costs cover the additional or incremental costs to deploy and support troops (e.g. combat pay, upgrading overseas bases, meals), conduct operations, and to repair, replace and sometimes upgrade equipment (see Appendix B). The cost of the regular salaries of military personnel, conducting normal training and maintaining equipment or running military installations at home are all covered in DOD's regular budget. CRS estimated the cost implications for FY2009-FY2012 war costs of changes in average strength in Afghanistan and in Iraq along with other changes.

The Effects of Timing

The timing of proposed increases and decreases in troop levels is important in estimating current and future war costs. In addition to the 33,450 troops deployed in Afghanistan at the beginning of this fiscal year, an additional 30,000 troops are scheduled to deploy during the remainder of FY2009. The increases in troop levels in Afghanistan include the following:

• one additional Brigade Combat Team of about 9,000 including support deployed in January 2009 as directed by then-President Bush;
• 21,000 more troops including a brigade of 8,000 U.S. Marine Corps personnel deployed in March 2009, and about 9,000 Army troops in June 2009 as directed by President Obama; and 4,000 additional trainers deployed this summer as directed by President Obama.

The war cost implications of these decisions depend on the amount of time that these personnel will be in-country: for example, 9 months for the brigade deployed in January, seven months for the Marine Corps personnel, four months for the additional Army brigade and five months for the trainers. ²³ Not until FY2010 would a full year's cost for all 30,000 additional troops be needed.

Similarly, for Iraq, the savings from the withdrawal in June 2009 of two Brigade Combat Teams that will not be replaced will reflect the three months that those roughly 18,000 troops will no longer be in-country in FY2009.

To estimate the effect of these troop plans on average troop strength and cost, CRS developed a schedule of the movement of units in and out of Afghanistan and Iraq based on Presidential and DOD announcements (see Appendix A and Table A-1). To simplify the metric, CRS assumes that 9,000 troops make up a BCT-equivalent including both combat and support troops. This reflects the Army's assumption that 10,000 troops is the average for a BCT including support forces, adjusted for the fact that Marine Corps units tend to be smaller.²⁴

CRS uses this metric to estimate the cost of maintaining a certain number of weighted average BCT-equivalents in country that reflect both the number of units and the amount of time that they are in Afghanistan or Iraq in each fiscal year. While this metric has the advantage that it can be tied to Administration and DOD plans to move units into or out of Afghanistan and Iraq, it implicitly assumes that the more than 100,000 other troops in the region rise or fall in proportion to troop changes in-country.

While many of the Navy and Air Force personnel conducting operations at sea or outside the country who support ground operations are likely to rise or fall proportionately, other personnel, such as headquarters personnel in Bahrain or Qatar, may adjust gradually or not at all. Army support troops in Kuwait who provide maintenance services may face reduced workload as troops leave partly offset by higher workload as equipment that has not rotated with troops is prepared to be re-deployed.

DOD has not provided any information about the likely effect of the planned increases and withdrawals on these additional 100,000 troops providing theater-wide support. For example, troop levels outlined in the FY2009 Supplemental and the FY2010 war request only include strength levels in-country.²⁵

Between FY2006 and FY2012, the balance of troops dedicated to Afghanistan and Iraq would gradually shift from the 88% for Iraq and 12% for Afghanistan that was typical for FY2006-FY2008 to 73% for Iraq and 27% for Afghanistan in FY2009 (see Table 2).

By FY2010, troops in Iraq would make up about 58% and those in Afghanistan about 42% of the total. The following year, Afghanistan would become the predominant mission with 60% of the troops, and by FY2012, Afghanistan would have 93% of all troops in-country.

These estimates reflect the Administration's plans as currently outlined. If plans change with different pacing of increases and decreases, then troop strength and costs would rise or fall as would the balance between the two operations.

Changes in Afghanistan and Iraq

For Afghanistan, troops in-country grew gradually from 5,200 in FY2002 to 20,400 in FY2006. Between FY2006 and FY2008, average strength there jumped by another 10,000 to 30,100. Under the Administration's plans, CRS estimates that average monthly Boots on the Ground in Afghanistan may increase to 50,700 in FY2009 with a further increase to 63,500 the following year once all new units are in-place (see Table 2).

Currently, additional increases have not been approved. Including some 31,000 coalition troops, the total number of foreign troops in Afghanistan could rise to about 95,000. Secretary Gates has suggested that going beyond that level could become "a hindrance rather than a help because we begin to look like occupiers to the Afghans," as did the Soviets with their 110,000 to 120,000 troops.²⁶

For Iraq, troops in-country nearly doubled between FY2003 and FY2004 reaching an average of 130,600. By the following year, average strength grew by another 13,000 to 143,800, with that level maintained in FY2006. During the surge in troops initiated by President Bush, average troop strength in Iraq initially grew by 7,000 or 6% in FY2007, less than the increase between FY2005 and FY2004. By the next year with another 9,500 troops, troops or another 9%, reaching a peak of 157,800 even as the additional combat brigades began to be withdrawn.

CRS estimates that troop strength in Iraq will average 135,600 in FY2009, 88,300 in FY2010, 42,800 in FY1011, and 4,100 in FY2012. Annual decreases range from 22,00 to 46,000 in these years.

Estimated Troop Levels in Afghanistan in FY2009

Based on the schedule outlined by the Administration, CRS estimates average annual U.S. troop levels in Afghanistan would likely grow from an average of 30,100 in FY2008 to about 50,700 in FY2009 or by 68% percent. The Administration plans to review troop levels at the end of this year including a request by former Commander of U.S. Forces in Afghanistan, General McKiernan, to deploy an additional combat brigade in the winter of 2009, partway through FY2010. DOD estimates troops in Afghanistan would average 45,000 in FY2009.²⁷

At a press conference in May 2009, Secretary Gates characterized the additional brigade as a "hard sell."²⁸ If no additional troops are approved, troop levels in Afghanistan would be likely to increase by another 25% to an average strength of 63,350 in FY2010, or close to DOD's estimate of 68,000 included in its war request.²⁹ According to Pentagon spokesman, Geoffrey Morrell, Secretary Gates has voiced concerns that "there is a tipping point in terms of the American footprint . . . on the ground in Afghanistan."³⁰

With no further changes, CRS estimates that troop levels in FY2011 and FY2012 would remain at that level. In this scenario, the number of BCT-equivalents would more than double from 3.3 BCT-equivalents in FY2008 to about 7.1 BCTs in FY2010 (see Table A-1).

If the additional brigade of 9,000 troops including support were added, troop levels would rise to 70,200 in FY2010 and peak at 72,450 or about 8 average BCTs in FY2011 and FY2012.³¹ Troop levels could, of course, increase or decrease in these later years if current plans change.

Estimated Troop Levels in Iraq in FY2009

In FY2008, troop strength in Iraq declined from a peak of 170,000 in November 2007 to 147,000 by September 2008 as additional units sent for the surge were withdrawn. Administration plans call for a gradual drawdown in FY2009 to be accomplished mainly by not replacing two of three BCTs slated to re-deploy from Iraq in June 2009, as well as withdrawals of other smaller units, and some downsizing of replacement units.³² This could bring monthly average troop strength to about 135,600 in FY2009, a 14% decrease from the previous year, close to DOD's estimate of 140,000 (see and Table A-1).³³

CRS assumes that additional re-deployments would not resume until the end of January 2010 after the Iraqi elections slated for December 2009. In order to reach the target of 35,000 to 50,000 troops in Iraq by August 2010 announced by President Obama, an additional 8 average BCTs would need to be withdrawn from Iraq.³⁴ Reflecting this schedule, average troop levels in Iraq would be about 88,300 in FY2010, a decrease of 35% that year (see Table 2 and Table A-1). DOD estimates average troop strength in Iraq would be about 100,000 in FY2010, somewhat higher than this CRS estimate.³⁵

In his speech to Marines at Camp Lejeune, President Obama announced that all troops would be out of Iraq by the end of 2011 as required by the Security Agreement.³⁶ To meet that deadline, CRS assumes, conservatively, that the last five remaining BCTs are withdrawn in the last five months of 2011, spanning the end of FY2011 and the first quarter of FY2012. This also reflects recent statements by Secretary Gates that the United States would have "a significant presence for another 18 months."³⁷

To meet this schedule, two BCTs would need to be withdrawn in August and September 2011 — the end of fiscal year — with the final three BCTs leaving between October and December 2011 during the first quarter of FY2012. This schedule would reduce average monthly troop strength from 88,300 in FY2010 to 42,750 in FY2011, and 4,050 in FY2012 (Appendix A and Table A-1). If troops were withdrawn sooner, or the agreement with the Iraq government was re-negotiated to extend DOD troop presence, average troop levels and costs would be lower or higher.

