Oil Imports: An Overview and Update
of Economic and Security Effects

John L. Moore, Carl E. Behrens, John E. Blodgett

Environment and Natural Resources Policy Division

December 12, 1997

98 -1

Contents

Introduction

Oil Imports as a Policy Concern

Framework for Analyzing External Costs

Potential External Costs and Recent Estimates

Economic Costs and Imported Oil

Effects of High Levels of U.S. Oil Imports

Effects on the Economy from Rapid Fluctuations in Energy Prices . . .

Military Costs and Imported Oil

Overview of Policy Issues

Measuring Persian Gulf Security Costs

Policy Options and Effects

Environmental Costs and Imported Oil

Conclusion

Introduction

Because the United States is both a large consumer and importer of oil, supply and price disruptions in world oil markets may have avoidable costs for the economy and hence effects on national security. Debate over policies to mitigate or prevent these adverse effects or external costs continues, despite many legislative visits to the issue since the 1973 energy crisis. Policy analysts, in attempting to quantify these potential adverse effects, translate estimated impacts into a premium associated with oil imports, usually on a per-barrel-basis. The premium is an attempt to approximate the monetary value of potential external costs to the nation associated with the large dependence on imported oil. Such estimates may provide information and guidance for policy decisions on what avoidance actions and costs may be warranted to reduce potential risks from import dependence. Avoidance costs may be direct, such as through tax incentives or disincentives, or indirect through programs to shift fuel sources or reduce consumption.

The policy issues and various estimates of external costs of oil imports were covered in a 1992 report by CRS entitled The External Costs of Oil Used in Transportation.¹ This report provides a brief update and extension to the 1992 report. That report was prepared at the request of Congress and summarized for the U.S. Alternative Fuels Council what was known about monetary estimates of the side effects (external costs) from oil used in transportation. Potential side effects covered were environmental, military, and economic, including the potential effects on the economy from large oil import dependence. This report focuses only on oil imports, but includes associated military and environmental issues. The report was prepared for the Honorable Lee Hamilton and is being released for congressional use with his permission.

Oil Imports as a Policy Concern

The federal role in oil and other energy markets is minimal today compared to the programs and regulations put in place after the 1973 Arab oil embargo and the oil market turmoil associated with the 1979 Iranian political upheaval. Programs that followed these two energy crises attempted to diversify and develop alternative domestic energy sources; to accelerate energy efficiency in the private and public sector; to establish a contingency reserve of oil (the Strategic Petroleum Reserve); and to fund a variety of research and development programs aimed at energy production and use.

While the United States is more dependent on imported oil in 997 than at the height of the energy crises, vulnerability to supply and price disruptions has changed substantially in the last 20 years.² The sharp increase in the relative price of oil following the two crises and the deregulation of domestic oil prices in the early I980s set in motion exploratory and technological investments that have increased available oil production capacity and diversified the sources. Indeed, proved world oil reserves were up 55 % in 1996 compared to 1980, whereas world production was up only 6% over the same time period.³ Today 20 % of U.S imports come from Persian Gulf sources compared to 37% 20 years ago.⁴

In addition, important changes have occurred since 1980 in the way oil is traded, making oil markets more efficient and thus less volatile. First, traditional long-term contract arrangements between producers and consumers gave way to a growing spot market in which large volumes of oil are traded on a daily basis. The dominance of the spot market has integrated the oil market to an international level, so that oil prices are set more directly by changing world supply and demand conditions. Thus, regional imbalances in supply and demand are shifted quickly to an international level based on the strength of buying in various consuming markets. Along with the spot market, the development of a futures market for oil and oil products has also reduced volatility and vulnerability of the oil market to large price shocks. Because prices are posted daily, the futures market adds valuable information for buyers and sellers about present and future market expectations. Perhaps more important, futures contracts reduce the need to physically hold speculative inventory, thus reducing the destabilizing role that inventory imbalances played in earlier periods of oil market instability.

Oil prices dropped sharply in late 1985 as a result of growing production and the adoption of market-based pricing. Since 1986, prices have fluctuated between $15 and $25 per barrel, spiking briefly to around $33 per barrel for one month in 1990 during the early phases of the Gulf conflict. Small releases from the Strategic Petroleum Reserve at the outset did not seem to affect rising prices and proposed sales after the bombing of Iraq began in early 1991 came as oil prices were dropping sharply. Since the Gulf conflict, oil prices have continued to fluctuate based on seasonal as well as world demand.

