XIII. Military Expenditure

Ke-young Chu, and Richard Hemming
Published Date:
September 1991
  • ShareShare
Show Summary Details

What information is available about military expenditure?

Can an appropriate level of military expenditure be defined?

Does military expenditure have a beneficial impact on the economy?

Is military spending undertaken at the expense of more socially desirable expenditure programs?

How can the efficiency of military expenditure be improved?

Military expenditure averages about 5 ½ percent of GNP across industrial and developing countries, and has grown more or less in line with average income between the late 1970s and late 1980s. Military spending is of a similar order of magnitude to that on each of the two other major functional expenditure categories—health and education. Military expenditure is a subject of special concern because of the widespread perception that it is mostly unproductive, and the resources it consumes could be more productively used to expand other economic and social programs. The most severe critics of large military budgets advocate sharp reductions in military expenditure across the world, with little or no such spending by many developing countries. Others contend that while most countries have legitimate defense objectives, these and other less compelling military activities require lower levels of expenditure than at present.

The Pattern of Military Expenditure


A prerequisite for any meaningful discussion of military expenditure is an appreciation of the current situation. However, the IMF’s Government Finance Statistics Yearbook (GFS) reports a series which is incomplete in its country coverage and there is a tendency for high or rapidly rising levels of military expenditure to be understated. Perhaps the most widely quoted data are those compiled by the United States Arms Control and Disarmament Agency (ACDA); these are derived from a variety of national sources and information collected by other US agencies, including USAID and the CIA. The principal shortcoming of the ACDA data is that they probably overstate military expenditure in Eastern Europe.

Expenditure levels and trends

ACDA data for the period 1977-87 are summarized in Table 1. These data reveal that over 80 percent of total military spending is undertaken by industrial countries (including Eastern Europe). Moreover, while expenditure in industrial countries has been growing, especially in the United States, expenditure in developing countries has been stable in real terms, and on average therefore fell relative to GDP. Among the main developing country regions, expenditure growth in Africa, East Asia and Latin America has been modest, and the share of military expenditure in GNP in the first two cases has fallen—especially in some of the more heavily indebted countries, including those in Latin America—while more rapid growth in South Asia has been swamped by a sharp decline in the heavy military spending undertaken by countries in the Middle East. Nevertheless, the list of heavy spenders (relative to income) is still dominated by countries in the Middle East (Oman, Israel, Jordan, Saudi Arabia, Egypt, Iraq, Syria, Yemen Arab Republic, and Libya are nine of the world’s ten heaviest spenders). Other heavy spenders include Angola and Mozambique in Africa, and Nicaragua, Guyana and Chile in Latin America. Except for relatively high military expenditure by China, there is little disparity between countries in Asia.

Table 1.Military Expenditure, 1977-87
In percent of GNPIn US dollars (billions)Annual real growth (in percent)
Industrial countries5.25.58442.7
North America4.86.23054.7
Western Europe (NATO)3.53.51421.7
Eastern Europe (Warsaw Pact)11.010.83641.6
Developing countries6.85.1173
Latin America2.02.0151.9
Middle East13.711.068-2.0
East Asia2.72.1702.1
South Asia3.64.1135.1
Source: World Military Expenditure and Arms Transfers (ACDA, 1988).
Source: World Military Expenditure and Arms Transfers (ACDA, 1988).

Determinants of Military Spending

Defense spending

For analytical purposes it is useful to distinguish between military spending and defense spending. National defense—that is, the provision of protection against the threat of external attack—is a classic public good. If the public sentiment is that such a threat does or could potentially exist, then it is the role of government to make the necessary provision. However, the question is how much to spend on providing protection. The magnitude of the threat a nation faces and the appropriate defense services to provide in response is difficult for most people to gauge; in practice, the most influential judges are usually the suppliers of defense services who have a vested interest in large defense budgets. Consequently, ends and means get confused, and an assessment of military spending relative to defense needs is difficult to make.

Internal security

In many countries, military spending is not only for the purpose of national defense, but also to maintain internal security. This may be undertaken by defense forces or may entail an expansion of the military. It may substitute for expenditure that would otherwise be required on the police, which is a separate expenditure category in the GFS and not covered by the ACDA data. It may be for purposes that the rest of the world regards as legitimate or illegitimate. For these and other reasons it is certainly difficult and probably inappropriate to make a judgement on the need for internal security.

