United States General Accounting Office GAO Testimony Before the Committee on Appropriations, U.S. Senate For Release on Delivery Expected at 10:00 a.m., EST, NATO ENLARGEMENT Thursday, October 23, 1997 Cost Implications for the United States Remain Unclear Statement of Henry L. Hinton, Jr., Assistant Comptroller General, National Security and International Affairs Division GAO/T-NSIAD-98-50 Mr. Chairman and Members of the Committee: We are pleased to be here today to help the Committee sort through some of the issues related to the cost and financial obligations of expanding the North Atlantic Treaty Organization (NATO). Our testimony today will address three issues: (1) current U.S. costs to support NATO’s common budgets and other funding that supports relations with Central and East European nations and promotes NATO enlargement; (2) NATO’s defense planning process, which will form the basis for more definitive cost estimates for an enlarged alliance; and (3) our evaluation of the recent Department of Defense (DOD) study of NATO expansion and a comparison of DOD’s study with studies of the Congressional Budget Office (CBO) and the Rand Corporation. The ultimate cost of NATO enlargement will be contingent on several Summary of factors that have not yet been determined. Specifically, NATO has yet to Observations formally define (1) its strategy for defending the expanded alliance, (2) force and facility requirements of the newly invited states, and (3) how costs of expanding the alliance will be financed. Also unknown is the long-term security threat environment in Europe. NATO’s process for determining the cost of enlargement is underway and expected to be completed by June 1998. In fiscal year 1997, the United States contributed about $470 million directly to NATO to support its three commonly funded budgets, the NATO Security Investment Program (NSIP), the military budget, and the civil budget. This is about 25 percent of the total funding for these budgets. It is through proposed increases to these budgets, primarily the NSIP and to a lesser extent the civil budget, that most of the direct cost of NATO enlargement will be reflected and therefore where the United States is likely to incur additional costs. Additionally, over $120 million was programmed in fiscal year 1997 for Warsaw Initiative activities in the three countries that are candidates for NATO membership and other Partnership for Peace (PFP) countries.1 This money was provided to help pay for Foreign Military Financing grants and loans, exercises, and other PFP-related activities. Funding for these 1 In 1994, NATO launched a wide-ranging cooperative effort—known as PFP—with nonmember countries to promote democracy, expand cooperation, and strengthen relationships between NATO and nonmember countries. Participation of countries in PFP plays a role in NATO’s decisions regarding expansion. For further information see NATO Enlargement: U.S. and International Efforts to Assist Potential New Member States (GAO/NSIAD-97-164, June 27, 1997). Page 1 GAO/T-NSIAD-98-50 activities will continue, but the allocation between the candidates for NATO membership and all other PFP participants may change over time. This funding is strictly bilateral assistance that may assist the candidate countries and other countries participating in PFP to meet certain NATO standards, but it is not directly related to NATO decisions concerning military requirements or enlargement. NATO defense planners are now developing military requirements through their defense planning process and are close to completing their analyses. These requirements will ultimately be translated into costs eligible for common funding. NATO officials plan to present their cost estimates for these items for approval at the NATO defense ministerial meeting in early December 1997. However, it will not be until June 1998 that NATO will make decisions about whether or how much to increase the common budgets, which would then be shared among current and new members. Until this has been done, the implications for the U.S. contributions to NATO’s common budgets will be unclear. As you know, DOD, CBO, and Rand developed cost estimates for enlarging NATO before invitations were extended to Poland, Hungary, and the Czech Republic and therefore before NATO had assessed its current military needs or developed military requirements that could be used to make more accurate cost estimates. Thus, the ranges of these estimates—from $10 billion to $125 billion—are substantially different, depending on the assumptions used. In some instances, the cost range estimates overlap; however, this may be coincidental, since the assumptions and force postures used to develop the estimates were different. Thus, it is not surprising that the debate on this issue has been surrounded by some confusion. Our analysis of DOD’s cost estimate to enlarge NATO indicates that its key assumptions were generally reasonable and were largely consistent with the views of U.S., NATO, and foreign government officials.2 In particular, the assumption that large-scale conventional security threats will remain low significantly influenced the estimate. However, DOD’s lack of supporting cost documentation and its decision to include cost elements that were not directly related to enlargement call into question its overall estimate. Because of the uncertainties associated with enlargement and