Despite a relatively stable oil market in the last several years, concern remains that excessive reliance on imported oil poses avoidable risks and costs to the U.S. economy. This concern may grow as the proportion of U.S. oil supply coming from imports grows in the future and the large, low-cost Middle East reserves again are a bigger portion of those imports.

Analysts studying the issues of oil import risks group these risks and their potential external costs in the following broad categories:

• The potential economic costs of less than fully competitive oil markets;
• The potential economic costs of large short term price shocks; and
• The potential for other social costs such as environmental damages.

Some also argue that some portion of military expenditures for Middle East operations should be considered a cost of heavy dependence on imports.

The federal government has responded to these potential risks and costs to the economy over the last two decades with several policies and long term programs including:

• Using political and military policy to minimize potentials for major oil market disruptions;
• Supporting public and private sector efforts to increase the energy efficiency of the economy;
• Promoting an efficient and competitive domestic oil industry through policies on oil and gas leasing, research and development, and equitable tax treatment;
• Supporting development of cost-effective alternatives to oil; and
• Developing and malntalning contingency oil supplies (Strategic Petroleum Reserve) and participating in international arrangements to blunt the impacts of oil price shocks.

The issue for public policy and congressional attention is finding the balance between the social benefits of government actions, such as those listed above, and the direct and indirect costs of such actions.

Framework for Analyzing External Costs

Most of the costs of producing, transporting, refining, and consuming oil are borne by the beneficiaries of oil use. However, the production, transportation, and consumption processes are not without side effects or external costs. With the presence of these market failures or external costs, beneficiaries may not fully pay the real costs of oil in the price of petroleum products they buy.

Policy prescriptions for dealing with external costs often start from an economic perspective. Economic theory holds that the price of a commodity such as oil should reflect its marginal cost to society, where costs include economic resources used to produce and deliver the commodity plus costs to environmental resources, human health, and energy security etc. which are termed external costs. External costs or externalities are defined as side effects of production or consumption activity that affect producers and consumers who are not part of the activity. Such costs are not considered by the generator, are not directly monetary, and may imply a misallocation of society's resources.

From a policy perspective, the ideal is to find a balance between the benefits of reduced harmful externalities and the costs of reducing the externality. In theory, the balance is found where incremental costs of avoidance are equal to the reduction in incremental external costs. At that point, damages to other activities have been internalized in the price of the externality-generating activity and additional expenditures to reduce remaining damages are not warranted on economic grounds.

Actually estimating external costs is fraught with conceptual as well as measurement problems, and estimating costs of avoidance can be equally contentious. The analytical community can offer approximations of the monetary value of various external costs, but such analyses often leave large areas for professional, public, and political disagreement.

Potential External Costs and Recent Estimates

Estimates of various external costs associated with oil are presented in the table in appendix A The table is from the 1992 CRS report⁵ and includes approximations of costs for oil in total and for the transportation sector. These estimates were assembled from a variety of studies and provide ranges of possible monetary value for various external costs. Cost components included in the appendix relevant to this report include:

• Effects of market power
• Risks of market disruption
• National security
• Oil spill risks.

Other environmental costs included in the 1992 report are not totally relevant to this review, since domestic environmental side effects arise from the transport and combustion byproducts of oil, not the source of the oil. Policy issues relevant to each of the above categories are sketched below along with a brief discussions of how and whether current circumstances have affected possible external cost estimates.

Economic Costs and imported Oil

Policy concerns during the energy crises of the I 970s dealt with the perception of a growing scarcity of world oil and sharply rising real oil prices as world demand increased in future years. With the oil market and oil supply changes occurring since that era, the focus of policy has evolved from a physical supply concern more to one focused on the economic and security implications of high levels of imports and price volatility. Analyses of U.S. energy security and the external cost of oil focus on two broad and interrelated themes. The first is the potential for gains to the U.S. economy by offsetting the possible distorting effects of OPEC market power through the exercise of market power as a large consumer. This area of potential economic gain is directly related to the level of U.S. imports. The higher the level of U.S. oil imports, the greater the potential effect on oil prices through possible cuts in U.S. demand The second area where policy intervention plays a role is offsetting the risk of sharp short term price shocks which may cause distortions and inefficiencies in investment decisions and the macro economy.