Military intervention

It is one thing to be militarily prepared, but quite another to engage in military conflicts. Preparations depend upon judgements about risks and appropriate responses. A prospective aggressor is usually aware of these preparations, and instigating a conflict will often require some military build-up. Fearing attack, potential victims generally increase their defense capabilities. Actual conflict then leads to a further escalation of military spending on both sides. With an end to conflict, a reassessment of risk and response will almost certainly leave military expenditure at a permanently higher level. For these reasons, it is hardly surprising that a principal determinant of military spending is current or past involvement in civil or international conflict. This is the case with a number of the heavier spenders identified above.

Other influences on military spending

A number of empirical studies have used ACDA and other data to identify the range of factors that determine both international differences in military spending and changes over time. In addition to civil or international conflict, other political factors are also important. For example, a country’s political regime also has a bearing on military expenditure; not only do nondemocratic governments tend to place a higher priority on military preparedness but they also often require military support to retain their power. In this latter connection, it has been suggested that part of the justification for a large military establishment is that a smaller, more cohesive military force is potentially more destabilizing. In general, however, the link between the military spending and political stability is not a strong one.

Political alliances

Clearly, some countries that have few political alliances, such as neutral Switzerland, give defense a low priority. At the same time, those with strong alliances can take advantage of international cooperation to reduce defense spending. For example, in return for allowing the US armed forces to locate in the Philippines, part of the burden of defense of the Philippines is shifted to the United States. As a result, military expenditure in the Philippines is less than in many other countries in East Asia, despite serious internal security difficulties. However, while a small country may be able to benefit from an alliance with a larger country, it is less clear that the major alliances between large countries (NATO, Warsaw Pact) have reaped the benefits of cooperation. Indeed, it is widely believed that polarizing potential adversaries into two alliances has led to an escalation of military spending.

Economic factors

Population size and income levels are important factors. There is little evidence that the smallest and poorest countries are devoting an unusually large share of national income to military spending. However, while some of the heavy spenders, and in particular the oil exporting countries of the Middle East, have the resources to finance their expenditure, it is becoming increasingly clear that the burden of military spending is a strain even on these richer economies. The composition of expenditure is also relevant. In countries with modest budgets and limited foreign exchange, military costs can be held down by building up personnel rather than importing sophisticated military equipment. The evidence suggests that variations in expenditure both between countries and over time are closely linked to the size of military imports. The availability of military aid plays a central role in this regard.

The preceding discussion clearly suggests that explaining differences in military expenditure lies outside the purview of economics, being primarily a reflection of political decision making. However, there are three aspects of military expenditure that do lend themselves to economic analysis. These concern: (i) whether military expenditure has beneficial spillover effects on the economy, as is often claimed by its proponents; (ii) the efficiency of military spending; and (iii) the extent to which resources are diverted from more productive programs to pay for military spending, as is often claimed by its critics.

Economic Analysis of Military Expenditure

Military spending, investment, and growth

There can be little doubt that military expenditure has some beneficial consequences for the economy beyond any additional political, social, and economic stability associated with the provision of core military services. At the same time it is difficult to identify what these are, and this is reflected in the way military expenditure is treated in fiscal accounts. In particular, while capital spending on military hardware and construction for purely military use is viewed as unproductive and therefore classified as current expenditure, to the extent that the private sector benefits directly from the construction of roads and other infrastructure, and because provision of military accommodation relieves pressure on private alternatives, this is regarded as the creation of productive capital assets (see the A Manual on Government Finance Statistics, IMF, 1986, for further discussion), although it may not be very efficient.

However, it is argued that the beneficial impact of military expenditure extends beyond the direct provision of productive capital assets. In particular, reference is frequently made to the spin-off benefits from military research and technological developments, and the value of military training. Certainly the former is explicitly recognized by the European aircraft industry, which defends the subsidies it receives by reference to the implicit subsidy its US competitors receive through the military research and development budget. And military recruitment campaigns in many industrial countries emphasize the enhanced employment prospects that result from military training. It is also argued that military expenditure that results in some military-related orientation in the private sector can promote military exports, and that the formation of military alliances increases foreign assistance, both military and nonmilitary.

Clearly, arguments concerning research and development and the growth of the military-industrial complex relate primarily to industrial countries. However, a few developing countries—Brazil and India, for example—produce significant quantities of military equipment. The arguments about military training and foreign assistance are more relevant to developing countries. But, even if military expenditure has beneficial consequences, the issue is whether these could have been achieved more effectively through other programs. It is difficult to believe that military spending is necessarily the best way to educate the work force in developing countries or to persuade donors that they are deserving of assistance. Moreover, even in industrial countries, the argument that improving the design of commercial aircraft and the promotion of exports are best served by military expenditure is not particularly convincing.

