National defense is in many ways a public, or “collective,” good, which means two things. First, consumption of the good by one person does not reduce the amount available for others to consume. Thus, all people in a nation must “consume” the same amount of national defense (the defense policies implemented by the government), although different people may value that common defense policy differently. Second, the benefits that a given person derives from the provision of a collective good do not depend on that individual’s contribution to funding it. Everyone benefits, including those who pay little or no taxes.

To say that everyone gains from defense is not to say that everyone gains from government expenditures that are labeled “defense.” There is no necessary connection between defense expenditures and actual improvements in protection from foreign threats. Some defense expenditures may not contribute to such security at all; some expenditures may contribute a great deal; and some expenditures may, by stirring up hornets’ nests, actually reduce security.

National defense, like other public goods, has an important “free-rider” problem. That is, because people benefit whether or not they contribute toward defense, each person has an incentive to wait for others to provide the collective defense good, and thus hopes to get a “free ride.” Also, because a free rider’s consumption does not reduce the amount of defense available for others to consume, even those who pay have little incentive to prevent free riding by others.

As a result of such free riding, individuals acting privately to provide national defense services would produce too little from the standpoint of society as a whole. Each person would provide defense until the incremental benefits to him equaled the incremental costs. But for society as a whole—that is, for all individuals collectively—the incremental benefits would exceed the incremental costs, precisely because once an individual provides some of the collective good, all people benefit from it and no one can be excluded. This free-rider behavior yields one of the important traditional arguments for government: By imposing taxes on all individuals and then providing collective goods, government, in principle, can eliminate freerider behavior and produce the “correct” amount of national defense and other collective goods.

How Can Government Be Induced to Provide the Optimal Amount of National Defense?

The traditional rationale for government taxation and spending on national defense is incomplete. It states that government can eliminate free-rider behavior—that is, achieve “efficiency” in the allocation of resources—but is silent on whether government has enough incentive to do so. Just as economists have shown that individuals acting alone have incentives to provide too few collective goods, public choice economists (see public choice) have shown that a democratic government acting under a majority decision rule also has an incentive to provide too few collective goods. The reason, in brief, is that the political majority can impose taxes on all citizens—or disproportionate taxes on the political minority—and then reduce spending on collective goods, such as national defense, while increasing spending on private (i.e. noncollective) goods that benefit the majority (or members of the majority coalition) but not the minority. Such transfer programs as Social Security and agricultural subsidies are examples of government spending on private goods.

If, however, the collective goods that government provides can be transformed into private ones—that is, if provision of collective goods yields ancillary benefits that some majority coalition of voters views as particularly beneficial to them—then the problem of governmental underprovision of collective goods can be at least partially offset. National defense spending does yield such ancillary benefits to special interests; defense contractors, defense-related workers, and communities with military bases are examples. The strong resistance from Congress that Defense Department officials encounter when they wish to close military bases—a resistance partly circumvented in recent years through the Base Realignment and Closing (BRAC) Commission process—illustrates the private benefits derived from such defense spending. Alternatively, defense spending can be bundled with such nondefense transfer programs as agricultural subsidies or water projects to induce majorities to fund sufficient amounts of such collective goods as defense.

Such interest groups benefiting from defense spending are located in a majority of congressional districts and in a majority of states. Their presence, therefore, offsets the bias toward too little defense spending under democratic institutions. Ironically, the geographic (and political) dispersion of military bases, defense contracts, and other seemingly “wasteful” dimensions of defense spending—not to mention transfer programs seemingly unrelated to defense—could make government budgeting more efficient rather than less.

Optimal Taxation When Government Provides National Defense

The traditional theory of optimal taxation (see taxation) posits that the kinds of taxes imposed and the rates levied should minimize economic distortions; that is, they should interfere as little as possible with the choices taxpayers make in the private sector. This traditional theory assumes implicitly that the size and composition of the government budget are independent of the kinds and magnitudes of the taxes imposed—a highly unrealistic assumption. If government programs benefit one group of voters but are financed by another, the beneficiaries will demand larger programs than would be the case if they were required to bear the tax burden themselves. For such collective goods as defense, political processes are more likely to yield the optimal amount of spending if each individual pays taxes proportional to the individual benefits derived from the defense services provided.