Effects of High Levels of U.S. Oil Imports. United States demand for world oil is large and some argue that reducing this demand by various government interventions could be directly and indirectly beneficial to the U.S. economy as a result of lower oil prices paid by U.S. consumers. The argument is that the United States could use its potential monopsony or large buyer power to offset the possible market power of OPEC. By cutting back on the use of oil, the United States could directly benefit from lower world oil prices and capture some of the gains or cartel rents that may exist if OPEC is actually restricting output below a competitive market level. The economy could indirectly benefit through reducing unfavorable effects on the U.S. merchandise trade balance which may affect the exchange value of the dollar;⁶ through possible positive effects on productivity due to reduced wealth transfer leading to more investment and innovation; and through reduction in possible structural unemployment in energy production sectors. Other effects cited as reasons for reducing imports include reduced potential for structural inflation due to rising energy prices; and the increased risk of price disruptions if world oil capacity is strained due to high U.S. import demand.

As discussed in the earlier CRS report, analysts disagree on both whether any premium at all should be attributed to the presence of market power and the possible magnitude of such a premium. The key issues in this category are whether the United States would take action counter to free trade practices and if the United States did, whether OPEC would retaliate by restricting supply, thus offsetting any benefits to the United States from reducing its demand for imports. Because of this range of conditions and assumptions, estimates of possible benefit to the economy of actions to reduce imports range from nothing to several billion dollars annually. If OPEC cut production to offset the U.S. cuts in demand for imports by an equal amount, then there would be no decline in price and no gain to U.S. consumers. If OPEC made no change in output in response to cuts in U.S. demand, savings to U.S. consumers could be several dollars per barrel of oil or on the order of tens of billions of dollars annually. It seems unlikely, however, that OPEC would not take some retaliatory action, meaning that benefits would be small. Such hypothetical ranges also beg the question of how and at what cost reduction in imports would be achieved. Beyond contentious international trade restrictions, it could cost more to achieve reductions than the savings realized to consumers.

In weighing the merits of arguments for considering OPEC market power in placing a premium on imported oil, policy analysts in the last few years take one of two views. One school argues that OPEC supply behavior resembles that of a traditional cartel; that world oil price is substantially above the marginal cost of production; and that any price premium to account for a depleting resource is small because of a large reserve base and the role of technology in offsetting any rise in reserve replacement costs.⁷ On this basis, some premium should be attributed to imported oil for policy purposes. On the other hand, some policy analysts argue that there is little evidence that OPEC prices reflect the exercise of any significant market power, given the group's internal rivalries and need to generate large amounts of foreign exchange to cover indebtedness. In this view, no policy relevant action is justified including attributing a premium to oil imports.⁸

Effects on the Economy from Rapid Fluctuations in Energy Prices. Even if the United States could reduce oil imports substantially from politically sensitive regions, the integrated nature of the world oil market means that the domestic economy could still be subjected to potentially wide swings in oil prices.

When prices spike unexpectedly, a number of impacts may hurt the economy. An immediate impact is increased payments for imports. Sharp increases in the cost of oil may not be adequately accounted for in the investment decisions of oil consumers and producers. While there are mechanisms in the economy for anticipating price changes, such as futures contracts, if these are inadequate, a jump in the payments for oil imports can be an economic cost not adequately reflected in the price of imported oil. The argument is that the various actions of the private sector to anticipate sharp price swings are not enough to protect the economy and that there are cost-effective public sector interventions that could reduce avoidable social or external costs. Others, however, argue that additional import costs should not be included as external costs since there are ways to protect against them and government intervention may not improve on the hedging and adjustment decisions of energy users.

Unexpected short term price swings can also produce economic impacts that may include external costs in the form of macroeconomic adjustments. If short term price movements are large and very rapid, adjustments in input and output markets may be more costly than if adjustment took place more gradually over a longer time period. The result is loss of potential GDP. This occurs because of adjustment problems that affect other sectors (for example, labor contracts affecting wage rigidity and unemployment, or obsolescence of energy intensive equipment). Other potential rigidities that may impede adjustment to short term energy price fluctuations include regulation of electric prices and possible differences in regional energy use and production.

As pointed out by Bohi and Toman ⁹, what is at issue for policy purposes are market failures that impair the private sector process of hedging against short-term price fluctuations. If the private sector takes precautions through adequate storage, purchase of flexible energy using or less-energy intensive equipment, and/or use of various financial instruments for hedging and balancing of price movements, then there are no external costs from short term energy price fluctuations.