A considerable amount of research has been devoted to the identification of the impact of military expenditure on growth. However, the evidence is contradictory, some studies finding that higher military spending is positively correlated with the growth of nonmilitary output, while others find that higher military expenditure is negatively correlated with growth. On closer inspection, it appears that the link may be negative for resource-constrained countries and positive for resource-abundant countries. The evidence therefore points to a negative impact of large military budgets on growth in developing countries. This contrasts with the so-called Benoit hypothesis, which argued that high military expenditure had been associated with more rapid developing-country growth during the 1950s and 1960s. These two opposing findings can, however, be reconciled by noting that developing countries utilized more underemployed domestic resources in this period; but more recently, the tendency to import military equipment appears to have had adverse consequences for growth.

Expenditure efficiency

Efficiency in general requires that objectives are met at least cost. Evidence from the industrial countries suggests that traditional economic analysis has rather limited applicability in the area of military expenditure. Cost-benefit analysis can in principle be applied where measurable objectives are involved, for example when choosing between alternative missile systems or deciding how many missiles to deploy. And cost-effectiveness analysis can be applied to specific defense objectives such as achieving second-strike capability in the event of a well-specified attack. But because military objectives cannot in general be defined with much precision, efficient resource allocation in the military sector is difficult to characterize.

It is clear, however, that military costs are a concern. For example, the lack of competition in US defense procurement policy has led to calls for increased public scrutiny and accountability. Insofar as most developing countries rely on imported military equipment, they may face less of a problem. The international arms market is highly competitive, and within the constraints implied by political alliances there are savings from shopping around. Perhaps the principal way in which developing countries can contain military costs is through the appropriate choice of equipment. Given the power of the military establishment, there is an inevitable tendency to confuse ends and means. This is especially true where high-technology equipment is concerned. While military prestige may well be served by having the latest weapons, this may be inappropriate given military objectives, manpower skills and available resources. A powerful military establishment may also be able to secure pay and benefits for military personnel that are substantially better than those of other public sector and private sector employees, which adds to excess costs.

Trade-off against other programs

While it is obvious that military spending is undertaken at the expense of some other programs (or lower taxes and borrowing), it is less clear that displaced expenditures are more productive. Military objectives may be perfectly legitimate, there may be beneficial spillover effects, and spending may be efficient. An optimal public expenditure program could then feature a fairly large military budget. It is, however, fanciful to believe that military budgets around the world can be justified by reference to such considerations. Critics who claim that defense expenditure ought to be reduced are probably on safe ground. But evidence is hard to come by. Attention has focused mainly on changes in the pattern of expenditure in the context of fiscal adjustment. The evidence reveals that, in the short term, military expenditure is more resilient to budgetary cutbacks than most other expenditure categories. In the long term, however, the often-heard claim that expenditure on social programs is the worst affected category is in general misplaced. The brunt of adjustment is borne primarily by public investment and other economically productive programs. Indeed, in some cases social expenditure appears to be more resilient than military spending; but both bear some of the costs of adjustment.

Policy Aspects of Military Expenditure

A sovereign country’s military objectives cannot be questioned, nor can the way in which it decides to respond to them. However, the following issues can legitimately be the subject of policy discussion: (i) the extent to which military spending is used to promote nonmilitary objectives that can be more efficiently met in other ways; (ii) the trade-offs that are implied for the size and composition of other expenditure programs given the size of the military budget; and to a limited extent (iii) the efficiency of military spending given military objectives. Issue (ii) in particular would appear to be directly relevant in many adjusting countries. Reductions in economically productive expenditure and inadequate protection of the poor during periods of adjustment imply that military spending could have a large opportunity cost in terms of lower growth and welfare. This may not be widely appreciated. There is then a strong case for arguing that military expenditure should bear at least its fair share of the costs of adjustment.


    Benoit, Emile, Defense and Economic Growth in Developing Countries (Lexington: Lexington Books, 1983).

    De Masi, Paula, and Henri Lorie, “How Resilient Are Military Expenditures?,” IMF Staff Papers, 36 (Washington: International Monetary Fund, 1989).

    • Search Google Scholar
    • Export Citation

    Hewitt, Daniel P., “Military Expenditure: International Comparison of Trends,” Working Paper 91/54 (Washington: International Monetary Fund, 1991).

    • Search Google Scholar
    • Export Citation

    Porter, R.C., “Recent Trends in LDC Military Expenditures,” World Development, 17, 1989.

    Other Resources Citing This Publication