DOD’s estimating procedures, the actual cost of NATO enlargement could be 2 See our report NATO Enlargement: Cost Estimates Developed to Date Are Notional (GAO/NSIAD-97-209, Aug. 18, 1997). Page 2 GAO/T-NSIAD-98-50 substantially different from DOD’s estimated cost of about $27 billion to $35 billion. Rand and CBO cost estimates are no more reliable than DOD’s, based on our comparison of the three studies. CBO and Rand developed a range of cost estimates for NATO enlargement, including estimates that employ a defense strategy similar to DOD’s. Several factors account for the differences between DOD’s estimate and the CBO and Rand estimates, including those estimates that employed defense strategies similar to DOD’s. For example, CBO assumed a much larger reinforcement force and much more extensive modernization, infrastructure, and training costs than DOD did. Rand assumed a somewhat larger reinforcement force and higher training and air defense modernization costs than DOD did. As it does now, the United States will fund its share of NATO enlargement U.S. Contributions to primarily through contributions to the three common budgets. NSIP pays Common Budgets and for infrastructure items that are over and above the needs of the member Other Funding nations, including communications links to NATO headquarters or reinforcement reception facilities, such as increased apron space at Sources existing airfields. The military budget pays for NATO Airborne Early Warning Force program and military headquarters costs, and the civil budget pays primarily for NATO’s international staff and operation and maintenance costs of its civilian facility in Brussels. For fiscal year 1997, the U.S. contribution for the three common budgets was about $470 million: $172 million for the NSIP, $252 million for NATO’s military budget, and $44.5 million for NATO’s civil budget. Any increases to the U.S. budget accounts would be reflected primarily through increased funding requests for the DOD military construction budget from which the NSIP is funded, the Army operations and maintenance budget from which the military budget is funded (both part of the National Defense 050 budget function), and the State Department’s contributions to international organizations from which the civil budget is funded (part of the International Affairs 150 budget function). While NATO will not have finalized its common infrastructure requirements for new members until December 1997 or decided whether or how much to increase the common budgets until June 1998, DOD and State Department officials told us that the civil and NSIP budgets are likely to increase by only 5 to 10 percent and the military budget will probably not increase at all. This would mean an increase of about $20 million annually for the U.S. contribution to NATO. However, as we indicated, NATO has yet Page 3 GAO/T-NSIAD-98-50 to make decisions on these matters. In addition, the United States could choose to help new members in their efforts to meet their NATO membership obligations through continued Foreign Military Financing grants and/or loans, International Military Education and Training grants, and assistance for training activities. The three candidate countries and other PFP countries have been receiving assistance through these accounts since the inception of the PFP program, and this has enabled some of these countries to be more prepared for NATO membership. In fiscal year 1997, over $120 million was programmed for these activities, and about $60 million of this amount went to the three candidates for NATO membership. Any increased funding for such assistance would be funded through the International Affairs and Defense budget functions. It is through NATO’s defense planning process that decisions are made on NATO’s Defense how the defense burden will be shared, what military requirements will be Planning Process satisfied, and what shortfalls will exist. NATO’s New Strategic Concept, adopted in Rome in 1991, places greater emphasis on crisis management and conflict prevention and outlines the characteristics of the force structure. Key features include (1) smaller, more mobile and flexible forces that can counter multifaceted risks, possibly outside the NATO area; (2) fewer troops stationed away from their home countries; (3) reduced readiness levels for many active units; (4) emphasis on building up forces in a crisis; (5) reduced reliance on nuclear weapons; and (6) immediate and rapid reaction forces, main defense forces (including multinational corps), and augmentation forces. Although NATO has not defined exactly the type and amount of equipment and training needed, it has encouraged nations to invest in transport, air refueling, and reconnaissance aircraft and improved command and control equipment, among other items. NATO’s force-planning and goal-setting process involves two interrelated phases that run concurrently: setting force goals and responding to a defense planning questionnaire. The force goals, which are developed every 2 years, define NATO’s requirements. The major NATO commanders propose force goals for each nation based on command requirements. Each nation typically has over 100 force goals. NATO and national officials frequently consult one another while developing force goals and national defense plans. NATO commanders are unlikely to demand that member nations establish units or acquire equipment they