Project on Defense Alternatives

Post-Cold War US Military Expenditure in the Context of World Spending Trends

Project on Defense Alternatives Briefing Memo #10
Carl Conetta and Charles Knight
January 1997

[Full Text] [Summary]

This Publication Available in Print
$4.00 buy

1. The Value and the Limits of International Spending Comparisons

International comparisons of military expenditure can help answer the question, How much national military capability and spending is enough? The determination of military requirements should rest on (i) a calculation of national interests and goals, (ii) an assessment of the potential military challenges to those interests and goals, and (iii) a choice among the various means -- military and nonmilitary, cooperative and independent -- for meeting the challenges. Of course, not all challenges to a nation are military (or even adversarial) in character. However, a survey of the military expenditures of likely or potential adversaries can indicate the magnitude of the specifically military challenges that do exist. A survey of world military spending also helps planners gauge the general level of military capability existent in the world. This can usefully reduce uncertainty about the future, illuminating the potential for the emergence of new or unexpected military threats, and clarifying how much nonspecific "military insurance" a nation may require. 1

For a variety of reasons static comparisons of military expenditure can provide only an inexact andpartial measure of relative military capability. Such comparisons are even less useful as predictorsof conflict outcomes. 2 Attention to specific contingencies and to the relative goals of the prospective contestants is always necessary. Even then, allowance must be made for a wide margin of error. Observation of multi-year spending trends, however, can enrich comparative analyses considerably. Any substantial and prolonged change in a "balance of expenditures" is suggestive of a change in relative strength -- even if the significance of the balance at any specific point in time is unclear.

Extrapolating from comparative spending data to an estimation of relative strategic position requires close attention to complicating factors. First, nations differ markedly in their ability to convert treasure into military capability. It is easy to spend money, but difficult to spend it wisely. Hence, defense expenditure must be qualified by a measure of cost effectiveness. 3 Second, countries differ in the technology and skill base upon which their military establishments draw, and they differ in the social and cultural context they provide for military development. Such "foundational" or "contextual" factors help determine how much capability any particular level of defense expenditure buys. 4

Just as defense expenditure does not equate directly with military capability, military capability does not equate directly with military strategic position. Because nations differ in their strategic and geostrategic circumstances, they also differ in their military requirements. The level of military capability that is adequate for one nation may be inadequate for the next -- even though their national interests and goals may be roughly similar. Some nations are more vulnerable to aggression than others and, thus, may have to spend more to achieve comparable levels of security. Some nations face special domestic constraints on their exercise of military power -- perhaps due to their form of government; others are less constrained in their resort to force. Some nations stand alone; others are parts of powerful alliance networks (which both impose costs and provide benefits). Some nations can amplify their military capabilities with non-military forms of power -- for instance, financial strength or diplomatic sway. Other nations are uniquely dependent on raw military power.

By taking into account the "complicating" factors reviewed above analysts can derive from comparative spending data a more accurate picture of relative strategic position. Some of the qualifying factors are fairly transparent -- for instance: a nation's special vulnerabilities, socio-economic conditions, and membership in alliances. Others -- like the cost-effectiveness of a nation's defense expenditures -- are more difficult to ascertain. Recent experience in war can help illuminate the cost-effectiveness of a nation's military expenditure. Also, cost-effectiveness should correlate with some obvious attributes of a nation's defense establishment and political process. Among those attributes conducive to cost-effective expenditure are (I) a high-level of military professionalism, (ii) open public debate on military policy issues, (iii) close scrutiny of military budgets by parliamentary authorities, (iv) multiple independent authorities for the testing and evaluation of procurement and other policy choices, and (v) a competitive defense industry.

2. US Military Spending in the Context of World Spending Trends

The tables in the following sections examine the change in US military spending between 1986 and 1994 against the background of changes in military expenditures worldwide. We chose 1986 as the baseline because it was the recent peak year for US military spending as measured by the US Arms Control and Disarmament Agency (ACDA) whose annual publication, World Military Expenditures and Arms Transfers, is the primary source for the spending data in this report. The most recent edition of World Military Expenditures (henceforth, WMEAT) provides data only through 1994. (All spending figures are in 1994 dollars). For a portion of countries outside the Western group (and especially those considered a threat to US interests) ACDA has made efforts to adjust its reports of expenditures to reflect real purchasing power rather than simple currency exchange rates. 5 Where ACDA data was not available for a country in a particular year we used an estimate based on data from a second source, The Military Balance (henceforth, MB). 6 We also used MB generally to cross-check ACDA's estimates; any significant discrepancies are addressed in footnotes.