Some argue, however, that private sector actions do not sufficiently hedge against the negative effects of short term price fluctuations. This argument is in part justification for investment in the Strategic Petroleum Reserve, but some hold that short term price fluctuations may still impose avoidable costs on the economy that warrant additional policy actions to deal with market failure.¹⁰ Under these circumstances, an oil import premium associated with price shocks may be called for.

Estimates for this potential component of a oil import premium in the 1992 CRS report were in the $6-9 billion range, based on a variety of studies. Such estimates are based on various macroeconomic-energy models using varying assumptions on probabilities and magnitudes of price shocks. The review and analysis by Bohi and Toman notes a Department of Energy estimate from 1991 in the range of about $3-8 billion annually.¹¹ They also point out that federal expenditure on energy research and development, the Strategic Petroleum Reserve, and other energy security-related DOE expenditures may already internalize much of any energy security premium.

Military Costs and Imported Oil

Although Persian Gulf oil has declined significantly as a percentage of U. S. oil imports, possible threats to the region's oil supply are seen as an important strategic concern in both diplomatic and military terms. Possible disruptions in Middle East oil supply affect world oil price, which directly affects the U.S. economy, and supply disruptions may affect important U.S. allies who are more directly dependent on oil from that area. Thus, some argue that some portion of the cost of U.S. military presence in the region should be attributed to our import dependence in terms of policy actions to reduce that dependence.

Overview of Policy Issues. In the wake of the 1990-91 Persian Gulf War, critics of U.S. energy policy argued that the price of imported oil did not reflect the military costs of defending the region. This argument formed part of the support for a variety of policy proposals, ranging from oil import fees and quotas, to subsidies for alternative fuels and stimulants to domestic oil production, and stringent oil conservation measures.

The Congress debated and discussed these issues at great length following the Gulf War, culminating in the Energy Policy Act of 1992. Despite the many diverse provisions of the final statute, however, the outcome was a clear rejection of a major government role in reducing oil consumption and imports. Proposals to impose stringent fuel economy standards on automobiles, for instance, were dropped from the legislation, as was leasing of potentially large oil deposits in Alaska's Arctic National Wildlife Refuge.

In 1992 CRS analyzed the national security costs of oil used in transportation, based on information and estimates available at that time. ¹² Since then, several studies of the issue have been published, but no substantively new information has been presented. In particular, current estimates of the military cost of defending access to Persian Gulf oil are not available. The Department of Defense consistently declined to make cost calculations on a regional basis, and continues to do so. The General Accounting Office, which published one such study, has not updated its figures; nor have the private analysts who previously made such estimates.

What follows, therefore is a summary of the conclusions published in the 1992 CRS study. The intervening years have revealed no reason to change them significantly.

Measuring Persian Gulf Security Costs.

As stated in the 1992 report, the security cost of Gulf oil imports "is either insignificant or ponderous, depending on the assumptions made."¹³

• If only those military activities are tallied which would not be undertaken in the absence of oil in the region, then the cost of defending the oil would be very small. If any military activity is counted that, undertaken for whatever reason, contributes to the security of the region, the cost calculation becomes much larger.
• The value of the resource being protected can be variously interpreted. Some analysts used the relatively small proportion of oil imported to the United States from the Gulf countries as the base for calculating a "per-barrel security cost" of import dependence. Others used the entire production from the region. Still others¹⁴ pointed out that oil has been a world market since the collapse of crude prices in 1986, and that an interruption of cheap Middle East oil supply would push up the price of oil for everyone. From that point of view, the proper denominator is world oil production, leading to a very small per-barrel figure.

Combined with the difficulty in identifying region-specific military costs, the wide differences resulting from these varying assumptions make "security costs of oil" a very uncertain policy justification.

Policy Options and Effects. Even assuming that a reliable estimate of security costs were available, the nature of the world oil market makes it difficult to find practical policies to compensate for them.