Some defense programs protect against the threat that foreign aggressors will kill and injure Americans or confiscate or destroy American property. Such programs defend individual liberty, political freedom, and the domestic political system against foreign threats. In addition, the U.S. defense budget is used to support many foreign policy objectives and so is dedicated, in part, to protecting foreigners and foreign-owned assets. The post–World War II commitment to NATO—however anomalous it has become with the collapse of the Soviet Union—is the most obvious example.

Nonetheless, the protection of domestic wealth from confiscation or destruction by foreigners is a large part of the service provided by national defense policy, particularly in an era characterized by important threats from international terrorists. Accordingly, a substantial part of the demand for national defense reasonably can be attributed to those U.S. residents owning assets threatened by foreign aggressors; lives (human capital) are foremost among them. Some kinds of assets are more vulnerable (or more threatened) than others: American-owned assets located overseas may be more vulnerable than identical ones in the United States, and ocean tankers that can be moved may be more vulnerable to confiscation but less vulnerable to destruction than office buildings. In any event, a rough surrogate for individual valuation of defense services may be individual wealth; accordingly, an important component of a tax system designed to yield appropriate democratic choices on the size of the defense sector is a tax on wealth. Taxes on income or consumption may provide good approximations for such a tax. Similarly, individual preferences for political freedom, for protection of the political system, and for foreign policy maneuverings are likely to be correlated positively with individual wealth. Again, taxes on wealth, income, or consumption may be appropriate, particularly if the revenues yielded by such tax instruments are earmarked for defense purposes.

International Defense Alliances

Nations facing a common threat often pool their defense efforts in alliances such as NATO. While NATO is a formal alliance, nations also can cooperate implicitly in informal alliances, sharing defense responsibilities and burdens without the trappings of an international organization. Indeed, in principle, little communication between such informal partners is necessary; each nation can undertake given defense activities knowing that the other(s) will pursue complementary activities in response. Whatever the nature of the alliance, the defense (or other) efforts aimed at common goals are, again, a collective good. Thus, nations, like individuals, face the free-rider problem and the resulting underprovision of defense. Because larger nations like the United States are likely to value the collective defense effort more highly than smaller ones like Belgium, small nations may attempt to exploit large ones by free riding on the larger countries’ defense efforts. Thus, members of alliances often bargain over “burden sharing,” or the specific efforts each is to make.

One problem with achieving an equitable and efficient sharing of burdens is the determination of the appropriate allocation of burdens in principle. Even if nations agree that contributions should be “proportional,” the obvious question arises: proportional to what? Population, GDP, per capita GDP, and physical proximity to the perceived threat are all plausible candidates. Moreover, even if nations agree on precisely what efforts should be proportional to, alternative measures of effort can yield very different answers. One reasonable measure is military spending; another is the contribution of physical defense assets (troops, equipment, etc.). Contributions of physical assets and manpower may not be proportional to spending because of differences in valuations or pricing, differences in efficiency, and a host of other factors. For example, U.S. defense spending during the mid-1980s (while the perceived Soviet threat was substantial) was roughly 60 percent of the NATO total, while the United States provided about 46 percent of the main battle tanks and about 40 percent of the division-equivalent firepower (a measure of military effectiveness). The German Federal Republic (West Germany) provided about 8 percent of NATO spending, but about 17 percent of the main battle tanks and 13 percent of division-equivalent firepower. At the same time, the U.S. naval contribution was vastly disproportionate. Such indexes are crude, but they illustrate the difficulty of measuring relative contributions.

No definition of “fairness” in burden sharing is obviously correct, and this ambiguity inexorably creates tensions within alliances. Furthermore, citizens of one nation may value the collective defense effort less or more than do citizens of another nation, and they also may perceive the seriousness of the threat differently. That was the case in NATO for many years, as the United States and West Germany perceived a greater threat than did, say, Greece. In the counterterrorism context, U.S. contributions appear vastly disproportionate; but other nations may perceive a smaller threat and/or may believe that U.S. policy will not reduce the threat in the long run.

Dependence Versus “Vulnerability” in Foreign Defense Procurement

Modern military forces combine many kinds of manpower and physical matériel. Inevitably, some of these inputs, such as rare metals and electronic components, are purchased from foreigners because they are cheaper abroad than at home. Some worry that foreign procurement makes the United States vulnerable to a cutoff of foreign-supplied items; they fear that cuts in foreign supplies may exceed in both number and variety the potential supply reductions from domestic firms.