Table 1 presents military spending data for the world, the United States, and selected groups of states for the years 1986 and 1994. The groups of states represented are partially overlapping: the "non-US World" includes all states but the United States; the "non-NATO World" includes all states but those belonging to NATO; and the "Non-OECD World" includes all states not belonging to the Organization for Economic Cooperation and Development (which incorporates all of the larger, fully-industrialized market democracies). 7

Table 1. World Military Spending Trends, 1986-1994

1986 1994 % Change
1. World 1297.0 840.3 -35.2
2. OECD 622.6 540.9 -13.1
3. Non-OECD World 674.4 299.4 -55.6
4. NATO 562.6 469.3 -16.6
5. Non-NATO World 734.4 371.0 -49.5
6. Non-NATO OECD 60.0 71.6 +19.3
7. United States 365.3 288.1 -21.0
8. Non-US World 931.7 552.2 -40.7
9. Non-US OECD 257.3 252.8 -1.7
10. Non-US NATO 197.3 181.2 -8.2
11. Potential Threat States1 550.0 167.0 -69.0

Notes: 1For 1986 includes member states of the Warsaw Treaty,China, Cuba, Iran, Iraq, Libya, North Korea, Syria, and Vietnam. For 1994 includes Russia, Belarus, China, Cuba, Iran, Iraq, Libya, North Korea, Syria, and Vietnam.

Sources:US ACDA, World Military Expenditures and Arms Transfers 1995(Washington DC: US Government Printing Office, 1996); International Institute for Strategic Studies, The Military Balance 1996/97, -1995/96, -1994/95, -1993/94, -1992/93(London: Oxford University Press, 1996-1995; London: Brassey's, 1992-1994.)

The category "potential threat states" (Row 11) includes those nations that have figured prominently in official US security threat calculations -- however, the composition of this category is different for 1986 and 1994. For 1986 "potential threat states" include all statesparticipating in the Warsaw Treaty as well as China, Cuba, Iran, Iraq, Libya, North Korea, Syria, and Vietnam. For 1994, Russia and Belarus have taken the place of the former Soviet Union and all other former participants of the Warsaw Treaty have been excluded. The principal finding of our review of spending trends is neatly capsulated in rows 1 and 7 of Table 1: although US military expenditures have declined, the reduction in spending worldwide is significantly greater in magnitude, yielding a substantial boost in relative position for the United States. Comparing Rows 1, 2, and 4 reveals that for the NATO and OECD countries as groups there is a similar growth in relative spending despite an absolute reduction in miliary expenditures. 8

The group of nations outside NATO and the slightly smaller group outside the OECD both show very steep declines in military expenditures. However, these declines derive entirely from the change in spending among a subset of nations within these larger groups: those states regarded commonly within US military planning as serious threats or potentially serious threats to the West -- a cluster of only 10-16 nations.

The "potential threat" states -- which in 1986 spent nearly as much on military power as did NATO -- have undergone a decline in aggregate spending of nearly 70 percent. This is due mostly to the collapse in spending by the former Soviet Union, which in 1986 had accounted for 68 percent of all spending for this group. In 1986 the Soviet Union spent the equivalent of $374 billion on defense, according to ACDA's CIA-derived figures; By contrast, ACDA records Russia's 1994 defense budget as 96.8 billion equivalent dollars -- a 74 percent decline. Also notable in the threat group are sharp reductions in spending by Cuba, Iraq, Iran, Libya, North Korea, Syria, and Vietnam, which were down 68 percent as a group. 9 Contributing marginally to the spending decline in the "threat group" was the defection of the East European nations. 10

The final significant factor in spending patterns for the "potential threat" group is the Chinese military budget, which showed 1.6 percent growth during the 8-year period under examination. 11 The increases in the Chinese budget during 1992-1994, which caused concerns in the West and across Asia, may have actually re-established a level of expenditures that existed prior to 1992. Indeed, according to ACDA, China's 1994 equivalent expenditures were slightly lower than its 1984 expenditures. 12

Other than the putative "threat states," the nations constituting the "world outside NATO" and the "world outside the OECD" show an aggregate increase in spending over the period 1986-1994. (These subsets might be called the "non-threat states outside NATO" and the "non-threat states outside the OECD.") Fifty-three "non-threat" nations outside of NATO showed real increases in defense spending during the period 1986-1994; 48 of these nations were outside the OECD as well. However, among these states only a few stand out as having instituted major spending increases in both absolute and percentage terms.