• Basic economics advises that external costs be internalized so that consumer demand will reflect the true price of the commodity.¹⁵ However, the most readily available tools for internalizing such costs consist of taxes and import fees, both of which have proved economically clumsy and politically unpopular.
• Mother policy option to ameliorate dependence on Persian Gulf oil is subsidizing alternative fuels or domestic oil production so they can compete with the lower cost of imported crude. Unfortunately, subsidizing these alternatives to the level required to eliminate the need for defense of the world oil market, even assuming it could be done, would be prohibitively expensive. It is sometimes argued that any reduction in U.S. imports would reduce the "pressure" on the world market, but that position is conceptually flawed. In the absence of a shortage of a commodity - and proved world oil resources have increased steadily since the 1973 energy crisis - market forces tend to balance demand and supply. Reducing U.S. consumption would reduce world production rate, but it is hard to see how this would increase the stability of the market

In summary, as the 1992 report concluded, there may be other reasons for supporting efforts to reduce imports, but they will have little effect on the amount spent for security in the Persian Gulf

Environmental Costs and Imported Oil

Since the bulk of imports are transported by tanker, there is increased potential for oil spills in ecologically sensitive areas in contrast to shipment by pipeline from domestic oil fields. However, some domestic oil is also shipped by tanker and some pipelines involve important environmental risks. Since any policy designed to reduce import dependence would only reduce oil imports marginally, the issue for policy purposes is what if any risk might be avoided at what cost to the economy.

Any external environmental costs of imported oil primarily arise from the need to transport it to the United States. Potential external costs of oil use -- such as air pollution, small spills occurring in the course of the distribution of fuels, and improper disposal of used oil -- are the same whether the oil originates domestically or is imported. But external costs associated with the transport of imported oil do not arise only with foreign-produced oil: some domestically produced oil also requires transport, including that from Alaska. The Exxon Valdez oil spill, the most costly to hit the United States, involved domestic oil.

The external costs from oil spills include the short- and long-term effects on local aquatic and terrestrial resources, and attendant direct and indirect losses in economic well-being for the values associated with those resources. The total loss of the Exxon Valdez spill ranged from $3 to $15 billion,16 included a basic natural resource damage settlement of about $1.3 billion, Exxon's response cost of $2.5 billion, and numerous court actions as well as federal fines.

Actions to reduce the risk of such spills, and to insure compensation for damages when they do occur, have been taken. The Ports and Waterways Safety Act, the Clean Water Act (CWA), and the Oil Pollution Act of 1990 (OPA) establish national policy to prevent, mitigate, respond to, and clean up oil spills. A key requirement is that tankers in U.S. waters be equipped with double hulls by 2015, a considerable phase-in period adopted in consideration of the anticipated costs. The American Petroleum Institute has estimated that the costs for complying with this mandate of the OPA will total about $4 billion. Other regulatory costs include contingency planning required by the CWA and propositioning cleanup equipment, as required by the OPA.

Restoration of damaged natural resources and prompt settlement of economic losses, such as loss of income and tax revenues, are also national goals. The Oil Pollution Liability and Compensation Fund (FUND), maintained by a $0.05 per-barrel fee on oil received at refineries, is now the vehicle for assessing the oil industry for many oil spill costs and for paying for such costs. The CWA and the OPA also set requirements for financial responsibility and for purchase of Oil spill liability insurance.

These regulatory and compensation programs already internalize much of the costs of oil spills (in most cases without distinguishing foreign and domestic oil). The FUND is maintained at $1 billion and is available to cover a wide range of costs including administrative, preventative, research, response, and mitigation It addresses an extensive list of damages, including natural resource restoration and replacement, real or personal property losses, loss of subsistence use, loss of revenues, loss of profits or earning capacity, and loss of public services.

Given the internalized Costs of oil spill prevention and mitigation, it is difficult to determine whether imported oil poses further external costs and, if so, whether the costs of avoiding those externalities would, on the margin, exceed benefits.

Conclusion

From the above, four general conclusions seem warranted. First, if there is a premium associated with imported oil, it would be lower today than during the period when many of the studies of the issue were undertaken. This is mainly because of greater efficiency in the world oil market and of the more diverse and abundant supply sources than during the two periods of energy crisis. Second, estimates of current premiums for imported oil depend extensively on what assumptions one makes about supplier behavior, severity and probability of market disruptions, and extent to which private hedging and government programs may already compensate for some or all of such external costs. Under varying scenarios, one can argue a remaining premium of nothing to several billion dollars. Third, any public sector efforts to reduce oil imports would only have a marginal effect, which means that possible costs attributed to environment or military expenditures would be little affected by policies to reduce oil imports. Finally, with rising world demand for oil and ultimate rises in the costs of finding and producing new reserves in future years, abundant and low cost Persian Gulf sources are likely to reemerge as a dominant part of world oil supply. Under these future circumstances, the existence of an oil import premium may become a more important policy concern.

