On March 24, 1999, President Bill Clinton publicly ruled out committing ground troops to address the military crisis in the Yugoslav province of Kosovo. The failure of the North Atlantic Treaty Organization's (NATO) air campaign to compel Yugoslavian President Slobodan Milosevic to withdraw his forces from Kosovo has prompted some Members of Congress to urge an intervention by NATO that utilizes ground troops. Their calls may be having an effect: the Clinton Administration seems to be softening its stance on the use of ground troops. On April 11, Secretary of Defense William Cohen admitted that contingency plans for ground operations could be "updated fairly quickly."
Before the United States commits ground troops to Kosovo, or elsewhere in Yugoslavia, it is vital that Congress and the American people have some appreciation for the likely cost, risk, and feasibility of such operations. This assessment will look at several possible scenarios and address three questions regarding each: First, how long would it take for NATO ground forces to deploy and execute their assigned mission? Second, how many ground troops would be necessary to execute each mission? And third, what would be the likely costs in terms of casualties and dollars?
In a worst-case scenario, NATO forces could be assigned the expansive mission of completely destroying Yugoslavia's military forces and occupying the entire country. The probable costs and risks of this operation would place tremendous strains on the NATO alliance.
NATO would face numerous logistical hurdles to support a ground offensive to destroy Yugoslavia's military forces and occupy the entire country. Serbia, the largest republic in Yugoslavia, is a landlocked territory slightly larger than the state of Maine. Its terrain ranges from gently undulating plains in the north to rugged mountain ranges in Kosovo, the federation's southernmost province. The terrain in the south would favor the defense.
Yugoslavia has a large military force for a country its size (see Table 2). Its military is technologically inferior to NATO forces; most of its weapon systems are more than 20 years old. Nevertheless, Yugoslavian troops are generally well-trained and indoctrinated for the purposes of defending against invading forces. The military's extensive air defense network, even after nearly a month of bombing, remains a danger to NATO pilots.
The NATO air campaign has reduced Yugoslavia's ability to maneuver conventional ground units, except on a piecemeal basis. The air campaign has hit its fuel reserves particularly hard, reducing them by an estimated 50 percent.2 President Milosevic's military commanders have dispersed tanks, fighting vehicles, and artillery pieces to reduce their vulnerability to air attack. Yugoslavia's extensive indigenous arms industry has allowed its forces to stockpile large stores of weapons and ammunition, thus blunting the impact of the United Nations arms embargo imposed in 1991. If Yugoslavia were invaded, NATO forces also would have to contend with the high likelihood of resistance by guerrillas and partisan forces. The need to pacify this resistance would require a substantial number of ground forces after the initial invasion.
Taking all these factors into account, a NATO ground offensive to destroy all of Yugoslavia's military forces and occupy the entire country would require up to 500,000 troops including at least 250,000 Americans.3
During Operation Desert Shield, the buildup of coalition forces prior to the 1991 Persian Gulf War, the U.S.-led forces had access to excellent airfields and port facilities in Saudi Arabia and other Arab countries. In contrast, the geography in the Balkans poses severe challenges for military planners; most of this region's ports, airfields, and roads are poorly suited for assembling heavy combat divisions. It probably would take NATO six to eight months just to mobilize a force capable of destroying Yugoslavia's military force and occupying its territory. NATO forces could assemble in various staging areas, to include Albania, Bosnia-Herzegovina, Hungary, and possibly even Croatia.4
It would take even longer to execute this military mission, considering the probable resistance by not only conventional forces but also partisan attacks supported by a hostile population. In effect, in seeking to destroy all of Yugoslavia's military forces and occupy the entire country, NATO would be committing its forces to an open-ended mission.
This operation would cost at least $40 billion to $50 billion the first year, possibly more, depending on the level of resistance by Yugoslavia's military and partisan forces.5 The United States would have to mobilize thousands of reservists to augment its active-duty forces. Considering the likelihood of NATO forces' being attack by unconventional and partisan guerrillas, NATO's occupation costs could reach as high as $10 billion annually. The United States could expect as many as 15,000 to 20,000 casualties the first year alone.
It is possible that NATO forces could be tasked with seizing and occupying Belgrade, the capital of both Yugoslavia and the Republic of Serbia. The purpose of this mission would be to remove President Milosevic from power and destroy his government, but to avoid completely destroying his forces or occupying all of Serbia. Although seizing and occupying Belgrade would be less difficult than occupying the entire country, this operation still would pose formidable challenges for NATO forces.
Resistance by Yugoslavian military forces probably would stiffen as NATO forces neared Belgrade. Urban combat is costly and manpower-intensive, thus raising the costs and risks to NATO forces. A large number of NATO troops also would be required to prevent guerrilla forces from cutting supplies lines and hitting soft targets behind NATO front lines, such as staging areas and supply depots.