The FY2009 Supplemental and FY2010 War Request

Changes in average troop strength may provide a benchmark that can be used to evaluate the reasonableness of DOD requests for war funds along with other considerations. Although the services develop war requests by evaluating specific funding requests from the field rather than from a "top-down" approach, few would argue that changes in the number of deployed troop levels play a major role in determining war costs.

Along with other adjustments, CRS estimates how costs in FY2009 and beyond would change based on the number and operational costs of average Brigade Combat Teams in Afghanistan and Iraq each fiscal year based on FY2008 data (see Appendix B and Table B-1). Operational costs include military personnel and Operation and Maintenance (O&M).

Factors other than changes in troop strength, of course, also affect costs, including one-time expenses in the previous year that would be unlikely to be repeated, transfers of costs previously funded in supplementals to DOD's baseline budget, policy changes, lags in when savings from withdrawals would occur, the timing of new demands, and programmatic changes only loosely tied to troop levels (e.g. funds to train Afghan security forces).³⁸

Adjustments Affecting FY2009

On April 9, 2009, the Obama Administration submitted a request for an additional $75.5 billion for DOD for FY2009 Supplemental Overseas Contingency Operations, abandoning the "Global War on Terror" label used by the Bush Administration.³⁹ This brings the total request for FY2009, the current fiscal year, to $141.4 billion. For FY2010, the Administration has requested $130 billion in FY2010 for both wars.⁴⁰

Based on these requests, the proposed FY2009 level would be $45.7 billion lower than the $187.1 billion enacted in FY2008. At first glance, this 24% decrease appears to substantially exceed the 1% decrease in troop levels between those two years. To estimate how FY2009 would change compared to FY2008, several one-time decreases may be appropriate including:

• $12.2 billion provided by Congress in FY2008 that DOD characterizes as not related to war needs that would not be requested the following year, ranging from C-17 transport aircraft added by Congress ($3.3 billion), base closure or BRAC funding ($1.3 billion), and compensation for higher fuel prices for DOD's baseline program ($4.0 billion);⁴¹
• $16.8 billion for a largely one-time purchase of the full requirement for Mine-Resistant Ambush Protected (MRAP) vehicles in order to speed the delivery of these vehicles considered to have greater protective qualities than High Mobility Uparmored Wheeled Vehicles (HMMWV);
• a policy change by the new Administration that restores the definition of reconstitution — the repair and replacement of war-worn equipment — to procurement strictly to replace war losses and replenish supplies within the next fiscal year, reducing war-related procurement by $27 billion; and
• adjustments to troop levels for increases in Afghanistan and decreases in Iraq (see Table 3).

With these adjustments, CRS estimates that DOD's FY2009 war needs would total $133.1 billion rather than the $141.1 billion, or $8.3 billion below the Administration's request (see Table 3).

For changes in troop strength, this estimate assumes that operational costs in FY2009 rise or fall in proportion with changes in troop levels. Other programmatic changes may not be related to troop levels. Increases could include:

• the proposed $1.9 billion increase for the training for both Afghan and Iraqi security forces;
• additional demands for unmanned aerial vehicles or light-weight MRAPS;
• an additional $300 million for Coalition Support; and
• $400 million for the new Pakistan Counterinsurgency Fund.
• At the same time, other areas may have lower costs in FY2009 as needs change or are already met, including:
• $400 million less for Military Construction; and
• $800 million less for the Improvised Explosive Device Defeat Fund.⁴²

DOD's request makes adjustments for several of the changes outlined above and assumes small changes in military personnel costs and the same level of Operation and Maintenance funding in FY2008 as in FY2009.⁴³

Potential Changes to DOD's FY2010-FY2012 Requests

The Obama Administration requested $130 billion for war funding in its FY2010 budget that Congress is considering along with DOD's baseline request.⁴⁴ The funding needed for that year depends on not only changes in troop strength but other adjustments made by DOD.

In FY2010, according to Secretary of Defense Robert Gates, DOD transferred to its baseline about $8 billion of programs previously funded in supplemental appropriations. Anticipating that supplemental or war funding would be reduced, DOD transferred programs considered necessary for DOD's long-term requirements such as additional funding for increasing the size of the Army, recruiting and retention, countering threats from Improvised Explosive Devices (IEDs), Intelligence Surveillance and Reconnaissance (e.g., unmanned aerial vehicles), and "Global Train and Equip" funds for countries facing terrorist threats.⁴⁵

Because of these transfers, the FY2010 war request would be expected to be $8 billion lower than in FY2009. In addition, CRS estimates that planned changes in troop levels would also reduce requirements by an additional $7.7 billion for a total of $15.7 billion altogether. This could reduce DOD's requirement from $133 billion in FY2009 to $117 billion in FY2010, or almost $13 billion less than DOD has requested (see Table 3).

For FY2011 and FY2012, the Administration included planning figures of $50 billion each year in its budget.⁴⁶ Based on currently announced plans, CRS estimates that war costs could total $92 billion in FY2011 and $70 billion in FY2012. These levels would be $42 billion and $20 billion, respectively, above the current planning figures. While other factors could affect costs, this suggests future war requests are likely to change (see Table 3).

Potential Costs of Withdrawal

Some observers would argue that the assumption that war costs in Iraq will fall in tandem with troop levels overlooks that some costs are likely to fall more slowly than others, and that there may be additional costs associated with the withdrawal itself. Budgeteers sometimes assume that in the short-term, only direct support costs (e.g., special pays, operating costs, meals, protective gear) change with troop strength. Then once some threshold is reached, indirect costs (e.g., base support, communication hubs) fall, and reductions will become proportional.

Current efforts to reduce and move U.S. troops outside major cities in Iraq, required by the Security Agreement, may have already begun the consolidation process, which could reduce base support costs sooner rather than later.⁴⁷ In fact, there is some evidence that base support costs are already declining. In addition, the transfer of responsibility to Iraqi Security forces may also reduce U.S. operating costs even before there are major declines in troop levels.

Other observers have suggested that the withdrawal process itself may entail additional costs. Government Accountability Office (GAO) has emphasized the complexity and need for coordination and planning to re-deploy forces and equipment, but has not estimated costs associated with a drawdown.⁴⁸ Army leaders have testified frequently that the cost of "reset" or repairing and replacing war-worn equipment could persist at the current annual level of $16 billion for two years after a withdrawal from Iraq though the specific rationale for this argument is unclear.

For the past two years, the cost of repairing and replacing Army equipment has been $16 billion each year, a cost covered in war funding received each year.⁴⁹ About three-quarters of the Army's equipment rotates with units each year, being repaired on its return to the United States. This suggests that about one-quarter of unit equipment remains in-theater to be used by newly deploying units (based on the value of the equipment).⁵⁰ Once U.S. forces in Iraq fully withdraw, this equipment would be returned to the United States for repair or replacement except for items left behind for the Iraqis or transferred to Afghanistan.

If the repair and replacement bill were proportional to the value of the equipment, the Army would face an additional bill of about $5 billion to $6 billion to reset the equipment that has remained in theater, well below Army estimates. At the same time, the annual bill for repairing equipment that rotates with troops would be expected to decline as troop levels fall, offsetting some of this additional expense.⁵¹ Similar factors would presumably apply to the other services.

Potential Questions About Troop Strength and Cost

In looking at the effect of changes in troop levels on cost, CRS took into account the following questions:

• How many and how quickly are increases and decreases in troop levels planned for each fiscal year?
• How many troops are being increased and decreased in-country?

CRS distributed all operational costs over the number of troops in-country and assumed that troops in the region would fall at the same rate as those in-country, which is supported by past changes (see Figure 5 and Figure 7).

To do a more precise estimate, additional information would be needed. These question include:

• How did the number of troops in the region change?
• How are changes in troop levels distributed among the services since the war cost of an Army soldier in-country differs from a Navy sailor or Air Force pilot operating from outside Afghanistan or Iraq?
• How quickly will operational costs (Military Personnel and Operation & Maintenance) adjust to decreases in the number of troops deployed to Iraq or increases in Afghanistan, i.e. will some types of costs fall slower than others?
• How large are additional costs associated with re-deploying troops, such as sending troops and equipment home and repairing, replacing, or upgrading equipment currently stored in theater?
• How large is any new investment to build or upgrade bases to support higher levels of troops in Afghanistan, and are bases intended to be temporary or permanent?

Thus far, DOD has provided only very general information about troop levels rather than the number, location, or distribution among the services of troop levels which would help Congress to evaluate its supplemental war requests. The FY2009 National Defense Authorization Act requires that DOD provide separate budget displays for Afghanistan and Iraq as well as "include a detailed description of the assumptions underlying the funding for the period covered by the budget request, including the anticipated troop levels, the operations intended to be carried out, and the equipment reset requirements necessary to support such operations."⁵² These separate budget displays and detailed assumptions are not included in DOD's FY2009 Supplemental or the FY2010 war requests.