Among the nine OECD states that did not belong to NATO in 1994, Japan increased its budget during the period under examination by $10 billion -- a 28 percent increase. Also in this group,Australia increased its spending by $1.8 billion -- similarly a 28 percent increase. 13

Turning to non-OECD states: South Korea (whose accession to OECD occurred after the time period of this review) and Taiwan stand out as having implemented substantial increases in both absolute and percentage terms. South Korea increased its spending by $4.35 billion -- a 50 percent jump. Taiwan increased its spending by $2.7 billion -- a 30 percent increase. 14 Also increasing their spending by between $1 billion and $1.5 billion annually during the period under examination were Burma, Brazil, India, Kuwait, Singapore, and Thailand. 15

3. Regional Trends: South Asia, Asia-Pacific, and the Middle East

Asia: World spending patterns viewed regionally show aggregate increases for East Asia, Oceania, and South Asia over the 1986-1994 period. The South Asia increase is very small in absolute terms -- $2.3 billion -- but large in percentage terms: 23 percent. Most of this change is due to increased spending by India, Pakistan, and Sri Lanka. The East Asian increase is much larger in absolute terms -- more than $18 billion dollars -- although smaller in percentage terms: 14 percent. The states belonging to the category "Oceania" add another $1.6 billion to the total for the Asia-Pacific region (comprising East Asia and Oceania). The mainstays of this regional increase are several nations closely aligned with (or considered part of) the West -- Australia, Japan, South Korea, and Taiwan. These together show an increase in annual spending during the period of $18.8 billion. 16

Interestingly, the growth in the mainland Chinese budget accounts for only 4 percent of the Asia- Pacific regional increase. Indeed the Chinese share of regional Asia Pacific spending declined during the period from 39 to 34 percent. 17 Another nation in the region of special concern to the West -- North Korea -- shows a 22 percent decline in spending during the period. In 1986 it spent 81 percent as much as South Korea on defense; today it spends 42 percent as much. Vietnam's military budget has undergone a virtual collapse according to ACDA: it is down 85 percent from its 1986 level of 3.1 billion. Thus, behind the trend of increased spending in the Asia-Pacific region,which has attracted much attention, is a more significant development: the military spending ratio between those states allied with the West and those the West considers potential threats has shifted decisively in favor of the former group.

In 1986 China, North Korea, and Vietnam together spent the equivalent of approximately $62 billion on defense; in 1994 they spent $58.7 billion. In 1986 the pro-Western group (comprising Australia, Japan, Malaysia, New Zealand, the Philippines, Singapore, South Korea, Taiwan, and Thailand) spent $67.5 billion. In 1994 they spent $89.5 billion. Thus the pro-Western group, which spent 8.5 percent more than the "threat group" on defense in 1986, spent 52 percent more than these states in 1994. 18 Implicit in this is a reduced requirement for US military support.

Middle East: In another area of long-term concern for the United States -- the so-called "arc of crisis" comprising North Africa, the Middle East, and the Persian Gulf -- military spending dropped by 47.5 percent during 1986-1994, from the equivalent of $94.5 billion to $49.6 billion. This sharp reduction in regional spending is in distinct contrast with developments in the Asia-Pacific region. However, there is a trend in the "arc of crisis" that mirrors developments in the Asia-Pacific: the spending ratio between pro-Western and "potential threat" states has shifted decisively in the former's favor. This remains true even when Algeria and Yemen are included in the regional "threat" group (otherwise comprising Iran, Iraq, Libya, and Syria.) This group spent the equivalent of $46 billion in 1986 and $14 billion in 1994. By contrast the pro-Western group -- including Bahrain, Egypt, Israel, Kuwait, Oman, Qatar, Saudi Arabia, and the United Arab Emirates -- spent $41.2 billion in 1986 and $32.9 billion in 1994. Although both groups contributed to the regional decline in defense expenditure, the reduction for the threat group (70 percent) is far greater than that for the pro-Western group (17 percent). The pro-Western group in 1994 outspent the "potential threat" group by more than two to one; in 1986 the pro-Western group had spent 90 percent as much as the potential threat group. 19 Once again, the improved relative position of the pro-Western group implies a reduced requirement for US military support.