Taking all these factors into account, a ground offensive to seize and occupy Belgrade would be likely to require between 150,000 and 200,000 NATO troops, to include at least 75,000 to 100,000 Americans. Although a smaller-sized force could be capable of executing this mission, it would risk a much greater number of casualties.
Depending on where NATO decided to stage its intervention forces, it would take between three and six months just to deploy the ground power capable of seizing Belgrade. NATO forces then would require anywhere from one to two months to defeat Yugoslavia's conventional ground forces en route to seizing Belgrade. After defeating the enemy's conventional forces, NATO then could find itself in an open-ending campaign to subdue guerrilla forces.
The cost of such a ground operation could range between $10 billion and $20 billion. President Clinton would have to mobilize thousands of reservists to supplement the active-duty forces. The number of casualties would depend largely on how fiercely the Yugoslavian forces resisted. Unlike the Persian Gulf War, in which coalition forces evicted Iraqi soldiers from Kuwait, the U.S.-led forces would be spearheading an operation into the enemy's capital. This suggests NATO forces would encounter stronger resistance and thus higher casualties.
A major ground offensive to take Belgrade could result in anywhere between 5,000 and 10,000 U.S. casualties. The longer-term costs are more uncertain: Much would depend on the willingness of paramilitary forces and defeated conventional forces to pursue guerrilla operations. In general terms, NATO could expect more casualties the longer it occupied Belgrade because the populace would view NATO as conquerors, not liberators.
NATO planners would face significant problems if tasked with fielding a ground operation to oust Yugoslavian forces from Kosovo, even though these problems would not be as formidable as an effort to occupy the entire country or seize Belgrade.
NATO's air operation has reduced President Milosevic's ability to reinforce his military forces in Kosovo with conventional units garrisoned in Serbia. The air campaign has hit the military's fuel reserves particularly hard, reducing them by an estimated 50 percent. NATO would intensify its efforts to attack Yugoslavia's ground forces in Kosovo prior to any ground offensive.
Despite these advantages, NATO ground troops would face several difficulties in prosecuting a ground campaign in Kosovo. The mountainous terrain in southern Kosovo, a province roughly the size of Connecticut, would favor the defenders. Anticipating a possible NATO ground offensive, Yugoslavian forces have mined the border with Albania, dug defensive trenches, and built bunkers (see Table 3).
The burgeoning refugee crisis could complicate the planning for any ground offensive significantly. More than half a million ethnic Albanians already have fled Kosovo, and an estimated 800,000 Kosovar Albanians remain dislocated within the province. The resulting humanitarian crisis already has prompted NATO to deploy 8,000 troops to Kukes, a town in Albania 20 miles from the Kosovo border. President Milosevic could delay and complicate NATO's plans for a ground offensive by continuing his policy of forced migration of refugees along Kosovo's limited road network.
Taking these factors into account, NATO probably would require between 50,000 and 70,000 troops, including 20,000 to 30,000 Americans or more, to defeat Yugoslavia's conventional forces in Kosovo. Faced with overwhelming NATO firepower, some Yugoslavian paramilitary units probably would resort to guerrilla warfare (see Table 4). Yet guerrillas would find it difficult to sustain their operations without the support of a sympathetic populace.
The geography around Kosovo does not lend itself to the speedy deployment of NATO forces. The difficulties NATO has experienced in dispatching Apache helicopter gunships to Albania provide a glimpse of the logistical hurdles a much larger ground force would encounter. Albania, the most likely staging area for NATO ground troops, is the poorest country in Europe; its roads, airfields, and ports are ill-suited to accommodate large numbers of combat troops.
Moreover, NATO would face political problems in deploying combat troops in anticipation of a ground offensive. Albania's parliament is giving NATO unrestricted access to its airspace and military bases, but Macedonia appears reluctant to allow its territory to be used as a staging area. At this writing, its appears unlikely that Greece would allow NATO forces to use its port facilities because of domestic political opposition to Operation Allied Force.
Depending on where it assembled its forces, NATO would need anywhere from one to three months to deploy a force capable of ousting Yugoslavia's military and paramilitary forces from Kosovo. There are only 14 roads leading into this province, a complication for military planning. A smaller ground force could be put into action sooner, but that would entail greater risk to NATO troops and it would be less likely to achieve its assigned mission.
Kosovo's rugged terrain militates against a lightening-quick ground offensive. NATO ground forces probably would require four to six weeks to seize control of the province. This assumes that the ground offensive would be preceded by an intensified air campaign designed to pummel Yugoslavian forces in Kosovo. It would take NATO forces much longer to flush out and destroy scattered paramilitary and guerrilla forces within the province.