do not have. Page 4 GAO/T-NSIAD-98-50 In its annual response to NATO’s defense planning questionnaire, each member verifies its commitment for the previous year, defines its commitment for the next year, and lays out plans for the following 5 years. Alliance members review each nation’s questionnaire and, in meetings, can question national plans and urge member nations to alter their plans. After finishing their reviews, generally in October or November, NATO staff write a report summarizing each nation’s plans and assessing national commitments to NATO. Once NATO members approve this report, it becomes the alliance’s consensus view on each country’s strengths and weaknesses and plan to support the force structure. It is through this process that NATO determines what shortfalls exist, for example, in combat support and combat service support capabilities. According to U.S. officials, NATO is preparing several reports to be presented for approval at the defense ministerial meetings in December 1997. One report will discuss the additional military capability requirements existing alliance members will face as a result of the alliance’s enlargement. According to officials at the U.S. mission and Supreme Headquarters Allied Powers Europe, it is unlikely that any additional military capability requirements will be placed on NATO members over and above the force goals they have already agreed to provide. In other words, if current force goals are attained, NATO will have sufficient resources to respond to likely contingencies in current and new member countries. Therefore, it can be concluded that although enlargement of the alliance is another reason for current allies to attain their force goals, it will not add any new, unknown costs to existing members’ force plans. Other reports resulting from this process will discuss the requirements for commonly funded items in the new nations and their estimated costs. These items include infrastructure that will enable the new allies to receive NATO reinforcements in times of crisis, communication systems between NATO and their national headquarters, and a tie-in to NATO’s air defense system. How these projects will be financed by NATO, for example, whether they will be financed within existing budgets or by increasing the size of NATO’s common budgets, will not be determined until June 1998. Therefore, the impact of these costs on the U.S. contributions to NATO’s common budgets and the U.S. budget will be unknown until next spring. Another report will present an assessment of the capabilities and shortfalls in the military forces of Poland, Hungary, and the Czech Republic. NATO does not and will not estimate the costs of the shortfalls of either the Page 5 GAO/T-NSIAD-98-50 current or the new member states, but once these shortfalls are identified, cost estimates can be made by others. However, even though new members’ capabilities and shortfalls will be identified in December, these countries’ force goals will not be set until the spring. These force goals will, in effect, be a roadmap for the new members on how to address their shortfalls. (See app. I for a timeline illustrating these events.) When the DOD, CBO, and Rand studies were completed, many key cost Key Assumptions and determinants had not been established. Consequently, each study made a Cost Estimates for series of key assumptions that had important implications for each studies’ NATO Enlargement results. Studies DOD made the following key assumptions: • Specific nations would be invited to join NATO in the first round of enlargement.3 • NATO would continue to rely on its existing post-Cold War strategy to carry out its collective defense obligations (that is, each member state would have a basic self-defense capability and the ability to rapidly receive NATO reinforcements).4 • NATO would not be confronted by a significant conventional military threat for the foreseeable future, and such a threat would take many years to develop. • NATO would continue to use existing criteria for determining which items would be funded in common and which costs would be allocated among members.5 Using these assumptions, DOD estimated the cost of enlarging NATO would range from about $27 billion to $35 billion from 1997 to 2009. The estimate was broken down as follows: 3 DOD assumed that four countries would be invited to join NATO, but the actual countries that were the basis for the estimate are classified information. 4 NATO adopted a new post-Cold War strategic concept at its Rome summit meeting in 1991. The concept provides for substantial reductions in the size and readiness of NATO’s forces but increased force mobility, flexibility, and ability to adapt to the changed threat environment. 