Ways to Measure Troop Levels for the Afghan and Iraq Wars

While CRS used Boots on the Ground or in-country troop levels to estimate future troop levels and potential effects on war costs, these figures do not accurately capture the total demand for troops for the Afghan and Iraq war because other troops deployed in the region are not included. To get a better sense of changes in the total demand for troops since the 9/11 attacks, CRS calculated average monthly strength for troops deployed for the Afghan and Iraq Wars using the following five DOD data sources, which range from including only troops in-country to all those deployed for OEF and OIF:

• Boots on the Ground or (BOG) Report, a once-a-month headcount limited to U.S. troops in-country that is compiled by the Joint Staff, Joint Chiefs of Staff (JCS) using inputs from the services;
• Operations Report, a more inclusive monthly headcount collected by Central Command, that counts service members under U.S. Central Command which captures some, but not all, military personnel deployed for OEF or OIF (e.g., Special Operations Command is excluded).
• Combat Pay Estimate, based on the funding for combat or Imminent Danger Pay that is reported in DOD's war cost reports and used in budget justification materials;
• Average Strength Report, a person-year estimate of all OEF and OIF military personnel collected by the Defense Manpower Data Center (DMDC) based on its Contingency Tracking System (CTS), which reflects service personnel records and the beginning and ending dates of each service member's deployment; and
• Location Reports, a monthly DMDC count of all those deployed for OEF and OIF that reports the country where service members are located.

Yet another demand for troops resulting from OEF and OIF operations is the activation of some 50,000 reservists to backfill some of the positions of deployed active-duty troops.⁵³ Although these reservists are not carrying out OEF and OIF missions, their activation would not have occurred without those operations. Adding those troops to the 294,000 included in the Location Report for December 2008 brings the total troops associated with OEF and OIF to 344,000, or almost twice the level in the Boots on the Ground report for December 2008 (see Figure 1 and Table D-1 and Table D-2).

Policy Issues Raised by Differences in Troop Counts

These different measures of troop strength for the Afghan and Iraq Wars raise several significant policy and cost issues:

• How will the Administration's plan to increase troops in Afghanistan and decrease forces in Iraq affect troops providing support in the region, i.e. will forces in the region change in proportion to changes in combat troops?
• What do war requests assume about reducing troops in-country as opposed to reducing troops in the region, and are there interactions between the two?
• How does the United States envision the overall U.S. military presence in the region in terms of deployed troops and their locations?

A number of other questions arise for each individual operation including:

• To what extent are increases in troop levels in Afghanistan dependent on decreases in Iraq, taking into account all troops deployed?
• What can we learn from the past about the effect of U.S. troop levels on the likelihood that the U.S., its allies, and local security forces can quell violence in Afghanistan or in Iraq?

Trends in Troop Levels From FY2002-FY2008 Using Five Alternative Sources

Although there are wide differences in troop levels reported by these five DOD sources, trends over time are similar (see Table 4 and Figure 2). Some of the differences between sources reflect definitions — for example, Boots on the Ground Reports for Afghanistan and Iraq include only those troops located within each country, whereas Operations Reports include some of the military personnel providing support in the region.

Other differences reflect whether figures are headcounts taken once a month as opposed to average strength which takes into account the number or days each member is deployed for OEF or OIF. These distinctions can have important implications for the cost and demand for troops. In general, more comprehensive measures are more useful.

• To get a better sense of monthly trends, CRS calculated averages for each DOD source for each fiscal year rather than using the more commonly cited Boots on the Ground figures which captures only a point in time. Using this method, average monthly troop strength for both wars in FY2008 ranged from:
• 188,000 for DOD's Boots on the Ground or troops deployed in Afghanistan and Iraq;⁵⁴
• 223,000 in the Central Command's Operations Report including some but not all troops deployed in the region, or 19% above the BOG total;
• 248,000 in average strength estimated from combat pay, or 32% above the BOG total;
• 294,000 in the Defense Manpower Data Center average strength or 56% above the BOG total; and
• 307,000 troops based on DMDC's Location Report, or 64% above BOG figures; (see Table 4).⁵⁵

Generally, figures in Average Strength Reports are higher than once-a-month headcounts because they include all military personnel providing theater support in the region, reflect the time personnel are deployed, and capture those present throughout the month. It is not clear, however, that these reasons are sufficient to explain the difference of over 100,000 between Boots on the Ground and Average Strength Reports.

Like the average strength measure, Location Reports are also inclusive measures because they capture all troops deployed at any point during a month. Location Reports are probably higher than Average Strength Reports because military personnel who are in-country for short periods of time are counted the same way as those present for the entire month.

There are also substantial differences by service between BOG and the more inclusive Average Strength Reports. Average Strength Reports figures are about:

• 30,000 to 35,000 or over 20% higher for the Army;
• 13,000 to 20,000 or 60% to 90% higher for the Navy;
• about 8,000 or about 30% higher for the U.S. Marine Corps; and
• 6,000 to 7,000 or over 25% higher for the Air Force.

Trends for both the Afghan and Iraq Wars

Despite these substantial differences in the number of military personnel who are counted, the overall trends over time in troop levels are similar for the five DOD sources. Overall, troop levels jumped sharply with the initiation of combat operations for the Afghan war starting in FY2002 and with the Iraq invasion in FY2003.

After that, troop levels generally remained fairly stable from FY2004-FY2006. The exception is Combat Pay Estimates, which are somewhat erratic probably because of the unreliability in DOD war cost reporting.⁵⁶ CRS included this measure because, if accurately reported, combat pay is a good measure of average strength and is often used by the services in its budget justification materials, including the current FY2009 Supplemental.

Figure 2. Average Monthly Troop Levels for the Afghan and Iraq Wars, FY2002-FY2008: Five DOD Sources

Sources: Joint Staff, Joint Chiefs of Staff, "Boots on the Ground" (BOG) reports; Central Command, "Operations Report;" Defense Finance Accounting Services (DFAS), "Supplemental & Cost of War Execution Reports;" Defense Manpower Data Center (DMDC), DRS 21198, "Average Number of Members Deployed on any given day by Service Component and Month/Year," January 2008; DMDC, DRS 11280, "Modified Location Country Report ," December 2008. Notes: CRS calculated monthly averages from the BOG, Operations, Average Strength and Location reports for each fiscal year. To estimate average strength from DFAS war cost reports, CRS divided the total amount reported for Imminent Danger Pay (IDP) by the annual amount that each member would receive.

In FY2007 and FY2008, overall troop levels increased moderately in response to the troop surge in Iraq announced by former President Bush in January 2007 and continued increases in troop levels in Afghanistan (see Table 4 and Figure 2).

Allocating Troops Between the Afghan and Iraq Wars

Despite the wide differences among DOD sources in the number of troops deployed for OEF and OIF since FY2002, the allocation of troops between the two operations is similar. Before the invasion of Iraq in March 2003, all troops deployed in the theater were allocated to OEF.

Since then, CRS found that the split for troops between the two operations was generally about 88% to 90% for the Iraq war and 10% to 12% for the Afghan war over the last several years based on an analysis of three DOD sources, the Joint Staff's Boots on the Ground, Central Command's Operations Reports, and DMDC's Location Report (Figure 3 and Figure 4).

Figure 3. Share of Deployed Troops for Iraq in Boots on the Ground and Operations Reports, FY2003-FY2008

Sources: Joint Staff, Joint Chiefs of Staff (JCS), "Boots on the Ground Reports," Summary Table, September 2001-October 2008; Central Command, "Operations Report," September 2001-November 2008. Notes: CRS calculations of shares from sources above. The "Operations Report" includes all personnel for OEF and OIF who are assigned to the U.S. Central Command including personnel in-country and some of those in neighboring areas.

Roughly, six or seven of every eight troops have been dedicated to the Iraq War and one or two of every eight to the Afghan war. While troop levels increased in Iraq during the surge or temporary buildup of troops in 2007 and 2008, troop levels also rose in Afghanistan so that the percentages remained similar to previous years (see Table 4 and Table 5 and Figure 3 and Figure 4).

Figure 4. Share of Deployed Troops for Afghanistan in Boots on the Ground and Operations Reports, FY2002-FY2008

Sources: Joint Staff, Joint Chiefs of Staff (JCS), "Boots on the Ground Reports," Summary Table, September 2001-October 2008; Central Command, "Operations Report," September 2001-November 2008. Notes: CRS calculations of shares from sources above. The "Operations Report" includes all personnel for OEF and OIF who are assigned to the U.S. Central Command including personnel in-country and some of those in neighboring areas.