Appendix

Table A-1 on the next page summarizes monetary estimates from the studies CRS reviewed in its 1992 report on the external costs of oil used in transportation. The only relevant categories for purposes of this review are the total estimates in the economic category, the military category, and the oil spill category.

In considering these types of estimates, several considerations are necessary. First, monetary estimates of external costs are at best rough approximation for complex non-market values. The studies used to estimate the totals and averages in the 1992 CRS report were done under a variety of circumstances and with a variety of assumptions and methods. Second, some studies estimate incremental damages, some are national or regional in scope, and others are site specific. Consequently, the totals presented in the accompanying table provide only a rough guide based on disparate and not always directly comparable estimates. Third, these estimates are from multiple studies in different years. Because only rough extrapolations are made from these monetary estimates, no attempt was made to translate to current year dollars. Thus, the numbers presented in the table give only a rough indication of where some social benefits might be realized from incremental reduction in imports.

Table A-1. Estimated External Cost of Oil Used in Transport

References

¹The External Costs of Oil Used in Transportation. CRS Report 92-574 ENR. June 17, 1992

² For a comprehensive discussion of the conceptual, analytical, and policy issues and empirical findings on this topic see: Bohi, Douglas R. and Michael A. Toman. The Economics of  Energy Security. Kluwer Academic Publishers. Norwell, Mass. 1996 For perspectives on concerns about U.S. energy security over the longer term see: Energy and National Security in the 21stCentury. Edited by Patrick L. Clawson. National Defense University Press. Washington, D.C. October, 1995

³ lnternational Petroleum Encyclopedia. PennWell Publishing Company. Tulsa, Oklahoma. 1996

⁴ Monthly Energy Renew. Energy Information Administration. U.S. Department of Energy. August, 1997 

⁵ The External Costs of Oil Used in Transportation. Loc. Cit.

⁶ As noted by several analysts, the effects of oil price increases on the balance of payments and depreciation of the dollar are not clear cut. Nominally, an increase in the price of oil means increased total import costs for oil, since oil demand is not price responsive. With an increased supply of dollars in foreign exchange markets, the value of the dollar may depreciate, making U.S. imports more expensive However, a number of analyses reviewed by Bohi and Toman (Op. cit.) suggest that a relationship between oil price increases and external costs due to balance of payments effects is ambiguous at best. Effects on terms of trade will depend on individual circumstances in each country, such as the rigidity of its oil use, and on how oil exporting countries choose to spend their increased revenues (i.e., return capital flows to importing countries such as the United States).

⁷ See for example: Dahi, C. and M. Acaule. Testing Alternative Hypotheses of Oil Producer Behavior. The Energy Journal. Vol. 12, No.4. 1991. pp.117-138; and Mabro, R. OPEC and the Price of Oil. The Energy Journal. Vol.13, No.2. 1992 pp.1-17

⁸ Bohi and toman, Op.cit.

⁹ Ibid

¹⁰ Among recent analyses supporting this perspective, see: Mork, K.A., 0. Olsen, and H.T. Mysen. Macroeconomic Responses to Oil Price Increases and Decreases in Seven OECD Countries. The Energy Journal. Vol.15, No.4. 1994. pp.19-35 Also see U..S. Energy Vulnerability: Macroeconomic Consequences of Oil Supply Shocks. Hill G. Huntington in Energy and National Security in the 21St Century, pp 117-124.

¹¹ Bohi and Toman

¹² The External Costs of Oil Used in Transportation. CRS Report 92-574 ENR. June 17, 1992. pp. 23-if The analysis was based on figures published by the General Accounting Office and by several non-governmental estimates of military costs in the Middle East.

¹³ Ibid., p. 32

¹⁴ Bohi, p.53. Bohi argues that even if the United States completely eliminated oil imports it would plausibly need to devote military resources to defending the market, because a disruption would cause worldwide economic harm.

¹⁵ in his comments on the 1992 CRS report, Bohi (p. A-19) points out that military costs are not true "externalities," but rather the cost of mitigating possible externalities. Equating the externality with the mitigation cost implies that any amount spent in mitigating an external cost is appropriate. The actual externality, he says, is the damage caused by a disruption.