5 NATO funds only those facilities or portions of facilities that are over and above the needs of an individual country’s national security requirements. For example, NATO would fund only the portion of infrastructure at an air base that is beyond the host nation’s own needs, such as hangars for reinforcing aircraft, but not hangars for the host country’s aircraft. Page 6 GAO/T-NSIAD-98-50 • about $8 billion to $10 billion for improvements in current NATO members’ regional reinforcement capabilities, such as developing mobile logistics and other combat support capabilities; • about $10 billion to $13 billion for restructuring and modernizing new members’ militaries (for example, selectively upgrading self-defense capabilities); and • about $9 billion to $12 billion for costs directly attributable to NATO enlargement (for example, costs of ensuring that current and new members’ forces are interoperable and capable of combined NATO operations and of upgrading or constructing facilities to receive NATO reinforcements). DOD estimated the U.S. share of these costs would range from about $1.5 billion to $2 billion—averaging $150 million to $200 million annually from 2000 to 2009. The estimated U.S. share chiefly consisted of a portion of direct enlargement costs commonly funded through NATO’s Security Investment Program. DOD assumed that the other costs would be borne by the new members and other current member states and concluded that they could afford these costs, although this would be challenging for new members. (See app. II.) DOD’s Key Assumptions In our review of DOD’s study of NATO enlargement, we (1) assessed the Were Reasonable, but Cost reasonableness of DOD’s key assumptions, (2) attempted to verify pricing Estimates Are Speculative information used as the basis for estimating enlargement costs, (3) looked into whether certain cost categories were actually linked to enlargement, and (4) identified factors excluded from the study that could affect enlargement costs. We concluded that DOD’s assumptions were reasonable. The assumption regarding the threat was probably the most significant variable in estimating the cost of enlargement. Based on information available to us, we concluded that it was reasonable to assume the threat would be low and there would be a fairly long warning time if a serious threat developed. This assumption, and the assumption that the post-Cold War strategic concept would be employed, provided the basis for DOD’s judgments concerning required regional reinforcement capabilities, new members’ force modernization, and to a large extent those items categorized as direct enlargement costs. DOD also assumed that during 1997-2009, new members would increase their real defense spending at an average annual rate of 1 to 2 percent. Page 7 GAO/T-NSIAD-98-50 Both private and government analysts project gross domestic product (GDP) growth rates averaging 4 to 5 percent annually for the Czech Republic, Hungary, and Poland during 1997-2001. Thus, projected increases in defense budgets appear affordable. Analysts also point out that potential new member countries face real fiscal constraints, especially in the short term. An increase in defense budgets at the expense of pressing social concerns becomes a matter of setting national priorities, which are difficult to predict. If these countries’ growth rates do not meet expectations, their ability to increase real defense spending becomes more problematic. DOD further assumed that current NATO members would on average maintain constant real defense spending levels during 1997-2009.6 Analysts have expressed somewhat greater concern about this assumption and generally consider it to be an optimistic, but reasonable projection. Some analysts indicated that defense spending in some current member states may decline further over the next several years. Such declines would partly be due to economic requirements associated with entry into the European Monetary Union.7 Despite our conclusion that DOD’s underlying assumptions were sound, for several reasons we concluded that its estimates are quite speculative. First, DOD’s pricing of many individual cost elements were “best guesses” and lacked supporting documentation. This was the case for all three categories of costs: direct enlargement costs, current members’ reinforcement enhancements, and new members’ modernization requirements. Most of the infrastructure upgrade and refurbishment cost estimates were based on judgments. For example, DOD’s estimate of $140 million to $240 million for upgrading a new member’s existing air base into a NATO collocated operating base was not based on surveys of actual facilities but on expert judgment. We were told that the actual cost could easily be double—or half—the estimate. DOD’s estimated costs for training and modernization were notional, and actual costs may vary substantially. DOD analysts did not project training 6 In 1996, defense spending as a percent of GDP was 2 percent for Italy, 1.7 percent for Germany, 2.9 percent for the United Kingdom, and 3 percent for France. If GDP increases in real terms, these percentages will decline under DOD’s assumption of constant real defense spending. 7 Under the European Monetary Union, scheduled to go into effect January 1, 1999, the European Union would have a common central bank and monetary policy and a single currency called the euro. According to the Maastricht Treaty, the primary goal of the common central bank is price stability. The treaty requires that the economies of the participating countries converge toward certain performance goals in terms of inflation, long-term interest rates, exchange rate stability, and budget deficits (no greater than 3 percent of GDP) and debt (no greater than 60 percent of GDP). Page 8 GAO/T-NSIAD-98-50 tempos and specific exercise costs. Instead, they extrapolated U.S. and NATO training and exercise costs and evaluated the results from the point of view of affordability. DOD’s estimate for modernization and restructuring of new members’ ground forces was also notional and was based on