CRS also calculated OEF and OIF troop shares using DMDC's Location Reports and guidelines developed primarily by the Joint Staff and Central Command that designate particular countries as part of OEF or OIF. This count showed the same proportions as the Boots on the Ground and Operations Report of about 12% for OEF and 88% for OIF for FY2006-FY2008.⁵⁷

While allocating personnel to an operation by country has certain limitations because some military personnel in a particular country may support either or both OEF and OIF, personnel in particular countries generally support only one operation (see Box 1).

Although JCS and Central Command have adopted guidelines for their own reporting that assign personnel to OEF and OIF by location, the Comptroller in the Office of the Secretary of Defense has not set specific guidelines for the Services about how or whether to use location as a way to assign war costs for military personnel. Because there is not a single standard for assigning personnel or other war costs to OEF or OIF, troop and cost allocations may not match.

This appears to reflect the assumption in DOD's financial regulations that "Ultimately, each DOD Component is responsible for the accuracy and completeness of financial information in the reports which present the financial effects of its operations."⁵⁸ Thus while the Comptroller's office works with the services to improve the accuracy of war cost reporting, and requires that each service develop and justify its method, it has not issued policy guidelines or regulations that would ensure commonality.

Troop Levels for the Afghan War Since October 2001

After initial combat operations in FY2002 were complete, troop levels for the Afghan war increased steadily from FY2003 to FY2008 though the pace of increases varies among DOD sources. The lowest DOD figure, Boots on the Ground, tripled from an annual average of 10,400 in FY2003 to 30,100 in FY2008 while the more inclusive estimate, average strength, doubled from 21,000 in FY2002 to 42,000 in FY2008 (see Table 5 and Figure 5.).

These changes in troop levels in the Afghan war appear to reflect several stages of military operations:

• initial combat operations supporting the Northern Alliance that resulted in the overthrow of the Taliban forces by November 2001;
• relatively low-level U.S. operations against the remaining Taliban insurgents from FY2003 – FY2006; and
• adoption of a counter-insurgency approach in response to increasing levels of violence in 2006 as the Taliban regrouped and adopted the suicide and Improvised Explosive Device (IED) tactics of Iraqi insurgent groups.

Initial Combat Operations

The Afghan War, or Operation Enduring Freedom (OEF), began on October 7, 2001 with U.S. troops launching air strikes and Special Operations forces supporting operations by the Northern Alliance against Taliban forces in response to the 9/11 attacks. President Bush announced that major combat operations were complete with the overthrow of the Taliban regime in November 2001. Under UN sponsorship, a transition government was formed, succeeded by an elected government on November 3, 2004.⁵⁹

During FY2002, the first year of the war, an average of 5,200 troops were deployed in Afghanistan according to Boots on the Ground figures, the report capturing troops most likely to face combat. For that same year, the Operations Report, which includes some but not all support troops in neighboring areas, shows an average of 8,800 troops. The higher figure in the Operations Report also more accurately shows the initial troop buildup beginning in October 2001 rather than in January 2002 as shown in the BOG reports.

Neither of these headcounts of 5,200 in the Boots on the Ground Report or 8,800 troops in the Operations Report capture all troops deployed to launch Operation Enduring Freedom, particularly Air Force and Navy personnel operating offshore or outside of Afghanistan. A CRS allocation of the Average Strength Report suggests that some 83,400 troops participated in OEF in FY2002, including 33,000 Navy personnel on ships, and 30,000 Air Force personnel. The Operations Report shows only 100 Navy personnel and 2,000 Air Force personnel (see Table 5 and Figure 5).⁶⁰

Troop Levels Grow Steadily From FY2003-FY2006

After initial combat operations were complete in November 2001, troop levels for the Afghan War grew gradually but steadily. Boots on the Ground headcounts increase by about 5,000 troops a year between FY2003 and FY2005, then inch upward to 20,400 in FY2006. According to the more inclusive Operations Report, OEF troop levels rose from 12,500 in FY2003 to almost 20,000 by FY2004, remaining at about that level for the next two years (Table 5 and Figure 5).

Because the Operations Report includes some military personnel in neighboring countries, these figures would be expected to be higher than Boots on the Ground reports, and generally OEF troop levels are between 1,000 to 5,000 higher than Boots on the Ground reports. This pattern is reversed for several months in FY2008, however, apparently because DOD changed the definition of personnel covered in the BOG report (see Table 5 and Appendix D and Table D-1).⁶¹ These variations suggest a need for better definitions that would clarify who is and is not counted as part of OEF and OIF.

Figure 5. Average Monthly Troop Levels for the Afghan War, FY2002-FY2008: Five DOD Sources

Sources: Joint Staff, "Boots on the Ground Reports", Summary Table, Sept. 2001-October 2008; Central Command, "Operations Report," September 2001- November 2008; Defense Manpower Data Center (DMDC) DRS Report 21198, "Average Number of Members Deployed on Any given Day, by Service Component and Month/Year," (computed by dividing the days deployed per month by the number of days in the month), January 2009; DMDC, DRS 11280, "Modified Location Country Report." December 2008. Notes: CRS calculated monthly averages from the BOG, Operations, DFAS, Average Strength, and Location reports for each fiscal year. To estimate average strength from DFAS war cost reports, CRS divided the total amount reported for OEF Imminent Danger Pay (IDP) by the $2,700 that each member would receive per year. CRS estimated the average strength for OEF by applying the percentage shares for each fiscal year in the Operations reports to the monthly averages reported in DMDC's Average Strength Report. Location Report figures for FY2002-FY2005 are not included because DMDC does not consider location data in its Modified Location Country Report to be reliable until January 2005 because service members often were shown in their first location in the region rather than in their actual deployment. After that time, a system was developed where members recorded each location. In April 2009, DMDC "scrubbed" the data to eliminate double-counts of members who were deployed more than once in one or multiple locations within the same month; this eliminated all but 400 in duplications. CRS allocated countries to OEF based primarily on guidelines developed by the Joint Staff and Central Command (see Box 1).

To capture all troops providing theater-wide support in the region or engaged in other OEF counter-terror operations, CRS applied the shares shown in the Operations Report to the more complete figures compiled in the Average Strength Reports. Because the Average Strength Reports reflect each service's reporting of benefits received and deployment dates, it is likely to be a more accurate measure of troop strength.

At the same time, the similarity in OEF and OIF shares shown in BOG, Operations and Location Reports, suggests that applying shares to the Average Strength Reports would be a fairly reliable way to estimate OEF and OIF troop strength. According to this Average Strength estimate, OEF troops grew from 21,100 in FY2003 to 28,400 in FY2006 and 42,500 in FY2008. The Location Report shows similar figures (see Table 5).

CRS also estimated troop levels by using Combat Pay reported in DOD's "Supplemental & Cost of War Reports." The sharp changes from year-to-year suggest that allocations between OEF and OIF may not be reliable, a criticism raised by GAO (see Table 5 and Figure 5).⁶² CRS includes this estimate despite its apparent inaccuracy because strength and combat pay would be expected to be consistent, and DOD uses this measure in its justification material for the FY2009 Supplemental.⁶³

Further Increases As Violence Rises After 2006

In 2006 and 2007, the level of violence in Afghanistan grew as Taliban insurgents adopted the suicide attacks and roadside bombing of insurgents in Iraq.⁶⁴ The renewed increase in troop levels in FY2006 and FY2007 appears to follow higher levels of violence as measured in:

• the number of security incidents and attacks, collected by DOD, the United Nations, and Central Command;
• attacks on non-combatants collected by the State Department;
• number of incidents involving IEDs and roadside bombs;
• number or suicide bombings; and
• number of U.S. and civilian casualties.⁶⁵

All the different measures show increases in OEF troop strength in FY2007 and FY2008. After the modest increases in the prior two years, headcounts in both Boots on the Ground and the Operations Report rose by 10,000 between FY2006 and FY2008, reaching over 30,000.

These figures probably do not capture all deployed troops. The more comprehensive Average Strength and Location Reports show an increase from 28,000 in FY2006 to 42,000 in FY2008, or roughly 5,000 more troops each year. (see Table 5).⁶⁶

Since May 2008, in response to a congressional reporting requirement, DOD has included figures showing monthly headcounts for OEF as well as military personnel in Afghanistan in its Boots on the Ground report.⁶⁷ For the last several months of FY2008, this new OEF figure shows an average of 47,790 for OEF, higher than either the Average Strength and Location Reports, and almost 18,000 above in-country headcounts.⁶⁸ The reasons for these higher figures, submitted by the services, are not apparent (see Table 5).