4. Developments Among the NATO and OECD Countries

A final interesting trend concerns the groups of states allied or aligned with the United States. Although both the "non-US NATO" group and the "non-US OECD" group have undergone spending decreases in aggregate, neither show retrenchment of a magnitude comparable to that of the United States -- at least for the period under examination. (Compare Table 1 rows 7, 9, and 10.) However, throughout the period 1986-1994 these states devoted much less of their GNP to defense on average than did the United States; they still do.

In 1994 the United States spent 14 percent more on defense than did all other OECD states combined, even though the United States produced only one-third of the aggregate OECD product. In 1994 the US devoted 4.3 percent of GNP to defense; the other OECD states invested only 1.8 percent of their GNPs in military power. But many of these OECD states border on areas of continuing or increased instability. Thus the fact that their pace of reduction has not matched that of the United States is understandable -- although they, like the United States, might absorb further cuts in military spending. At any rate, the relative change in defense expenditures among the United States and the other NATO and OECD states accomplishes a long-time goal of the United States: a more equal sharing of the burden of defense. 20

5. Assessing the Change in America's Relative Spending Position, 1986-1994

Table 2 shows the change in shares of world defense spending for the United States and several groupings of states. By presenting expenditure data in terms of percentage shares of world spending for both 1986 and 1994 the table clarifies the changes in relative spending during the period. Table 3 utilizes the same data but focuses specifically on the United States. It shows the change in the ratio of US defense expenditure to that of each of the select groups. For each group in each year Table 3 uses the group's spending figure as divisor and the US figure as dividend. The resulting quotient expresses US spending as a fraction of that of the comparison group. With each group's spending set to "1" in both of the comparison years, the change in America's relative position appears as an increase in America's spending quotient.

Table 2. Shares of World Military Spending for the United States and Selected Groups, 1986-1994

1986 1994
1. World 100.0 100.0
2. USA 28.2 34.3
3. Non-US World 81.8 65.7
4. Non-NATO World 56.6 44.2
5. Non-OECD World 52.0 35.6
6. Potential Threat States 42.4 19.9

Sources: Same as Table 1

Column 3 of Table 3 shows the percentage change in America's relative position vis a vis each of the state groupings under examination. In effect this column shows how much the United States would have had to increase its spending in 1986 to achieve the ratios of spending that exist today. Thus, Table 3 reveals that, despite the real 21 percent reduction in US spending during the period 1986-1994, the change in what "potential threat states" spend produced a change in America's relative position equivalent to a 1986 US spending increase of 157 percent.

Viewed from another perspective: had the United States cut its military spending in 1994 by 60 percent, its position relative to potential threat states would still have compared favorably with the ratio of spending that existed in the Cold War year of 1986. 21 By ACDA's accounting, the United States spent more on defense in 1986 than in any other peace-time year in the nation's history.

Table 3. US Military Spending as a Fraction of Spending by Selected Groups of States, 1986 and 1994

Ratio of US Spending to Group Spending
Ratio of US Spending to Group Spending
Change in Relative US Spending
1986 > 1994
United States Group United States Group
1. Non-US World .39 1 .53 1 +35.0%
2. Non-NATO World .50 1 .78 1 +56.0%
3. Non-OECD World .54 1 .96 1 +77.0%
4. Potential Threat States .67 1 1.72 1 +157.0%

Sources: Same as Table 1

6. Conclusion

Two important conclusions supported by this review of world spending trends are that:

  • Despite post-Cold War spending reductions, the United States and its friends and allies today have a spending edge over potential adversaries that is far greater than existed during the Cold War, and

  • The burden of defense born by the United States, its allies, and close friends is today more equitably distributed among the members of this group -- even though the United States continues to devote more of its GNP to defense than is the average for the group.

In 1994 the 25 OECD industrial democracies accounted for almost 65 percent of all military spending worldwide, NATO accounted for more than 55 percent, and the United States accounted for almost 35 percent -- in all cases a dramatic increase in spending share since 1986. "Potential threat states" today account for less than 20 percent -- down from 58 percent 10 years ago. But what is the strategic import of this change? Section 1 of this report examined a number of factors that complicate any equation of relative spending levels with strategic position. Among these were:

  • The efficiency with which nations or groups of nations convert financial resources into military power;

  • National social-economic and cultural conditions that influence military capability, but do not register as defense expenditures or costs; and,

  • Special strategic and geostrategic conditions that affect military requirements.