This view is misguided. Suppose that some defense good is purchased from foreign suppliers and that this arrangement is subject to easy but unpredictable cutoffs. Suppose, also, that such interruptions are easy to insure against, with stockpiles, alternative suppliers in other parts of the world, or excess production capacity in the United States. If so, foreign dependence does not yield vulnerability. The central question, therefore, is not the source of the defense goods but the ease with which interruptions in supply—whether foreign or domestic—can be insured against or hedged. If domestic dependence is more difficult to insure against than foreign dependence, then, ironically, domestic dependence may yield greater vulnerability.

What could make insurance more difficult for domestic purchases than for foreign ones? One possibility is the expectation of price controls during future conflicts. Producers of defense-related goods know that the prices of such goods can rise dramatically when a government at war or preparing for war increases its purchases of those goods. These price increases serve an important function: they reward domestic producers for stockpiling goods in advance, for maintaining excess production capacity, and for increasing production quickly. But domestic producers also know that governments attempting to constrain budget increases and reacting to political pressures on “war profiteering” and the like often impose explicit or implicit price controls on just such goods. The imposition of price controls on petroleum products during some past wars is but one example. Taking anticipated price controls into account, domestic producers would not invest as much in stockpiles or excess production capacity as is optimal for society. Nor are they likely to increase production as much when price controls are imposed. But governments for the most part cannot impose price controls on foreign producers. On net, therefore, foreign producers actually may have stronger incentives to stockpile and to maintain excess production capacity. The “vulnerability” issue is thus far more complex than the common foreign/domestic dependence view suggests.

Efficient Delivery of Defense Services

With the collapse of most systems of state socialism, the U.S. Department of Defense (DoD) may be the largest centrally planned economy in the world. And there is little reason to believe that central planning works better in the United States than elsewhere.

Central planning in the DoD creates the two problems common to all centrally planned systems. First, the decision makers who design weaponry or who specify the characteristics and performance features of equipment designed by contractors have relatively weak incentives to respond to their consumers’ preferences. These “consumers” are the soldiers in the foxholes, the airmen facing dogfights and antiaircraft fire, and so forth. A good proxy for these ultimate consumers may be the theater commanders charged with winning battles. But few institutional incentives at the DoD systematically align the preferences of the decision makers with the preferences of even the theater commanders. The absence of a profit motive weakens incentives at the DoD to shape decisions according to the perceived preferences of users, and the pressures from Congress, while difficult to describe in simple terms, do not unambiguously move DoD decision making in that direction. The relative absence of competition in the provision of defense services weakens these incentives further.

As a result, the DoD and the individual armed services often promote weapons designs with dubious combat features and effectiveness. Consider, for example, the U.S. Air Force A-10 Warthog aircraft: the Warthog, designed for the support of U.S. Army and Marine Corps ground operations, has performed well in that mission. Yet the Air Force for years has tried to eliminate funding for the A-10 precisely because it supports the other services, and thus yields few bureaucratic benefits for the Air Force. The Air Force has preferred to use F-16s and other more glamorous aircraft for ground support, although their great speed may make these planes less suitable for such missions. One way to get weapons and other equipment that conform more to the demands of users is to give the users a larger voice or a direct veto in design decisions. This overall incentive effect should not be exaggerated—certainly the DoD civilian leadership has personal, professional, and political incentives to win conflicts and to minimize casualties—but centralized planning causes important adverse effects.

The absence of a profit motive and of an individual or group with a claim to the economic benefits from cost efficiencies weakens the DoD’s incentive to minimize costs in the pursuit of given objectives. Contracts for the design and production of weaponry often are written on a cost-plus basis, under which the contractor receives a payment from the government equal to (approved) costs plus some predetermined “profit.” This weakens incentives to minimize costs. On the other hand, if the contractor simply receives a fixed price for the delivery of specified defense goods, incentives to minimize costs are strengthened, but the risks of increases in the prices of inputs, unforeseen contingencies, and the like are borne by the contractor, which may not yield an efficient allocation of risk bearing.