If these new DOD figures better reflect current OEF troop levels, then total troops for the Afghan War, including U.S. and coalition troops in Afghanistan, and U.S. troops in the region, would average 97,000 in FY2009. That total could grow to about 110,000 by the following year.⁶⁹

Seasonal Variations

While average monthly troop levels may more accurately reflect strength levels, once-a-month headcounts may better capture buildups and draw downs within a year as well as some longer-term trends. For example, between January and September 2002, troop levels shown in the Operations Report more than doubled from 4,100 to 10,400 as U.S. forces grew during initial combat operations. Despite month-to-month variations, a clear upward trend emerges for OEF in the Boots on the Ground and Operations Reports headcounts (see Figure 6 and Table D-1).

During FY2003, U.S. troop levels hovered between 11,000 and 13,000 from month to month. Monthly variations may reflect seasonal variations and planned operations. For example, troop levels temporarily spiked in the fall and winter of 2003 and 2004 when the U.S. and Afghan forces conducted several operations against Taliban insurgents – Mountain Viper (August 2003), Operation Avalanche (December 2003) and Operation Mountain Storm (March-July 2004). In FY2004, troop levels grew to around 20,000 while the U.S. and Afghan troops continued operations against remaining Taliban insurgents at relatively low levels.⁷⁰

Typically, troop levels have increased for several months of the year -- sometimes early in the year in order to prepare for spring Taliban offensives when the weather is better. Between FY2003 and FY2008, troop levels generally rose in the winter and spring, then declined slightly in the summer and fall, settling back at somewhat higher troop levels each year (see Figure 6 and Table D-1).

Figure 6. Monthly Boots on the Ground in Afghanistan and Troops Deployed for OEF Operations, September 2001-November 2008

Source: Joint Staff, Joint Chiefs of Staff (JCS), "Boots on the Ground Summary Table" provided to CRS, September 2008; Central Command, "Operations Report," September 2001-November 2008. Notes: Trend line inserted by CRS. Boots-On-the-Ground" captures deployed military personnel in-country while OEF also includes military personnel in the region that support that operation and are assigned to Central Command. DOD did not send Congress BOG reports for October and November 2008.

After falling by several thousand in the fall of 2006, U.S. troop levels rose to over 25,000 the following spring and summer as the situation in Afghanistan worsened. After a brief decline in the fall of 2007, troop levels resumed their upward climb reaching about 33,600 in the spring and summer of 2008 where they remained for the rest of the fiscal year (see Figure 6 and Table D-1).

Troops from Other Countries

Just as U.S. troop levels have risen since 2006, contributions from other NATO countries have also increased. Average troop levels from NATO countries other than the United States have grown from about 20,000 in FY2007 to 28,000 in FY2008, and 31,000 in FY2009 thus far.⁷¹ While the United States has requested additional troops from NATO countries, increases are likely to be small.

Troop Levels for the Iraq War Since the March 2003 Invasion

In the fall of 2002, the United States began to build up troop levels in the Gulf region in preparation for the invasion of Iraq that took place in late March 2003. After peaking in the spring and summer of 2003, troop levels then fell rapidly.

Between FY2004 and FY2006, average troop levels reached a steady-state of about 130,000 to 140,000 on the ground in Iraq and from about 210,000 to 230,000 for OIF as a whole according to estimates of DOD average strength.

During the surge in troops initiated by President Bush in January 2007 and continuing through much of FY2008, troop levels in Iraq increased somewhat more steeply — from 140,000 to 158,000 in-country and from 233,000 to 248,000 (see Table 6 and Figure 6).

Changes in troop levels in Iraq appear to be loosely related to changes in military strategy including:

• a quick peak for the invasion itself followed by a rapid drawdown;
• limits on troop strength reflecting the military strategy endorsed by General Abizaid, the commander in-country, to minimize military presence in order to prevent insurgents from using the U.S. occupation as a way to gain popular support;⁷²
• a shift in the U.S. strategy in October 2005 announced by Secretary of State Condoleezza Rice to "clear, hold, and build" so as to "clear areas from insurgent control," that would gradually be implemented as U.S. troops moved out of large bases to work more closely with Iraqi forces;⁷³ and
• the surge in troops announced by President Bush in January 2007 that temporarily increased the number of troops by 30,000 and the number of Brigade Combat Teams from 15 to 20 in order to implement the new counter-insurgency policy to help Iraqis clear and secure neighborhoods.

Initial Combat Operations

While Boots on the Ground headcounts show average strength of 67,00 in FY2003, other measures that capture Navy and Air Force personnel outside Iraq show strength levels that are almost two or three times as high (see Table 6 and Figure 7).

Month-to-month headcounts show the buildup for initial combat operations, followed by a fairly rapid draw down. Troop levels for the invasion peaked at 149,000 according to DOD's Boots on the Ground reports while troop levels for OIF reached a highpoint of 285,000 in April 2003 — almost twice as high as the number of U.S. troops in-country (see Table D-2).

Troop levels in Iraq itself may not have reached the 285,000 peak deployed in the region partly because many U.S. troops who were en route, at sea, or deployed to Kuwait may not have been sent into Iraq. For example, Secretary Rumsfeld and General Franks decided to turn back the 1^st Cavalry Division, which had been scheduled to enter Iraq.⁷⁴ Although monthly averages smooth out the peaks and valleys of the invasion and its aftermath, the OIF Operations Report for FY2003 is still twice as high as the Boots on the Ground (see Table 6).

After former President Bush declared an end to major combat in Iraq on May 1, 2003, troop levels in Iraq dropped from the invasion peak of 149,000 to 130,000 in September 2001. OIF troop levels fell from 285,000 in April to 170,000 in October 2003. The gap between those in Iraq and those part of the OIF operation narrowed from 140,000 during the invasion to 40,000 by that summer, possibly because of the re-deployment of forces (see Table D-2).⁷⁵

In March 2004, OIF troop levels spiked again to 270,000 compared to 120,000 in Iraq for reasons that are not clear. One possibility is that additional troops may have been deployed to the region in anticipation of problems with the transfer of power from the U.S. occupation force to an Iraqi government by June 30, 2004.⁷⁶

Based on the Average Strength estimate, troop levels for OIF reached about 216,000 in FY2003, considerably higher than either headcount measure, probably because this approach captures troops outside Iraq. The estimate for OIF based on Imminent Danger Pay may be unreliable because of inaccurate war cost reporting (see Table 6 and Figure 7).

Troop Levels Plateau From FY2004-FY2006

From FY2004 through FY2006, average troop levels for the Iraq War changed little under all measures, generally varying by several thousand to about 10,000 from year to year. While Boots on the Ground headcounts grew from 130,600 to 141,100, the Operations Report for OIF fell from 196,900 to 181,300 in those years for reasons that are unclear. The OIF share of average strength rose by over 20,000, from 211,600 in FY2004 to 234,100 in FY2006 (see Table 6).

The gap between Boots on the Ground and the Average Strength estimate ranged from about 70,000 in FY004 to 90,000 in FY2006. This gap probably reflects the buildup of in-theater support as the war continued. For OIF, the major regional support and headquarters countries include about:

• 45,000 to 50,000 in Kuwait;
• 16,000 to 20,000 on ships afloat;
• 9,000 to 10,000 in the Qatar headquarters base;
• 2,000 in the United Arab Emirates;
• 4,000 in Bahrain, a Navy regional headquarters; and
• 500 to 600 in Saudi Arabia.⁷⁷

Figure 7. Average Monthly Troop Levels for the Iraq War, FY2002-FY2008: Five DOD Sources

Sources: Joint Staff, "Boots on the Ground Reports", Summary Table, Sept. 2001-October 2008; Central Command, "Operations Report," September 2001- November 2008; Defense Manpower Data Center (DMDC) DRS Report 21198, "Average Number of Members Deployed on Any given Day, by Service Component and Month/Year," (computed by dividing the days deployed per month by the number of days in the month), January 2009; DMDC, DRS 11280, "Modified Location Country Report." December 2008. Notes: CRS calculated monthly averages from the BOG, Operations, DFAS, Average Strength, and Location reports for each fiscal year. To estimate average strength from DFAS war cost reports, CRS divided the total amount reported for OEF Imminent Danger Pay (IDP) by the $2,700 that each member would receive per year. CRS estimated the average strength for OEF by applying the percentage shares for each fiscal year in the Operations reports to the monthly averages reported in DMDC's Average Strength Report. Location Report figures for FY2002-FY2005 are not included because DMDC does not consider location data in its Modified Location Country Report to be reliable until January 2005 because service members often were shown in their first location in the region rather than in their actual deployment. After that time, a system was developed where members recorded each location. In April 2009, DMDC "scrubbed" the data to eliminate double-counts of members who were deployed more than once in one or multiple locations within the same month; this eliminated all but 400 in duplications. CRS allocated countries to OIF based primarily on guidelines developed by the Joint Staff and Central Command (see Box 1).