Any one of these factors could alter the significance of the changes in world spending patterns. However, to the extent that these factors have changed clearly in the past ten years, they mostly reinforce the trends apparent in the comparative spending data. Several developments stand out: first, the group of "potential threat states" no longer has a "coordinating center" in Moscow. It is today not even a loose association of like-minded states. Several members of this category, such as Vietnam, are eager to improve their relations with the West; others, such as Syria, have only a narrow basis of contention with the West. Most, with the notable exception of China are undergoing economic crises that make the mid-term rejuvenation of their armed forces unlikely. 22 Russia, in particular, remains stuck in a morass of economic stagnation, political stalemate, and social disruption. Its very integrity as a nation is in question.

The second development that is especially germane to the interpretation of the spending data is the allying of most former Warsaw Treaty states with the West and the withdrawal of Russian troops behind borders 600 miles east of the former "central front." This provides the heartland of Western Europe with a degree of security that cannot be easily reversed. Even if Moscow can rebuild a Soviet-scale military on a smaller economic foundation, aggression against the West would involve much greater operational challenges than before. Western Europe's substantially improved strategic situation permits increased self-reliance in defense matters and, thus, gives the United States greater freedom to use its military power elsewhere.

Several factors counterbalance America's profound edge in defense investment. As always, long- range power projection involves unique costs. Although the end of the Cold War relaxes the external constraints on America's military power, it also weakens the rationale for foreign military involvements. Outside of the framework of global superpower contention, most distant regions now arguably involve "less than vital" US interests. This has prompted concern about sustaining public support for regional interventions. 23 Hence, some policy-makers see a new requirement to conclude such interventions very quickly and with exceptionally low casualties.

The significance of America's spending edge seems to lessen when viewed in light of the combined requirements of rapidly projecting overwhelming force over great distances in order to win wars quickly, decisively, and at low cost. To this some add a requirement to handle two or more contingencies simultaneously, with or without strong allied support, and possibly in regions lacking accessible bases. Working within these parameters Pentagon planners have produced contingency plans -- force packages, deployment schedules, and operational goals -- that far surpass what the nation would have attempted in regional conflicts during the Cold War. Indeed, they substantially exceed the standards set in the Gulf War.

The objectives that define America's current "regional military strategy" are uniquely ambitious, and they can give the impression that America's military expenditure, although 35 percent of the world total, is "not enough" or "just barely enough." Judging the sufficiency of America's military expenditure depends partly on an accurate assessment of the military challenges to the nation's interests, which this report has addressed, and partly on whether America's present national and military strategies offer the best way to achieve the nation's goals. Thus, there are two separate but related questions that policy-makers must consider: How much have the revolutionary changes of the past half-decade benefitted the United States? And how best can the United States use these benefits to secure a prosperous future and facilitate the evolution of a more peaceful and cooperative international system?


1. Implicit in our use of the term "requirements" is the proviso that nations use military power for defensive ends, in defensive ways, and as a means of last resort. Of course, nations can use military power for other ends and in other ways, ranging from outright aggression to coercion. Nations can also seek to maintain military power in excess of their defensive needs in hopes of using it as a substitute for other forms of influence. The choice among these various uses of military power, and the different national strategies they imply, naturally affect the quality and quantity of military power that a nation may seek to accumulate. By our definition, however, military capabilities developed or maintained for purposes other than national or collective defense, narrowly defined, should not be considered "requirements."

2. There are analytical tools available for directly estimating the combat power of a nation's armed forces -- but these, too, have characteristic shortcomings. Simple "bean counts" of weapon systems fail to take differences of quality into account. One methodology, called TASCFORM, developed by the Analytical Sciences Corporation and used by many US defense planners, tries to account for both the quality and quantity of military power by scoring weapon systems for attributes such as lethality, mobility, and protection. Still, even sophisticated methodologies such as this usually fail to account for a variety of vital attributes, such as: overall force readiness and sustainability, the quality of personnel and leadership, and Command, Control, Communication, Computation, and Intelligence (C4I) capabilities. By contrast, attention to military expenditure (adjusted to reflect purchasing power parity) has the advantage of providing some measure of overall defense effort. Thus, the prodigal tendency of many nations to over- invest in major combat systems while short-changing readiness, sustainability, and C4I does not register as a relative strength. Reviews of national military expenditure and direct assessments of potential combat capability are best seen as complementary methods. For a discussion of the difficulties of analyzing combat capability and for an account of the TASCFORM scores of selected nations, see Michael O'Hanlon, The Art of War in the Age of Peace (Westport: Praeger Publishers, 1992), pp. 63-70.