improving 25 percent of the new members’ forces. However, it did not specify what upgrades would be done and how much they would cost. Second, we could find no linkage between DOD’s estimated cost of $8 billion to $10 billion for remedying current shortfalls in NATO’s reinforcement capabilities and enlargement of the alliance. Neither DOD nor NATO could point to any specific reinforcement shortfalls that would result from enlargement that do not already exist. However, existing shortfalls could impair the implementation of NATO’s new strategic concept. DOD officials told us that while reinforcement needs would not be greater in an enlarged NATO, enlargement makes eliminating the shortfalls essential. This issue is important in the context of burdensharing because DOD’s estimate shows that these costs would be covered by our current NATO allies but not shared by the United States. Finally, NATO has yet to determine what military capabilities, modernization, and restructuring will be sought from new members. Consequently, DOD had little solid basis for its $10 billion to $13 billion estimate for this cost category. Moreover, DOD and new member governments have noted that new members are likely to incur costs to restructure and modernize their forces whether or not they join NATO. Indeed, some countries have indicated that they may need to spend more for these purposes if they do not become NATO members. DOD showed these costs as being covered entirely by the new members. Potential Additional Costs NATO enlargement could entail costs in addition to those included in DOD’s of Enlargement estimates, including costs for assistance to enhance the PFP or other bilateral assistance for countries not invited to join NATO in July 1997. In addition, the United States may provide assistance to help new members restructure and modernize their forces. For example, Polish officials said they may need up to $2 billion in credits to buy multipurpose aircraft. While not an added cost of enlargement, such assistance would represent a shift in the cost burden from the new member countries to the countries providing assistance. DOD did not include such costs in its estimate of the U.S. share, though it acknowledged that the cost was possible. Moreover, U.S. and NATO officials have stated that additional countries may be invited to join NATO in the future, most likely in 1999. DOD’s cost estimate did not Page 9 GAO/T-NSIAD-98-50 take into account a second or third round of invitations. If additional countries are invited, cost of enlargement would obviously increase. Comparison of the DOD, CBO and Rand estimated the cost of incorporating the Czech Republic, CBO, and Rand Estimates Hungary, Poland, and Slovakia into NATO. They based their estimates on a range of NATO defense postures, from enhanced self-defense with minimal NATO interoperability to the forward stationing of NATO troops in new member states. However, they also noted that the current lack of a major threat in Europe could allow NATO to spend as little as it chose in enlarging the alliance. Because of the uncertainties of future threats, and the many possible ways to defend an enlarged NATO, CBO examined five illustrative options to provide such a defense. Each option built on the pervious one in scope and cost. CBO estimated that the cost of the five options over the 15-year period would range from $61 billion to $125 billion. Of that total, CBO estimated that the United States might be expected to pay between $5 billion and $19 billion. CBO included in its range of options a $109-billion estimate that was predicated on a resurgent Russian threat, although it was based on a self-defense and reinforcement strategy similar to that used by DOD.8 Of this $109 billion, CBO estimated that the United States would pay $13 billion. Similarly, Rand developed estimates for four options to defend an enlarged NATO that build upon one another, from only self-defense support at a cost of $10 billion to $20 billion to the forward deployment of forces in new member states at a cost of $55 billion to $110 billion. These options include a middle option that would cost about $42 billion that was also based on a self-defense and reinforcement strategy. Rand estimated that the United States would pay $5 billion to $6 billion of this $42 billion in total costs. Several factors account for the differences between DOD’s estimates and the CBO and Rand estimates, even those that employed defense strategies similar to DOD’s. (App. III illustrates the major results and key assumptions of the three estimates.) CBO’scost estimate is significantly higher than DOD’s for the following reasons: 8 CBO’s lowest estimate is based on a low-threat assessment; the additional costs are predicated on a resurgent Russian threat. Page 10 GAO/T-NSIAD-98-50 • DOD assumed reinforcements of 4 divisions and 6 wings, whereas CBO assumed a force of 11-2/3 divisions and 11-1/2 wings and a much larger infrastructure for this force in the new member states. • CBO’s modernization costs are much higher than DOD’s and include the purchase of 350 new aircraft and 1,150 new tanks for the new member states. DOD assumed that about 25 percent of the new member states’ ground forces would be modernized through upgrades and that each nation would procure a single squadron of refurbished Western combat aircraft. • CBO assumed much higher training costs, $23 billion, which include