firms do not, however, pay taxes or fees on emissions that regulations still allow, although major point sources are expected to pay approximately $400 million annually in user fees to cover program costs of state operation permits under the Clean Air Act Amendments of 1990. The 1990 amendments also adopted a new acid rain control program that introduces a market-based system for emission allowances to reduce SO2 emissions. An emission allowance is a limited authorization to emit a ton of SO2. Affected electric utilities are allotted tradable allowances based on their past fuel use and statutory limits on emissions. Once the allowances are allotted, the act requires that annual SO2 emissions not exceed the number of allowances held by each utility plant. Firms may trade allowances among each other, bank them for future use, or purchase them through periodic auctions held by the EPA. The market for allowances is structured to encourage firms with relatively low costs of abatement to reduce their emissions and sell surplus allowances to firms that have relatively high costs of abatement.

The 1990 Clean Air Act Amendments strengthened components of the earlier law for mobile sources of pollution. The tailpipe standards for cars, buses, and trucks were tightened, and inspection and maintenance programs were expanded to include more regions with pollution problems and allow for more stringent tests. The amendments also introduced several regulations to reduce air pollution from mobile sources. The act mandated that improved gasoline formulations be sold in some polluted cities to reduce levels of carbon monoxide and ozone. It also provided new programs that set low emission standards for vehicles to encourage the introduction of even cleaner cars and fuels. Despite the progress to date in controlling air pollution from motor vehicles, mobile sources continue to have a significant impact on national air quality. On average nationwide, highway motor vehicles contribute one-quarter of all VOC emissions, almost one-third of NO, emissions, and over 60 percent of CO emissions. A tax related to emissions from mobile sources could provide an additional incentive for consumers to purchase cleaner cars and trucks.

The incremental cost of controlling pollution from stationary sources varies, given the numerous sources. The four options that tax pollution from

stationary sources would base the tax rates on an estimate of the average cost of reducing an additional ton of pollution. Consequently, some firms with low abatement costs might reduce pollution below allowable standards. The option that taxes emissions from mobile sources could also reduce pollution levels. (See REV-38 and REV-40 for other taxes that might reduce emissions of air pollutants.) Reductions in emissions as a result of the taxes would increase welfare if the additional abatement costs were less than or equal to the social benefits. However, accurate estimates of additional social benefits from reducing pollution levels do not exist in many cases. The revenue estimates for the options discussed below all assume that some reduction in emissions occurs as a result.

Tax Emissions of SO, and NO, from Stationary Sources. Imposing taxes of $400 per ton of SO2 emissions and $3,000 per ton of NO, emissions from all stationary sources would raise roughly $28 billion for SO, and $132 billion for NO, from 1997 through 2002. Basing the tax on the terms granted in air pollution permits, which all polluting firms must acquire, would minimize the costs of administration for the Internal Revenue Service. The present monitoring and reporting system for stationary sources that the EPA and state regulators operate could be used to enforce the tax. If the actual emission levels of polluters were lower than permitted levels, polluters could apply for revised permits based on those actual levels. If the tax was based on permitted emission levels, it would be equivalent to the government's selling pollution permits at their fair market price.

The proposed tax on SO2 could reduce pollution below the mandated amounts contained in the 1990 amendments. Some electric utilities and manufacturing plants might switch to coals with lower levels of sulfur because that would be less costly than paying the tax, and others might choose to operate their most heavily emitting plants less frequently or to install new SO2 control devices. The tax system could interact with the tradable allowance system, thereby allowing the government to collect revenues based on emission levels and firms to collect the proceeds from the sale of allowances. (The average sale price of allowances would probably adjust downward in the event of a tax.) The tax on NO, could also reduce emissions below mandated levels contained in the

1990 amendments if some firms adopt currently available abatement techniques whose capitalized costs are lower than the tax they would otherwise pay.

Tax Emissions of PM-10 from Stationary Sources. A tax of $2,000 per ton of particulate matter would raise about $23 billion from 1997 through 2002. Some electric utilities and manufacturing plants might install improved electrostatic precipitators, wet scrubbers, or other equipment that reduces PM-10 emissions to lower their tax burdens. This tax could be administered in the same manner as the taxes on SO, and NOx.

Tax Emissions of VOCs from Stationary Sources. Stationary sources of volatile organic compounds range from huge industrial facilities such as chemical plants, petroleum refineries, and coke ovens to small sources such as bakeries and dry cleaners. Their vast number and diversity make it difficult to estimate emissions and the costs of abatement. A tax of $5,000 per ton on all VOC emissions from stationary sources might promote some abatement and would generate almost $200 billion in revenues from 1997 through 2002.

The advantage of a broad-based tax on VOCs is that it would capture small sources, which the EPA estimates are responsible for approximately 80 percent of all emissions from stationary sources. Because stationary sources emitting less than 2.5 tons of

VOCs per year are not currently subject to federal regulation, a broad-based VOC tax would be administratively harder to carry out than a tax on large sources alone. Assessing the tax on small sources through technology-based estimates of emissions rather than measured emissions would reduce administrative costs but make the incentives less precise.

Tax Emissions of NO, VOCs, and CO from Mobile Sources. A one-time tax imposed on new automobiles and light trucks could be based on grams of NO, VOCs, and CO emitted per mile as estimated under the EPA certification tests for emissions that are required on every new vehicle. The tax could be administered like the "gas guzzler" excise tax. The EPA would determine the tail-pipe emissions for each new model of light-duty vehicles, and the tax would be based on those emission rates. The auto dealer would collect the tax on behalf of the Internal Revenue Service from the vehicle's purchaser.