3. The cost-effectiveness of military expenditure depends on many factors: How well does a nation's military posture match its national goals and circumstances? How sound is the military's doctrine? Has the military achieved a balance between the quantity and quality of its personnel and material that makes sense in light of the nation's circumstances? How well have planners reconciled the goals of modernization, readiness, and sustainability? How efficient are the military's organizational forms? What degree of synergy has the military achieved among its various arms, branches, and services? Detracting from cost effectiveness are bureaucratic mismanagement, pork-barrel spending and status purchases, and non-military uses of the armed forces (such as in the role of national police or as an auxiliary industrial/agricultural workforce). Also degrading military cost-effectiveness is the tendency of some nations, especially poorer ones, to use their armed forces as a means of redistributing national product and absorbing surplus labor. In such cases the size of a country's armed forces may bear little relation to national military requirements or capability.

4. Public and social expenditures can contribute to the military potential of a nation without appearing as line items in its military budget. A country with superior education and health care systems will have a larger pool of skilled and healthy individuals available for service. Similarly, a high degree of social cohesion or a strong military culture can strengthen a nation's armed forces substantially. Conversely, a nation's social problems -- such as communal tensions and political alienation from the governing order -- can infect and weaken its military establishment.

5. In order to provide military spending time series for foreign countries in constant US dollars, ACDA takes several steps: (I) it adjusts all country data for inflation using a country-specific deflator, and (ii) it converts all country data into constant US dollars using the average market exchange rate for the chosen baseline year. This method has the advantage of insulating the time series from exchange rate fluctuations -- thus producing a more accurate measure of change. For some foreign currencies, however, the market exchange rate does not reflect true indigenous purchasing power. Hence, for nations of particular concern to the United States, ACDA relies on alternative methods. For the USSR (up to 1991) ACDA uses CIA estimates of what it would have cost the United States in dollars to build, operate, and maintain a military like that of the Soviet Union. Of course, if the USSR had had a pricing pattern like that of the United States -- reflecting very high labor costs, for instance -- it most likely would have built a very different military. Recognizing this, ACDA notes that, "Estimates of this type -- that is, those based entirely on one country's price pattern -- generally overstate the relative size of the second country's expenditures..." (WMEAT 1995, p. 180) Estimates for Russia (after 1991) are derived from the earlier estimates for the USSR.

WMEAT uses a different method for estimating China's effective military budget (described in Note 11) -- but it also has the likely effect of inflating the value of Chinese military expenditure. Finally, for a number of countries (including five others that we designate as "potential threat states"), ACDA adds to their nominal military expenditures the value of arms imports because it believes that the nominal figures exclude capital expenditures -- such as weapon purchases. See Statistical Notes in US ACDA, World Military Expenditures and Arms Transfers 1995 (Washington DC: US Government Printing Office, 1996), pp. 179-186.

6. International Institute for Strategic Studies, The Military Balance 1996/97, - 1995/96, -1994/95, -1993/94, -1992/93 (London: Oxford University Press, 1996-1995; London: Brassey's, 1992-1994). The principal second source is The Military Balance 1995/1996 because it states its military spending estimates in 1994 USD.

7. The sixteen members of NATO are Belgium, Canada, Denmark, France, Germany, Greece, Iceland, Italy, Luxembourg, the Netherlands, Norway, Portugal, Spain, Turkey, the United Kingdom, and the United States. All are members also of the OECD. In addition the OECD now includes Australia, Austria, the Czech Republic, Finland, Hungary, Ireland, Japan, Mexico, New Zealand, Poland, South Korea, Sweden, and Switzerland. Our calculation of OECD military expenditures excludes Mexico for 1986 and excludes completely those states joining after 1994: the Czech Republic, Hungary, Poland, and South Korea.

8. Checking the ACDA data against data from the Military Balance strongly corroborates the direction and magnitude of these trends. Moreover, the MB data indicates that the trends apparent in the ACDA data for 1986-1994 continue through 1995. See IISS, The Military Balance 1996/97 (Oxford: Oxford University Press, 1996), Table: International Comparisons of Defence Expenditure and Military Manpower, pp 306-311.

9. Although The Military Balance uses estimates of "threat state" spending in 1986 and 1994 that are lower than those employed by ACDA, the MB figures confirm the magnitude of decline recorded by ACDA. Moreover, MB records continued decline for most "threat states" through 1995 and at a pace substantially greater than the US spending reduction in 1994-1995. IISS, The Military Balance 1996/1997, pp 306-311.

10. The real military strategic significance of the recently won independence of the Eastern European states resides in the operational obstacles it places in the way of any prospective Russian conventional military challenge to Western Europe. Moreover, the addition of these states to the Western camp and their possible inclusion in NATO places Russia in a strategic circumstance that it has not had to face since 1939.