While troop levels were relatively stable --- hovering around 140,000 in Iraq itself and around 230,000 for OIF, the number of security incidents per week grew from about 500 in March 2005 to about 900 in September 2005. After the Samarra mosque bombing in March and April of 2006, security incidents rose more steeply to 1,000 per week, almost twice the level of the previous year.⁷⁸

Troop Surge in FY2007 and FY2008

By October 2006, weekly security incidents in Iraq reached 1,400, peaking at close to 1,600 in June 2007.⁷⁹ In reaction, then-President Bush announced in January 2007 that there would be a surge in the number of troops deployed to Iraq, an increase of 30,000 or five brigade combat teams over the next several months.

Because the Army plans deployments six to 12 months in advance in order to prepare and train units, the services have limited options to increase troop levels quickly in reaction to a deteriorating situation. One option, unpopular with the troops but which was used during the surge is to extend the tours of troops already deployed or to assign specialized troops for short assignments.

To carry out this troop increase, Secretary Gates formally extended tour lengths for all active-duty Army units in-country and those slated to deploy from 12 to 15 months, and coupled that extension with an assurance that troops would have 12 months "dwell time" at home between tours.⁸⁰

Because the increase in troop levels took place gradually over seven months and then was gradually reversed over the next ten months, average monthly troop strength for the Iraq War does not show an increase of 30,000. Between FY2006 and FY2007, average troop strength grew from 234,000 to 243,00 or by 9,000 for the year. Because the drawdown was also gradual with higher troop levels maintained for most of the fiscal year, the monthly average in FY2008 reached 251,000 or about 7,000 more than the previous year (see Table 6 and Figure 7).⁸¹

In April 2008 as the number of Brigade Combat Teams was falling from 20 to 15, President Bush announced that the Army would return to 12 month tours.⁸² DOD withdrew one additional Brigade Combat Team from Iraq in the fall of 2008, bringing the total to 14 BCTs as of FY2009 and deployed an additional combat brigade of about 9,000 troops to Afghanistan in January 2009. Further troop increases are underway in Afghanistan.

Changes from Month-to-Month

While Boots on the Ground figures do not capture all personnel dedicated to OIF operations, these are the figures, along with the number of Army Brigade Combat Teams, that are typically cited by policy makers in describing changes in troop levels. Although monthly reports give the most up-to-date information, the variations from month-to-month tend to reflect rotation patterns rather than military plans with increases in the fall and winter generally offset by decreases in the following months in Iraq (see Figure 8).

Figure 8. Monthly Boots on the Ground (BOG) in Iraq and Troops Deployed for OIF Operations, March 2003-November 2008

Sources: Joint Staff, Joint Chiefs of Staff, "Summary Table," September 2008; Central Command, "Operations Report," September 2001-November 2008. Notes: "Boots-On-the-Ground" captures deployed military personnel in-country while OIF also includes military personnel in the region who support that operation and are assigned to Central Command. DOD did not send Congress BOG reports for October and November 2008.

During the surge, the number of troops for the Iraq War grew by 30,000 from 138,000 to 169,000 in Iraq itself and from 185,000 to 219,000 for OIF between February and September 2007 (see Table D-2). By June 2007, with most of the first new troops in place, the military launched a series of operations to clear insurgents from other areas.⁸³

Between September 2007 and September 2008, this increase was gradually reversed with troop levels falling from 169,000 to 147,000 in Iraq itself and from 219,000 to 181,000 for the operation as a whole (see Figure 8 and Table D-2).

Using Different Troop Strength Figures

For those most concerned about the number of troops present in Afghanistan or in Iraq at a point in time or those most likely to be exposed to combat, Boots on the Ground Reports may be the best measure. On the other hand, these reports do not capture Navy and Air Force personnel operating from outside Afghanistan or Iraq, troops providing regional support for the Afghan or Iraq Wars or troops not present on the day of the head count.

While the Operations Report gives a more complete picture of those involved in OEF and OIF than the BOG Reports because it includes some personnel deployed outside of Afghanistan and Iraq, it excludes personnel who are not assigned to Central Command, and also has the limitations of once-a-month headcounts.

To capture the full demand for manpower, a key factor in estimating costs, Average Strength Reports—which measure person-years—may be the best measure because they capture not only all deployed troops in each fiscal year but also how long each one stays, as well as short-term tours (e.g., Air Force personnel flying mission, personnel assigned for special tasks). DOD routinely uses average strength in its budgeting for military personnel. On the other hand, the Average Strength Reports do not allocate service members between OEF and OIF, requiring CRS to estimate those allocations using other DOD figures.

While Location Reports capture all deployments for OEF and OIF, these reports are not reliable for each operation prior to FY2006, a disadvantage in looking for trends over time.⁸⁴

If DOD were to require that the services followed a standard set of guidelines about allocating military personnel between OEF and OIF, measures of troop strength would be more reliable and consistent. That, in turn, would give Congress and DOD better tools to assess troop demands for the Afghan and Iraq Wars, as well as the effects of changes in troop levels.

Few would dispute that changes in troop levels have a significant effect on war costs, whichever measure is used. The following section discusses potential ways that troop levels may be used to assess future war cost requests.

Service Roles and Readiness Concerns

Many observers and military spokesmen have raised concerns about frequent deployments of units for the Afghan and Iraq Wars, characterizing the Army, in particular, as "stressed" or "almost broken." Long and frequent deployments are cited as harming readiness not only because of effects on morale, but also because of the effects on having other units available and ready if another crisis arises. Although how often a unit is deployed is one way to look at the burden, it is useful to look at the effect on service members because the make-up of a unit changes between deployments as individuals leave for other assignments, or retire and are replaced by new personnel.

Concern has focused primarily on the Army, both active-duty, guard, and reserve units. Reflecting the ground focus of both wars, Army troops (active-duty, National Guard, and Reserves) made up 61% of all OEF and OIF troops in FY2008 with Marine Corps troops contributing another 12%. The remaining 27% are split between the Navy (15%) and the Air Force (12%) (see Figure 9).

Overall, active-duty forces made up 80% and reserve forces 20% of average strength during FY2008. For example, 47% were active-duty Army, 10% Army National Guard, and 5% Army Reserve. The Air Force made up another 12%, the Marine Corps another 12%, and the Navy 15%, almost all active-duty personnel (see Figure 9).

Figure 9. Service and Component Shares of OEF and OIF Troops in FY2008 As percent of total average strength

Source: Defense Manpower Data Center, DRS 21198,"Average Number of Members Deployed on Any Given Day, by Service Component and Month/Year," January 31, 2009; Defense Manpower Data Center, DRS 26354 "Average Military Strength by Service/Component and Fiscal Year, FY2001-FY2008; February 9, 2009 from Active Duty and RCCPDS Master Files. Notes: CRS calculations based on sources above.

Measuring the Burden of Deployment

Looking at the share of each service and component that is deployed gives one measure of the burden of deployment. Active-duty forces had substantially higher deployment rates than reservists. Of the 2.2 million active-duty and reserve U.S. military personnel serving in FY2008, some 12% or 260,000 troops were deployed for OEF and OIF based on average strength reports. Overall, some 17% of active-duty forces are deployed compared to 7% of reserve forces (see Table 7).

The highest deployment rate is for Army active-duty forces where over one-quarter are deployed (26%). Some 7% of Army National Guard and Army Reserve strength are deployed, similar to the overall rate for reservists.

Deployment rates are lower for the other services including:

• 16% for the Marine Corps;
• 12% for the Navy; and
• 8% for the Air Force (see Table 7).

Deployment rates for reserves in the other services were generally half or less than the rate for active-duty forces. For example, about 7% of the Army National Guard and the Army Reserve were deployed compared to 26% for active-duty Army in FY2008. Similarly, 4% of the Air Force Guard and Air Force Reserve were deployed compared to 8% for active-duty (see Table 7).

Deployment rates alone, however, do not capture the full impact of OEF and OIF on the services. In addition to those currently deployed, other troops are preparing or "training up" to deploy or recovering from a recent deployment. Taking those troops into account suggests that roughly twice as many troops are affected by OEF and OIF as the number deployed. In the case of the active-duty Army, probably half of the active-duty Army is either deployed, undergoing intensive training to deploy, or recovering from a deployment, and about one-third in the case of the active-duty Marine Corps.⁸⁵ The large share of the Army active-duty forces dedicated to OEF and OIF presumably underlies concerns about the Army's readiness to carry out other operations should they arise.

Appendix A. Actual and Estimated Troop Levels in Afghanistan and Iraq, FY2007-FY2012

CRS estimated average monthly troop levels in Afghanistan and in Iraq for FY2009-FY2012 based on the announcements made by the White House on February 17, February 27, and March 27, 2009, and Department of Defense press releases and conferences. ⁸⁶ Where future levels have not been identified, CRS made the conservative assumptions described below.