Such a tax averaging $250 per new vehicle could raise $11.7 billion in revenues from 1997 through 2002. Vehicles made in earlier years have been excluded from the estimate because of the administrative problems of collecting a tax on older vehicles. A disadvantage of excluding them, however, is that vehicles from earlier years represent more than 90 percent of the light-duty vehicles in use and an even greater share of emissions. In addition, the tax would encourage people to delay purchases of new vehicles by raising their price.

T

he nation's spending on military and foreign affairs has fallen dramatically since the Cold War era. As a result of cuts in forces and acquisition programs, spending for defense has been reduced by nearly two-fifths from its recent peak in the mid-1980s. Outlays for defense and international affairs programs account for about one-sixth of the total federal budget in 1996 compared with nearly one-half in 1966 and about one-fourth in the 1980s. Nonetheless, those two functions still represent about one-half of all spending subject to annual appropriations.

Defense cuts were a major element of the deficit reduction packages the Congress passed in 1990 and 1993. By contrast, the budget plan the Congress adopted last year kept defense spending relatively constant while eliminating the deficit by 2002. International affairs, however, was not so favored: its budget for 1996 was cut by 10 percent from the 1995 level.

The National Defense Budget

National defense budget authority more than doubled in the early years of the Reagan Administration-from $144 billion in 1980 to $295 billion in 1985-then was held at roughly its 1985 level through 1990 before beginning to decline. But when inflation is taken into account, a different picture emerges. Since 1985, real (inflation-adjusted) defense budget

authority has dropped sharply, for a cumulative decline of 35 percent (see Figure 3-1).

That decrease was made possible by reductions in the number of military and civilian personnel, the closing of bases, and the cancellation or deferral of many acquisition programs. But savings from eliminating forces have yet to be matched by proportionate decreases in operating costs for the Department of Defense's (DoD's) extensive infrastructure of bases, supply and repair depots, and other facilities.

Some of the options in this chapter would reduce military forces or capabilities in specific areas; others would trim spending for support activities. Although this volume focuses on ways to reduce the federal deficit, the savings from those options could be applied in any number of ways. For example, the savings could fund additional spending for higher-priority military functions without increasing overall budgetary allocations for defense.

Size and Structure of
U.S. Military Forces

One aim of U.S. national security policy is to maintain military forces that are powerful enough to deter potential adversaries from attacking the United States directly or to defeat them, should deterrence fail. The collapse of the Soviet Union and the Warsaw Pact removed the single greatest military threat to the United States and its allies in Europe and the Pacific.

On entering office, the current Administration initiated a broad review--termed the Bottom-Up Review--of the national security situation and U.S. military strategy and forces. That review established goals for major elements of U.S. forces based on the scenario of fighting two major regional conflicts nearly simultaneously. Those goals are 30 percent to 40 percent below force levels of the Cold War era. Reductions of that magnitude have been controversial. Some Members of Congress and certain retired military leaders have argued for keeping more forces than the Administration plans, either because they reject the Administration's postulated two-conflict scenario or because they believe greater numbers of forces would be needed to actually meet that threat.

Other military analysts and policymakers, however, believe that the two-conflict strategy overstates the likely magnitude of security risks the United States will face in coming years. They argue that further reductions in military forces are possible with little risk to national security and that the technical capabilities of current U.S. forces are overwhelmingly superior to those of any likely adversary. Furthermore, in most conceivable situations, U.S. allies would join the military campaign. To accommodate

Figure 3-1.

Budget Authority for National Defense (By fiscal year)

those views, additional reductions in military forces are among the options presented in this chapter.

Strategic Forces

Strategic forces are much reduced from Cold War levels. Since 1990, the United States has nearly halved its force of land-based intercontinental ballistic missiles, reduced the number of bombers committed to strategic missions and taken them off alert status, and reduced the number of submarine-based missiles from 584 to 360 (see Table 3-1). Most strategic analysts believe that those forces still provide a robust deterrent to a direct nuclear attack. All parties have now ratified the first Strategic Arms Reduction Treaty (START I). In 1995, the Congress ratified START II, which would commit the United States and Russia to make even larger reductions in strategic forces. But Russia's parliament has yet to do likewise. Option DEF-01 examines the savings that would result from accelerating planned cuts in U.S. strategic forces, and DEF-02 looks at an early cancellation of D5 missile purchases.

Conventional Forces

In its Bottom-Up Review, the Administration determined the conventional forces it believes the United States would have to deploy to win two nearly simultaneous conflicts. Those forces include 10 active Army divisions supplemented by 15 Army National Guard brigades and other reserve combat and support units. The eight National Guard divisions that represent the largest component of reserve combat units were not allocated a role in meeting the two-conflict threat; instead, they were defined as the nation's strategic reserve. The Navy will retain 11 active carriers plus one reserve carrier for training and local contingencies. And the Air Force will keep 13 active tactical fighter wings, with another seven in the reserve forces. By September 1997, most conventional military forces will have been cut to their target levels (see Table 3-1). Several options would further reduce forces. DEF-17 would eliminate two of the 10 active divisions, and DEF-18 would cut four of the eight National Guard divisions. DEF-06 would