11 By ACDA standards the Chinese military budget and its effective purchasing power are much greater than suggested by Chinese government figures. ACDA relies on a Defense Intelligence Agency calculation of the dollar cost of Chinese forces and activities. The DIA calculation utilizes a World Bank estimate of the Purchasing Power Parity (PPP) of the yuan. However, PPP are estimated with reference to a specific "basket" of goods, and the World Bank estimates are not designed to capture PPP ratios for military goods. According to ACDA, China spent the equivalent of $52 billion on defense in 1986 and $52.8 billion in 1994 (both in 1994 dollars). Other commonly-used sources set the real Chinese budget lower. For instance, MB-95/96 estimates China's 1994 military expenditure to be the equivalent of only $28.5 billion. The Stockholm International Peace Research Institute (SIPRI) estimates China's gross military expenditure to have been the equivalent of $45 billion in 1993. MB employs a PPP ratio developed by the International Monetary Fund, which estimates China's 1994 military expenditure as $36.3 billion. MB argues that this conversion is more in line with estimates derived from costing defense inputs. However, MB further corrects this figure to account for the fact that between 5- 10 percent of the Chinese expenditure requires hard currency and this portion should not be subject to PPP conversion at all. For a discussion of the problem of estimating China's military budget see IISS, The Military Balance 1995/96 (Oxford: Oxford University Press, 1995), pp 270-276; and, World Military Expenditures and Arms Transfers 1995, pp 180-183.

12. ACDA and MB disagree on the magnitude and direction of change in China's budget. MB- 95/96 records a six percent increase in Chinese expenditures during 1985-1994 -- albeit from a much lower baseline. (See Note 11). MB-96/97 also estimates a further 10 percent increase in 1994-1995. The difference in the trend lines that the two sources report may be due to different valuations of qualitative changes in the Chinese military during the period under examination -- for instance, the replacement of older weapon systems with newer, more capable ones.

13. The Military Balance 1995/96 records a much larger spending increase by Japan over the slightly longer span of its review: $13.5 billion growth from 1985 to 1994 -- an increase of 48 percent. Between 1994 and 1995 MB-96/97 records an additional 7.7 percent increase. For Australia, MB records a 16.7 percent increase during the period 1985-1994 and another 1.4 percent growth between 1994 and 1995. MB and ACDA's World Military Expenditures are in rough accord on the absolute level of military spending by the two countries in 1994.

14. The Military Balance review of spending trends 1985-1994 is in close accord with ACDA estimates. For 1994-1995 MB-96/97 reports further substantial increases in the South Korean and Taiwanese military budgets. A 13.5 percent increase for South Korea (to a $14.4 billion annual budget) and a 14 percent increase for Taiwan (to a $13.1 billion budget).

15. WMEAT also records Yemen as having increased its defense expenditures by the equivalent of more than $1 billion during the period 1986-1994. However, we do not include Yemen because much of the reported "increase" seems to be an artifact of a change in WMEAT's accounting procedure. In 1990, following the merger of North and South Yemen, WMEAT records a 65 percent boost in their combined defense budget and in their combined GNP. Furthermore, the MB data on Yemen disagrees sharply with WMEAT's. For the period 1985-1994 successive editions of MB estimate declines in Yemeni military expenditures ranging from 38 percent to 51 percent. MB and WMEAT also differ significantly on the absolute level of Yemeni military spending in 1994. Whereas WMEAT reports $2 billion in spending, MB reports $324 million (which expressed in terms of Purchasing Power Parity would be the equivalent of $700 million). By regional standards, the size of Yemen's armed forces and their equipment holdings are more compatible with the $700 million estimate. This estimate is also more in line with that of a third source, the Stockholm International Peace Research Institute.

16. MB records a greater increase from a lower baseline over the period of its review -- 1985-1994. This is due largely to the differences between how MB and WMEAT treat Chinese military expenditure. Nonetheless, the two sources agree on the general trend in the area. Moreover, MB strongly affirms the importance of the key pro-Western states to the regional trend. The MB data also shows that the key pro-western states continued to increase their spending during 1994-1995 at a significantly higher rate than the regional average.

17. Although differing on the particulars, MB affirms these general trends: the Chinese budget is growing at a much slower rate than the average for the region and, thus, its share of regional spending has fallen. According to MB-95/96 the Chinese share of regional spending has gone from 26.5 percent to 22.8 percent during the period 1985-1994.