According to these announcements, the Administration is deploying 21,000 more troops to Afghanistan in FY2009. These troops are in addition to a combat brigade with support of about 9,000 that was approved by President Bush in December 2008. Together, this would increase troop levels in Afghanistan by 30,000 by the end of FY2009. Because Department of Defense Secretary Gates suggested that increases beyond that are unlikely, CRS assumes continuation of that level through FY2012 in this table.

In the case of Iraq, the Administration plans to reduce troops in Iraq by two Brigade Combat Teams in FY2009 by not replacing two brigades coming home in June 2009. In addition, several smaller units were withdrawn between December 2008 and April 2009.⁸⁷ DOD is also planning to downsize some replacement units.

According to Administration statements, troop levels in Iraq would decline to 35,000 to 50,000 by August 31, 2010, which CRS estimates as 45,000 troops or 5 BCT-equivalents (roughly the midpoint). These troops would be withdrawn by the end of 2011 as required by the Security Agreement between the United States and Iraq.

Conservatively, CRS assumes that these U.S. troops remain in-country until the last five months of 2011, the latest possible date that would meet the December 31, 2011 deadline for the withdrawal of all U.S. troops in the U.S.- Iraq Security Agreement. Based on that assumption, two BCTs would be withdrawn at the end of FY2011 (August and September 2011) and three in the first quarter of FY2012 (October – December 2011).⁸⁸

Based on these plans, CRS set up a schedule that estimates how long BCTs will be in-country. Taking into account the number of troops and their time in-country, CRS calculates the number of weighted average Brigade Combat Team equivalents in Afghanistan and Iraq in each fiscal year in Table A-1. CRS assumes that each BCT has 9,000 troops, 1,000 less than the Army's planning assumption in order to reflect the fact that Marine Corps units are generally smaller.

The schedule also assumes that troops are added or withdrawn at the end of each month, and that all Navy and Air Force troops in-country are providing support to Army and Marine Corps. This schedule does not reflect changes in the number of troops deployed to countries in the region or related operations.

To check the validity of this approach, CRS used this method to estimate troop levels in Iraq in FY2007 and FY2008 and then compared the estimate with reported BOG figures. The CRS estimates were close to reported average monthly Boots on the Ground. Estimates for FY2009-FY2012 build on the reported numbers of Boots on the Ground as of September 1, 2008, close to the beginning of FY2009 (see Table A-1).⁸⁹

Appendix B. The Cost of an Average Brigade Combat Team: FY2005 and FY2008

To estimate future war costs, budgeters typically look to past experience, and then make adjustments for changes in troop levels, new requirements, savings in infrastructure costs already covered or one-time expenses, or pricing changes. To estimate the effect of increases and decreases in troop strength, CRS used the cost of an average Brigade Combat Team in FY2008.

Between FY2005 and FY2008, overall war costs rose more rapidly than increases in the number of troops in-country so the cost of an average BCT has risen. This reflects both higher operational costs, and particularly rapid growth in investment spending. In some cases, such as the rise in investment spending, the Administration's new policy to limit war-related procurement to replacement of war losses and replenishment, is likely to reduce future costs. In other cases, such as the rise in infrastructure spending in Iraq, it is not clear whether to expect that experience to be duplicated in Afghanistan.

Because the Administration only provided information about future changes in the number of troops in Afghanistan and Iraq, CRS spread all war costs over the number of troops in-country. This essentially assumes that changes in the number of troops in-country will be matched by proportional changes in troops in the region, which may or may not be the case. While it is likely that Navy and Air Force personnel who support ground operations, and some of the Army support personnel in Kuwait would change in tandem with changes in forces in-country, other headquarters personnel may not change proportionately.

Definition of War Costs

War requests cover funding intended to cover the incremental or additional costs tied to paying and deploying U.S. military personnel to a war zone. For military personnel, this has included:

• the costs of special pays like Imminent Danger Pay and Separation Allowances;
• recruiting and retention bonuses;
• activating reservists (paying full-time rather than part-time salaries and benefits); and
• growing the size of the Army and Marine Corps.

Some of these expenses, like "growing the force" are to be transferred to DOD's baseline budget in FY2010, reducing war costs. Recruiting and retention bonuses could also be transferred and are likely to be lower because of the recession.

For Operations and Maintenance (O&M), war costs include:

• transporting personnel and equipment to and from the war zone;
• higher operating tempo in a war zone;
• repairing war-worn equipment;
• providing force protection gear to personnel;
• setting up and operating communication and intelligence assets; and
• building, maintaining and providing security at bases.

For procurement and Research, Development, Test & Evaluation (RDT&E) war costs have included the following types of items:

• replacement and upgrading of a wide range of war-worn equipment, defined broadly to include replacement with new weapon systems and upgrading of current systems already part of DOD's modernization plans (e.g. F-22s, Stryker brigades);
• equipment to accelerate the Army's plans to convert to more standardized modular units and the Marine Corps' plans to restructure its units;
• equipment for additional units in the Army and Marine Corps as part of DOD's "Grow the Force" initiative;
• new and upgraded force protection equipment based on experience in the field (e.g. Mine-Resistant Ambush Protected, or MRAPs and up armored High Mobility Wheeled Vehicles, or HMMWVs); and
• RDT&E related to war-related threats such as Improvised Explosive Devices (IEDs).

In its FY2009 Supplemental Request, DOD significantly changed its definition of reconstitution, restoring the traditional criteria which limited procurement to replacement of war losses and replenishment of war munitions and stocks. This resulted in a halving of DOD's procurement request in FY2009.⁹⁰

For military construction, war costs include not only building and upgrading bases in Afghanistan with new air strips, generating and water purification plants, and other structures, but also building roads.

Changes in Afghanistan Costs Between FY2005 and FY2008

Between FY2005 and FY2008, average monthly troop strength in Afghanistan rose by 58% while overall Operation Enduring Freedom (OEF) war costs almost doubled from $14.6 billion to $27.5 billion.⁹¹ Because costs outstripped the rise in troop levels, the average cost of a BCT in Afghanistan rose by 19% from $6.9 billion to $8.2 billion (see Table B-1).

Operational costs in Afghanistan — covering both military personnel and Operation & Maintenance (O&M) — rose less rapidly than investment costs. The 10% increase in overall operational costs reflects an increase of almost 50% in O&M costs offset by an 8% decrease in military personnel costs. The decline in average personnel costs in FY2008 may reflect less reliance on activating reservists, the largest single personnel cost.⁹²

In Afghanistan, O&M costs more than doubled for combat-related expenses (e.g. fuel and spare parts, depot maintenance) and trebled for base support costs, increasing far more rapidly than troop strength.⁹³ Increases in operating tempo costs may reflect the U.S. response to rising levels of violence while higher support costs may reflect the build-up of infrastructure in-country. It is not clear whether average costs will continue to rise or fall as the number of troops in-country increases.

Investment costs for replacing and upgrading equipment also more than trebled in the past three years, which may reflect the expanded definition of war-related costs more than changes in the operating tempo (see discussion of reconstitution below).

Changes in Iraq Costs Between FY2005 and FY2008

Between FY2005 and FY2008, average monthly troop strength in Iraq grew by 10% in Iraq while the average cost of deploying a BCT rose by 72%. The average cost of a BCT in Iraq rose from $4.6 billion to $7.2 billion, a 57% increase, a steeper increase than in Afghanistan (see Table B-1).

Like Afghanistan, overall average operating costs in Iraq rose by about 40% though both military personnel and O&M costs grew faster than troop strength. Unlike Afghanistan, average military personnel costs also increased, primarily because of the cost of growing the size of the Army and Marine Corps (no longer considered a war cost, transferred to DOD's FY2010 baseline budget).

O&M costs in Iraq grew at about the same rate as in Afghanistan – about 50%, with growth concentrated more in support than combat-related operating expenses. For example, the cost of base support rose by 41% while troop strength grew by 10%.

Most dramatically, investment costs doubled over three years, again more a reflection of a re-definition of war-related reconstitution rather than more intense wartime operating tempo (see discussion below).

Future War Costs

There are a variety of possible explanations for increases in average costs — higher benefits for deployed personnel, differences in combat intensity, rising spending to repair equipment, substantial support costs for large, well-developed bases, or expanded definitions of reconstitution.⁹⁴ Some of these increases may continue in the future and some may not.