The air operation preceding a NATO ground assault could cost the United States $2 billion to $3 billion.6 The NATO ground operation in Kosovo would be far more expensive; it probably would cost the United States anywhere between $5 billion and $10 billion, in addition to the $6 billion President Clinton projects the air campaign and humanitarian operations to cost through September 1999.7
Instead of committing ground troops to evict President Milosevic's forces from Kosovo, NATO could seize a sliver of the province and declare it a liberated zone. Refugees could be protected in this sanctuary. Such action in effect would partition the province and leave NATO forces' controlling only part of Kosovo. Unlikely to satisfy either Kosovar Albanians or nationalists in Belgrade, this outcome would sow the seeds of future conflict.
It would take at least 15,000 NATO troops, including 10,000 Americans, to establish and secure a liberated zone along the Albanian and Macedonian borders that reached approximately 20 miles deep into Kosovo. If, at some future point, President Milosevic decided to inundate the liberated zone with yet another deluge of Kosovar Albanian refugees, still more NATO troops would become necessary to cope with the humanitarian crisis.
The terrain along Kosovo's southern border is mountainous and features only a limited number of roads. The rugged terrain would favor the defenders. It would take at least a month to build up the combat forces necessary to carve out a liberated zone in southern Kosovo. The ground offensive to secure the territory probably would take two to four weeks. This assumes that NATO warplanes would destroy most of Yugoslavia's conventional forces in the area before the ground offensive began.
The dollar cost of this scenario would be much less than either an attempt to seize Belgrade or oust Yugoslavian forces from Kosovo. Even so, this operation would not be cheap. It probably would cost at least $5 billion to establish a liberated zone.
Considering the limited nature of this operation, the initial risk to NATO ground troops would be less than that posed by the more expansive ground scenarios discussed earlier. Nevertheless, this operation could result in as many as 500 U.S. casualties.
Much of the Kosovar resistance movement is operating out of Albania. A serious effort to assist resistance fighters would require the deployment of U.S. special operations forces, which could provide help to arm, train, and equip the ethnic Albanian resistance movement for operations in Kosovo. This approach would be in keeping with the tradition of the Reagan Doctrine, whereby Americans helped freedom fighters to defend themselves.
The rugged terrain along the Albania-Kosovo border dictates the deployment of a sizable ground force for security purposes.8 Yugoslavian forces have shelled Kosovar Albanian resistance forces in Albania, including the village of Tropoje. They also have demonstrated their ability to mount small-scale cross-border operations.9
A concerted effort to assist the growing number of displaced Kosovar resistance fighters located in Albania would require the deployment of at least 10,000 NATO troops, half of them American, beyond those already deployed in Albania and Macedonia for humanitarian purposes. The U.S. troops would be necessary for training and logistical support. NATO forces would not be intended for incursions into Kosovo, but rather to protect U.S. special operations forces as they train and organize the Kosovar resistance movement.
The United States could begin moving special operations forces into Albania for the purposes of arming, training, and equipping the Kosovar Albanian resistance within days. Yet it would take at least one month to deploy a security force capable of deterring further cross-border attacks by Yugoslavia's military forces. A long-term effort would be necessary to build up the Kosovar Albanian resistance fighters into a force capable of bleeding dry President Milosevic's military and paramilitary forces in Kosovo.
This operation would be the least costly of any of the ground scenarios. A U.S.-backed effort to assist the Kosovar Albanian resistance could cost $1 billion to $2 billion a year. This operation also would pose the least risk to U.S. ground troops. NATO's security force and air capabilities could limit the danger of a large-scale cross-border attack by Yugoslavian forces. But the capture of three U.S. soldiers near the Macedonia-Kosovo border by Yugoslavian forces on March 31 suggests this type of operation would not be free of danger.
It is possible, although unlikely, that an intensified air campaign alone may yet compel President Milosevic to withdraw his forces from Kosovo and accept an international peacekeeping force, as called for by NATO's political leaders. Although this outcome would obviate the need to deploy U.S. troops as combatants, it would be likely to result in the presence of a sizable number of ground troops as peacekeepers.
The defunct Rambouillet peace proposal, as discussed in February 1999, called for a NATO-led peacekeeping force of 28,000 troops, to include 4,000 Americans. The manpower requirements of a peacekeeping force would be considerably higher after a sustained bombing campaign; the force probably would require 30,000 to 40,000 troops. NATO would ask for a commitment from the United States to increase the size of its contribution to 8,000-10,000 troops instead of the 4,000 in the Rambouillet proposal.
The peacekeeping force probably would be assigned the mission of escorting tens of thousands of refugees safely back to their villages. Moreover, it is likely that NATO forces would be tasked with providing security for those civilian agencies working to reconstruct Kosovo, and with protecting the minority Serb civilian population from possible reprisal attacks.
