This report is the result of a concerted effort to estimate the full costs of a layered ballistic missile defense and to assess the budgetary and economic implications of building it. Many tens of billions of dollars have already been spent on developing missile defenses, going back to the early years of the Cold War, and much more will be spent. But how much more and what will be the consequences?

Our cost analysis differs in several ways from what others have done with regard to missile defense, and it differs from the way the costs of other military programs have been analyzed. First, we examine all the systems that logically could go into what the Administration refers to as a layered ballistic missile defense. These include systems for which there is only a partial indication of the likely architecture and about which official cost figures or related information is scanty at best.

Second, in making our estimates we found it necessary to modify or enlarge the configurations of some of the systems for which there is some published information. This was done in cases where we concluded that the systems described in official sources could not perform their intended missions. For example, we believe that a robust sea-based mid-course system will require additional dedicated Aegis ships, rather than only those in the existing fleet, and that those drawn from the existing fleet would require extensive modifications. We also conclude that larger, faster interceptor missiles will be required for some systems, and that larger constellations of satellites would be required for the Space Based Laser boost-phase system.

In addition to the costs of acquiring the systems, we also calculate the long-term costs of ownership, sometimes referred to as life cycle costs. This requires estimates of the annual costs to operate and support the systems for their expected lives. The expected life of most systems is 20 years.

It will take many years to build the layered ballistic missile defense envisioned by the Administration. To array the outlays over time, we took as a starting point the assumptions in the January 2002 missile defense report of the Congressional Budget Office. CBO assumed that the acquisition and post deployment operational testing for the ground based and seabased midcourse systems will be completed in 2015, and that the space based laser system will not reach this stage until 2025. In our analysis, we assume that land, sea, and air based systems are completed by 2015, and that the space based laser is completed in 2025. However, we also discuss the possibility that some of the systems will be completed later than 2015 and that there will be a slower rate of increase of expenditures.

Ballistic missile defense systems are categorized according to the phase of the trajectory of offensive missiles that they are intended to intercept. BMD systems are therefore labeled boost phase, mid-course phase, or terminal phase defenses. We examine the midcourse defenses first because they are furthest along in their development and the experiences and costs of the development, testing and acquisition of those systems contain lessons that may have a bearing on what to expect of other types of systems.

Our cost estimates are based on all the systems that logically could be included in a layered BMD, based on easily foreseeable technologies and in accordance with the stated intentions and goals of the administration. We do not predict that each of the systems we examine will actually be built. But we believe either that the administration intends to build them or that it would be necessary to build them in order to improve the chances that the overall program will be effective. Our estimates are intended to show how much it will cost if the systems we examine are, in fact, built.

We have been careful to avoid double counting of assets or equipment that can be used for complementary purposes, or to include in our totals the costs of inherently alternative systems. Thus, we conclude, as have others, that the administration would not build both the ground based and that the stand-alone sea based midcourse systems. But we believe it is likely that both a ground based system and an adjunct sea based system will be built. We also believe that the administration will go forward with some version of each of the boost-phase systems examined in our report.

It needs to be emphasized that we do not conclude that BMD will be effective in fully protecting military assets, let alone urban populations, against a missile attack even if all the systems under consideration are built. An impenetrable shield against weapons of mass destruction, has not been devised despite past and present efforts and doubts remain as to whether it will ever be possible to reach this goal. We are used to solving problems requiring new technology in the military area and t,here have been significant technological gains in the search for an effective missile defense. Unfortunately, the technology of penetrating missile defenses is also moving forward and these advances add to the uncertainties of missile defense. Questions about the effectiveness of the various systems are noted throughout this report. The costs and the accelerated pace of the Ballistic Missile Defense program should be viewed from that perspective.

The full costs of a layered missile defense, as should be expected, are enormous. BMD will be perhaps the first trillion dollar defense program. Given the analytical approach we employ, one might ask whether other major weapons system would prove to be as expensive if their costs were calculated in the same way, including 20 years of operations and support costs. It is our view that other weapons programs should also be examined in the context of their full costs including the long term costs of ownership. Government officials and the general public should have this kind of information. Indeed, objective estimates of the full costs need to be known to make intelligent decisions about any military program.

But the full costs of individual major weapons systems are not comparable to a layered ballistic missile defense. The reason is that BMD is not a single weapons system or even a new generation of a class of weapons systems. BMD is a vast complex of weapons systems designed to have global reach, similar to major components of the structure of forces such as army ground forces, naval warships, and strategic offensive forces. It includes large numbers of missiles, ships, aircraft, space satellites, radar and other monitoring, detection and interception activities. It is aptly called a layered program because, in addition to the complexity of its configuration, it represents a new layer in the structure of forces.

The extraordinary costs raise questions about the consequences for the Federal budget and the economy not usually associated with individual military programs. The issues are not just about affordability but are also about serious tradeoffs against other crucial priorities. Depending on how spending for BMD is financed, significant portions of other defense spending could be displaced over the next 20 years, or non-defense discretionary programs could face significant reductions. In the case of the more vulnerable domestic programs, the effects could be very serious. It would be possible to finance BMD through tax increases, but this seems unlikely under the present Administration. If this remains so, and non-defense programs are not substantially reduced, missile defense will be financed through borrowing. Our report shows that the effects of borrowing on capital markets could have widespread negative effects on the economy.

It could be argued that certain economic benefits would result from BMD spending by way of increased employment and income. Past experience with spending for weapons systems shows that such gains are modest and distributed among a narrow range of interests. Finally, the report examines some of the possible alternative uses of future missile defense expenditures. We find that if some of the planned expenditures were reallocated elsewhere it would be possible to more effectively address some of the same threats.