Dramatic Growth in Reconstitution Costs Ends

The steepest increase in costs has been in investment accounts, particularly procurement, where the definition of resetting or reconstituting units after their war-related deployments was substantially changed between 2004 and 2008. These changes led to a dramatic increase in war-related procurement from $5 billion in to $7 billion in FY2004 to $45 billion in FY2008.CBO, GAO, CRS, and the Center for Strategic and Budgetary Assessment (CSBA) have all questioned whether much of this funding in supplemental appropriation acts for procurement, in fact, reflected war-related needs.⁹⁵

Beginning in 2005, the definition of war-related procurement was broadened to include paying for the conversion and upgrading of Army and Marine Corps units to a new modular or more standardized composition. In FY2005 and FY2006, DOD requested and received $5 billion each year for these restructuring initiatives underway before the Afghan and Iraq wars. The services argued that these changes would make rotations easier because of the greater similarity among units though the impact appears to be small.⁹⁶

A further expansion in the definition of war-related procurement also occurred in 2006 when the services were given new guidance that permitted them to include in war requests items needed for the "long war" and not be "strictly limited to Operation Enduring Freedom (OEF) and Operation Iraqi Freedom (OIF)."⁹⁷ This made it easier for the services to request and receive funding for new and upgraded equipment rather than to replace war losses or replenish munitions or other war stocks. DOD also changed its traditional guidance that limited reset or reconstitution to restoring units to their pre-war condition (e.g. with replacements for war losses) to upgrading equipment to meet new threats.

In a study of the Army's reset , CBO estimated that more than 40% of the Army's reset funding was not, in fact related to the repair and replacement of war-worn equipment.⁹⁸ A December 2008 CSBA study concluded that the adoption of the "long war" framework for evaluating war-related requirements "allows the Services to include virtually anything in their request for war-related appropriations," and "removed any principled distinction between what should be included in special war-related appropriations and what should be included in the base defense budget."⁹⁹

DOD also allowed the services to include in war funding requests some of the cost of increasing the size of the Army and Marine or "growing the force," which was originally intended to be temporary with a return to pre-war strength once the conflicts were over. With President Bush's decision in January 2007 to make these increases permanent, the rationale for considering these war costs weakened.

In its FY2009 Supplemental Request and FY2010 War Request, the Obama administration reversed this policy and limited war-related procurement to replacing war losses and replenishing war supplies, war-related procurement requests dropped in half in the FY2009 Supplemental and FY2010 war request.¹⁰⁰

In addition, DOD has objected to some Congressional additions for C-17 transport and C-130 aircraft, additional equipment for reserves, and other items contending that these systems are not related to war-time needs, and has not included those systems in its requests.¹⁰¹

Operational Cost Increases Slow

On the operational side, there are some indications that increases in operational costs may be slowing or in some cases, decreasing. For example, in the past two fiscal years during the surge, O&M costs rose only somewhat faster than in-country troop strength.¹⁰² The initial months of FY2009 also show decreases in OIF operational costs outpacing decreases in strength, which may reflect either lower levels of violence or consolidation of bases. At the same time, some operational costs for OEF continue to rise more rapidly than troop strength, which may reflect more intense combat and expansion of bases for more troops.¹⁰³

Estimating the average cost of a BCT based on troop strength in Afghanistan or Iraq could also overstate costs and savings if the number of troops deployed in the region did not rise or fall at the same rate as those deployed in-country. In FY2008, an additional 7,300 U.S. troops were deployed outside Afghanistan as part of OEF, and an additional 80,000 were deployed outside of Iraq as part of OIF.

At the same time, if troop levels are increased or reduced primarily in Afghanistan or Iraq, those costs are generally higher than the average because of combat and security costs associated with a hostile environment.

Potential Cost Effects of Changes in Troop Levels

In Afghanistan where troop levels rose substantially and combat has intensified, the average cost of a BCT rose more moderately than in Iraq. If this experience continues, the average cost of a BCT in Afghanistan could be expected to rise slightly unless there are offsetting savings as infrastructure costs are spread over more troops. On the other hand, if additional facilities are needed for higher troop levels and if Afghanistan becomes a more developed theater like Iraq, then the cost of support could grow.

In Iraq, it's not clear how quickly savings from withdrawals will match or be proportional to changes in troop strength. If about one-third of the average cost of a BCT did not decrease in the first year, then savings would not be proportional to changes in troop strength. Those savings, however, would be likely to occur the following year. So each year, savings from the previous year could make up, at least in part, for lags in savings in the current year. It is also possible that costs may become proportional, or even exceed changes in troop strength, as the pace of redeployments from Iraq picks up.

Although there are some additional costs associated with the withdrawal itself, such as repair and replacement of equipment that has remained in-theater rather than rotating with units, these costs do not appear to be as large as some have suggested and could be spread over two to three years (see discussion of withdrawal costs above).

Appendix C. CBO's Projections of War Costs, FY2009-FY2018

The Congressional Budget Office (CBO) has published widely used ten-year projections of war costs for the past several years. CBO projects future costs using two different scenarios assuming more and less gradual decreases in total number of troops deployed for OEF and OIF that some observers have suggested encompass the likely range of alternatives. CBO does not distinguish between OEF and OIF.

Starting from a deployed troop strength of 210,000 (about 30,000 above Boots on the Ground report), CBO's scenarios assume:

• decrease to 30,000 by FY2011; or
• decrease to 75,000 by FY2013 (see Table C-1).¹⁰⁴

In both the more rapid and the gradual drawdown, CBO's projections assume that savings initially fall slower than troop levels, catching up or exceeding changes in troop levels by the second or third year (see Table C-1). CBO does not provide explanations or rationale for its projections.

Although CBO does not calculate a per person or per BCT-equivalent cost, it appears that a 9,000 troop BCT would initially save about $7 billion a year under the more rapid scenario, rising to $9.6 billion per BCT during a steady-state. In the more gradual drawdown, annual savings per BCT would grow from $7 billion to about $8.4 billion once a steady-state is reached.

Based on DOD's reported obligations, CRS estimates that withdrawing an average BCT in Iraq would save about $7.2 billion and one in Afghanistan about $8.2 billion. CBO's figures may be higher because State/USAID and VA Medical costs may be included as well as DOD costs.¹⁰⁵ While there is uncertainty about how costs change with troop increases and decreases, as well as whether the costs of an average BCT will remain the same, troop strength plays an important role in setting DOD's future costs (see Appendix A and Appendix B for further details).

Appendix D. Headcounts in Boots on the Ground and Operations Reports

The Defense Department makes a distinction between Boots on the Ground (BOG) or troops deployed in-country in Iraq and in Afghanistan and troops deployed for Operation Enduring Freedom (OEF) and Operation Iraqi Freedom (OIF) which also include troops deployed to neighboring regions or for other small counter-terror operations.

Although DOD has not adopted a single definition of the countries included in either OEF or OIF, the Joint Staff and Central Command have both developed guidelines or "business rules" that allocate troops in individual countries to one or the other mission. The services follow individual guidelines in reporting war costs for each operation.¹⁰⁶

In many cases, the Joint Staff and Central Command guidelines are the same. In other cases, the Joint Command guidelines cover countries outside of Central Command. In only three cases do the Joint Staff and Central Command allocations differ, affecting about 770 military personnel; in these cases, CRS selected the operation most closely associated with that country. In some cases, countries have assigned missions but no personnel are shown as deployed.

The Joint Staff guidelines were developed to meet a new congressional reporting requirement to submit monthly "the total number of troops deployed in support of OIF and OEF, . . . delineated by service and component (active, Reserve or National Guard)." Although the services largely agreed with the country designations, there were some differences, and some concerns were raised that linking personnel and operations would not capture cases where military personnel at one location worked on both OEF and OIF (e.g. Qatar) or where the focus shifted over time (e.g., Navy Afloat personnel).

According to these allocations, the OEF mission includes primarily troops deployed in Afghanistan and about 5,000 to 15,000 troops in other countries.

The OEF operation includes military personnel deployed in the following countries: Afghanistan, Djibouti, Egypt, Ethiopia, Jordan, Kazakhstan, Kenya, Kyrgyzstan, Lebanon, Oman, Pakistan, Philippines, Seychelles, Sudan, Tajikistan, Turkmenistan, Uzbekistan, Egypt, and Yemen.

For the OIF operation, while most troops are in Iraq, another 30,000 to 40,000 are deployed in Kuwait, about 15,000 other troops in Bahrain, Qatar, UAE, plus an additional 15,000 to 20,000 on ships afloat in the region.

The OIF mission includes: Iraq, At Sea, Bahrain, Kuwait, Oil platforms, Qatar, Saudi Arabia, Spain, Turkey, and UAE, Israel, Jordan.¹⁰⁷

Table D-1 and Table D-2 show monthly troop levels for the Joint Staff's Boots on the Ground in Afghanistan and Iraq and for OEF and OIF as reported in Central Command's Operations Reports to compare the number of military personnel located in-country and a total that also includes those providing theater-wide support, or conducting other counter-terror operations related to OEF. This data is also shown in Figure 6 and Figure 8.