18. The MB accounting of change among Asian "threat states" differs from WMEAT's in several significant ways. First, due to its lower estimates for Chinese spending, its estimates for the "threat" group as a whole are lower as well. Second, it records a less pronounced decline in spending by these states over its slightly longer time line. This is due to its assertion of a greater increase in the Chinese budget than reported in WMEAT and a less pronounced decrease in the North Korean budget. Interestingly, MB concurs with WMEAT on the recent size of the North Korean budget -- approximately $5.5 billion (1994 dollars). MB and WMEAT disagree, however, on the size of the North Korean budget in the mid-80s. While WMEAT sets the 1985 North Korean budget at $7 billion, MB sets it at approximately $5.7 billion (1994 USD). Thus, MB records little change in North Korean military expenditures during the past decade, which is implausible. This discrepancy may be due to differences in the handling of arms import data, which WMEAT in some cases adds to national defense budget figures to correct for a failure to report capital expenditures. One problem with such corrections is that they tend to overstate defense expenditures for those years in which major arms deals are reported or when major arms shipments occur, while understating expenditures in other years. Despite these differences, the MB accounting confirms the most significant development apparent in the WMEAT data: the pro-Western states of the region have expanded significantly their spending edge over the so-called "threat states." MB-95/96 records the pro-Western group outspending the "threat group" by 81 percent in 1985 and 162 percent in 1994. For 1995 it records further growth in the pro-Western edge -- to 171 percent.

19. Data from The Military Balance 1996/97 generally confirm these conclusions.

20. Data from MB confirms the general trend of a growing defense share for NATO and OECD members other than the United States, and shows a continuation of this trend into 1995. For 1995 the MB shows non-US members of NATO spending 70 percent as much as the United States on defense; the corresponding figure for 1985 is 59 percent. For 1995 the MB shows non-US members of the OECD spending 97 percent as much as the United States on defense; the corresponding figure for 1985 is 75 percent.

21. Data from MB confirms the magnitude and direction of this trend, and shows it continuing and deepening into 1995.

22. The remarkable growth of the Chinese economy has drawn much attention in the past few years. By some accounts, if current rates continue, the Chinese economy will double in real terms every 8-10 years. But some observers argue that China's economic expansion is not of the sustainable type. See Jack A. Goldstone, "The Coming Chinese Collapse," Foreign Policy (Summer 1995), pp 35-54; and, Richard Hornik, "The Muddle Kingdom? Bursting China's Bubble," Foreign Affairs (May/June 1994), pp 28-42.

23. The hyper-sensitivity of the American public to war casualties probably has been overstated. One recent review of US public opinion by the Program on International Policy Attitudes (PIPA) at the University of Maryland found that, while Americans certainly are sensitive to casualties in war, they are not easily deterred from a course of action that they view as necessary and just. ABC and NBC polls conducted after US troops suffered fatalities in Somalia found majorities ranging from 58 to 71 percent favoring continued involvement. PIPA tested the response of focus groups to the prospect of comparable US casualties in Rwanda and Haiti and found that an overall majority of 57 percent opted for some type of active response. One of the study's authors, Steven Kull, concluded that "PIPA's finding is consistent with a 1994 Rand study by Benjamin Schwarz assessing the effect of casualties on public opinion since World War II." Kull, "Misreading the Public Mood," Bulletin of the Atomic Scientists, (March/April 1995), pp 55-59.

Citation: Carl Conetta and Charles Knight, Post-Cold War US Military Expenditure in the Context of World Spending Trends, Project on Defense Alternatives Briefing Memo #10. Cambridge, MA: Commonwealth Institute, January 1997.

E-mail This Article


powered by FreeFind

US Defense Policy | Regional Security | Terrorism
Iraq & Afghanistan | Military & Strategic Studies
Alternative Security & Defense | Chronological

Buy Publications | Home | What's New | About PDA
Links | Search This Site | E-mail PDA

War Report | RMA Debate Page
Defense Strategy Review Page | Chinese Military Power Page
Terrorism, Counter-Terrorism, Homeland Security | Occupation Distress

Become a PDA Sustainer

Donate Now to Support PDA

The Project on Defense Alternatives, The Commonwealth Institute
P.O.Box 398105, Inman Square Post Office
Cambridge, MA 02139, USA
Phone 617/547-4474, Fax 617/868-1267
Email: pda(at)

Copyright © The Commonwealth Institute. All Rights Reserved.