The DoD’s own incentives for efficient operations are weak relative to those in the private sector. Because the military services have sharply defined tasks with limited overlap, each service is, in effect, a “monopolist” in its defined missions. The Army, for example, is proscribed from flying fixed-wing aircraft in combat missions, thus giving the Air Force greater monopoly power in the provision of close air support for ground operations (although Army helicopter operations are an obvious substitute for many such missions). Because each service is likely to have better information than Congress about the minimum cost of providing given defense services, the services’ efforts to maximize their budgets or their discretionary budgets can make national defense more costly than necessary. An important way to reduce these costs would be to have the services compete to provide various defense operations. The Army and Marine Corps could be required to compete on a much broader scale in the production of ground combat operations. The Army and the Air Force could compete in the provision of close air support for ground combat. The U.S. Navy could compete on a much broader scale with the U.S. Coast Guard in terms of sea operations, perhaps, in particular, in the context of counterterrorism efforts. Such budget competition, similar in many respects to competition in the private sector, would likely reduce the cost of defense programs.

The Defense Establishment’s Threat to the Polity

An armed defense establishment always poses some level of threat to the civilian government and to people’s rights and liberties. The Founding Fathers recognized this threat clearly; in August 1789, when Congress was considering the Bill of Rights, Congressman Elbridge Gerry wrote, “What, sir, is the use of militia? It is to prevent the establishment of a standing army, the bane of liberty.” That is why the Constitution empowers Congress “to provide and maintain a Navy,” but, in contrast, only “to raise and support Armies.” The founders recognized that the threat is posed mainly by a standing army, not by a permanent navy. Thus, two institutional arrangements have evolved from the earliest days of the American Republic. First, the ground forces are divided into the army, the Marine Corps, and the states’ national guards; this means that any officers planning a coup must confront the potential opposition posed by the other ground combat services. Second, the Founding Fathers recognized from their own experience that an armed citizenry is better able to resist efforts of governments—of centralized military establishments—to impose autocracy or dictatorship. Accordingly, they specified the “right of the people to keep and bear arms” in the Second Amendment to the Constitution. Unlike the First Amendment, which places constraints only on the Congress, the Second Amendment decrees that the right to keep and bear arms “shall not be infringed.”


National defense, while not a separate field of study within economics, raises a vast range of economic issues, and like other areas of public policy, is suitable for the prescriptions yielded by economic analysis.

About the Author

Benjamin Zycher is a senior fellow at the Manhattan Institute for Policy Research and was previously a senior economist with the RAND Corporation.

Further Reading

Brennan, Geoffrey, and James M. Buchanan. The Power to Tax: Analytical Foundations of a Fiscal Constitution. Indianapolis: Liberty Fund, 2000.
Cooper, Charles A., and Benjamin Zycher. “Perceptions of NATO Burden-Sharing.” RAND Corporation paper no. R-3750-FF/RC, June 1989.
Enthoven, Alain C., and Wayne K. Smith. How Much Is Enough? Shaping the Defense Program, 1961–1969. New York: Harper and Row, 1971.
Halbrook, Stephen P. That Every Man Be Armed: The Evolution of a Constitutional Right. Albuquerque: University of New Mexico Press, 1984.
Hitch, Charles J., and Roland N. McKean. The Economics of Defense in the Nuclear Age. Cambridge: Harvard University Press, 1960.
LaPierre, Wayne. Guns, Crime, and Freedom. Washington, D.C.: Regnery, 1994.
Lott, John R. The Bias Against Guns. Washington, D.C.: Regnery, 2003.
Malcolm, Joyce Lee. To Keep and Bear Arms: The Origins of an Anglo-American Right. Cambridge: Harvard University Press, 1994.
Stockfisch, Jack A. Plowshares into Swords: Managing the American Defense Establishment. New York: Mason and Lipscomb, 1973.
Thompson, Earl A. “Taxation and National Defense.” Journal of Political Economy 82, no. 4 (1974): 755–782.
Wolf, Charles Jr., and Benjamin Zycher. “European Military Prospects, Economic Constraints, and the Rapid Reaction Force.” RAND Corporation paper no. MR-1416-OSD/SRF, 2001.
Zycher, Benjamin. “A Preliminary Benefit/Cost Framework for Counterterrorism Public Expenditures.” RAND Corporation paper no. MR-1693-RC, 2003.
Zycher, Benjamin, Kenneth A. Solomon, and Loren Yager. “An ‘Adequate Insurance’ Approach to Critical Dependencies of the Department of Defense.” RAND Corporation paper no. R-3880-DARPA, 1991.