It would take less time to insert peacekeepers into Kosovo than it would combatants because the former would not face the same security threats. Some peacekeepers could be dispatched within days of a negotiated settlement, although deploying the entire force would require four to six weeks or longer.
NATO's peacekeeping mission in Bosnia has lasted much longer than expected; the NATO forces deployed there are, in effect, locked in an open-ended commitment. NATO troops would be likely to remain in Kosovo for anywhere from 5 to 10 years.
The danger of casualties to NATO troops if they deployed as peacekeepers would be significantly less than any forcible attempt to occupy all of Yugoslavia, seize Belgrade, or oust Yugoslavian forces from Kosovo. But a peacekeeping mission would not be risk-free. The peacekeeping force would have to defend itself against the possibility of terrorist reprisals.
For security purposes, this peacekeeping force would have to include combat forces capable of deterring any aggression by Yugoslavia along the Kosovo-Serbia border. It probably would cost the United States $5 billion the first year and roughly $2 billion to $3 billion per year thereafter to deploy a robust peacekeeping force within Kosovo and a combat force along Kosovo's border with Serbia.10
Estimating likely risks and costs of ground operations is an inexact science. These calculations depend on friendly and enemy military capabilities and necessarily more subjective assessments of intangible forces, such as unit cohesion and willingness to fight. In all the scenarios examined in this Backgrounder, the deployment of NATO ground troops would involve a combination of foreseen and unforeseen costs. One or more of these wildcard developments could raise significantly the costs and risks associated with the deployment of NATO ground forces into Yugoslavia or its neighboring countries. These include:
A decision by Russia to provide Yugoslavian forces with direct or indirect military assistance.
Such assistance could take various forms, to include intelligence-sharing or the shipment of arms and materiel to Yugoslavian forces.
A crisis elsewhere in the world that requires a military response would make it difficult for the United States to mount or sustain operations in the Balkans.
To support Operation Allied Force, the United States has redeployed air assets enforcing the no-fly zone in northern Iraq and left its naval forces in the Pacific Ocean without a carrier presence.
A decision by one or more NATO partners to limit its participation in or contributions to a major ground offensive.
Italy and Greece already have indicated their discomfort with the existing air operation. Mounting a ground operation would place additional stress on the NATO alliance.
An extended period of foul weather could complicate NATO's ability to conduct ground operations.
The cloud cover already has limited the effectiveness of Operation Allied Force.
A decision by President Milosevic to attack NATO forces with chemical weapons.11
Any major effort by Yugoslavia along these lines would slow the pace of NATO's ground operations and raise the number of expected casualties.
A decision by President Milosevic to open up yet another front, perhaps against the Republic of Montenegro.
A NATO ground war against Yugoslavia would greatly tax the military capabilities of the United States. The rugged terrain and lack of easy entry points for heavy combat forces would make it very difficult to achieve any quick deployment of NATO troops for a major ground offensive. Clearly, NATO forces would need time to build up staging areas capable of projecting tens of thousands of ground troops.
A ground offensive to destroy all of Yugoslavia's military and paramilitary forces and occupy the entire country certainly would be the most costly, risky, and time-consuming option. A NATO effort to arm, train, and equip the displaced Kosovar resistance fighters in Albania would be the least costly operation.
As the political leader of the NATO alliance and its most capable military member, the United States probably would be asked to bear the onus of any ground operation. That certainly would be the intention of many Europeans and some Americans. Any major ground commitment of U.S. troops directed against Yugoslavia would stress the ability of the United States to meet its security commitments elsewhere in the world. Members of Congress would do well to consider these costs, risks, and uncertainties as it debates the merits of placing U.S. ground troops in harm's way.
James H. Anderson, Ph.D., is a former Defense Policy Analyst in The Kathryn and Shelby Cullom Davis International Studies Center at The Heritage Foundation.
1. The author thanks Jack Spencer, Research Assistant for The Kathryn and Shelby Cullom Davis International Studies Center at The Heritage Foundation, for his assistance with preparing the order-of-battle tables.
3. For an assessment of Nazi Germany's experience in Yugoslavia during World War II, see Mark F. Cancian, "The Wehrmacht in Yugoslavia: Lessons of the Past?" Parameters XXIII, No. 3 (Autumn 1993), pp. 75-84.
6. The Center for Strategic and Budgetary Assessments estimates that the U.S. portion of the first 22 days of Operation Allied Force cost between $400 million and $600 million. See Steven Kosiak, "Cost of Allied Force Air Campaign: Day 22," April 15, 1999. This estimate is available at http://www.csbahome.com.