annual, large-scale combined exercises. DOD included $2 billion to $4 billion for training. • CBO included the purchase of Patriot air defense missiles at a cost of $8.7 billion, which is considerably higher than DOD’s assumed purchase of refurbished I-HAWK type missiles at $1.9 billion to $2.6 billion. • CBO’s infrastructure costs were much higher than DOD’s and included new construction, such as extending the NATO fuel pipeline, which CBO assumed would meet U.S. standards. DOD assumed planned refurbishment of existing facilities that would meet minimal wartime standards. Rand’s cost estimate is somewhat higher than DOD’s, although both were based on similar threat assessments. First, its reinforcement package was larger—5 divisions and 10 wings—and therefore infrastructure costs were higher. Second, it assumed new members would purchase the more expensive Patriot air defense system rather than the refurbished I-HAWKs. Finally, it assumed greater training costs than did DOD. The author of the Rand study stated that if he had used DOD’s assumptions, the cost range would have been almost identical to DOD’s. Mr. Chairman, this concludes our prepared remarks. We would be happy to answer any questions you or the Committee members may have. Page 11 GAO/T-NSIAD-98-50 Appendix I NATO Enlargement Timeline Date Activity September 1995 NATO issues study on enlargement. July 1997 NATO issues invitations to Poland, Hungary, and the Czech Republic to begin accessions talks. October/November 1997 NATO prepares several reports: • additional military capability requirements for existing alliance members that will result from the alliance’s enlargement; • requirements for commonly funded items in the new member nations, including • infrastructure that will enable the new allies to receive NATO reinforcements in times of crisis, • communication systems between NATO and their national headquarters, and • a tie-in to NATO’s air defense system; • cost estimates for items eligible for common funding presented by NATO officials; and • the capabilities and shortfalls in the military forces of Poland, Hungary, and the Czech Republic. Early December 1997 NATO defense ministerial meeting to approve the above reports. Spring 1998 New members’ force goals set. June 1998 NATO decides whether or how much to increase the common budgets, which would then be shared among current and new members. April 1999 Target date for new member accession into NATO. Page 12 GAO/T-NSIAD-98-50 Appendix II Categories and Share of Costs Table II.1: Estimates From the Executive Branch Report to the Congress, February 1997 Dollars in billions New members’ Current allies’ Cost category share share U.S. share Total New members’ military restructuring and modernization $10 to $13 0 0 $10 to $13 Current members’ reinforcement enhancements 0 $8 to $10 0 8 to 10 Direct enlargement 3 to 4.5 4.5 to 5.5 $1.5 to $2 9 to 12 Total $13 to $17.5 $12.5 to $15.5 $1.5 to $2 $27 to $35 Page 13 GAO/T-NSIAD-98-50 Appendix III Comparison of DOD, CBO, and Rand Estimates Dollars in billions Assumption DOD CBO Rand Total cost $27-$35 in $61-$125 in $10-$110 in constant 1997 constant 1997 constant 1996 dollars dollars ($109 for a dollars ($42 for a defense strategy defense strategy similar to DOD’s) similar to DOD’s) U.S. cost share $1.5-$2.0 $13.1a $5-$6a Notional new NATO A small group Poland Poland members (details classified) Hungary Hungary Czech Republic Czech Republic Slovakia Slovakia Time period 1997-2009 1996-2010 Approximately 1995-2010 Threat assessment Low threat A resurgent Russiaa Low threata Comparable force 4 divisions/6 wings 11.7 divisions/11.5 5 divisions/10 wingsa posture options wingsa a These assumptions correspond to the estimate based on a defense strategy similar to DOD’s. (711305) Page 14 GAO/T-NSIAD-98-50 Ordering Information The first copy of each GAO report and testimony is free. Additional copies are $2 each. Orders should be sent to the following address, accompanied by a check or money order made out to the Superintendent of Documents, when necessary. VISA and MasterCard credit cards are accepted, also. Orders for 100 or more copies to be mailed to a single address are discounted 25 percent. Orders by mail: U.S. General Accounting Office P.O. Box 37050 Washington, DC 20013 or visit: Room 1100 700 4th St. NW (corner of 4th and G Sts. NW) U.S. General Accounting Office Washington, DC Orders may also be placed by calling (202) 512-6000 or by using fax number (202) 512-6061, or TDD (202) 512-2537. Each day, GAO issues a list of newly available reports and testimony. To receive facsimile copies of the daily list or any list from the past 30 days, please call (202) 512-6000 using a touchtone phone. A recorded menu will provide information on how to obtain these lists. For information on how to access GAO reports on the INTERNET, send an e-mail message with "info" in the body to: firstname.lastname@example.org or visit GAO’s World Wide Web Home Page at: http://www.gao.gov PRINTED ON RECYCLED PAPER United States Bulk Rate General Accounting Office Postage & Fees Paid Washington, D.C. 20548-0001 GAO Permit No. G100 Official Business Penalty for Private Use $300 Address Correction Requested
NATO Enlargement: Cost Implications for the United States Remain Unclear
Published by the Government Accountability Office on 1997-10-23.
Below is a raw (and likely hideous) rendition of the original report. (PDF)