6 07 October 2, 1987 STRATEGIC DEFENSE HOW MUCH WILL IT REALLY COST RODUCIION Opponents of rovidine the United States with a defense against a Soviet missile attack have 1 een trymg to sink Ronald Reagan's Strategic Defense Initiative SDI, or as it is popul a rly known, Star Wars--under the weight of a putative $1 trillion price tag. Some critics warn that SDI actually could cost two and one-half times that. These claims, almost entirely unsubstantiated, have made the cost issue a major stumbling block to deve lopment of SDI.

In truth, however, SDI will cost American taxpayers just a fraction of these deliberately inflated estimates. While it is unlikely that SDI will be as cheap as the 40 billion claimed by some SDI backers, the price tag probably will be in th e range of $115 billion to $120 billion spread out over ten years.

Some variance in defense program estimates is normal because of the differences in researcher expertise, disagreements over details, or simple miscalculations. In the case of SDI, the huge estimate gap exists because there are fundamental differences in the assumptions, the biases, and very important, the motives of those preparing the estimates there must be a common understanding of the mission and parameters of the program. This is miss i ng from the debate. Neither the Reagan Administration nor Congress has focused on a specific SDI proposal. There is no widespread agreement on SDI's mission or on which technologies should be pursued to achieve this mission. In the absence of specific pro g ram outlines or technical proposals, SDI critics are free to base cost .estimates on the most expensive, and in many cases least likely, strategic defense scenarios. As a result, their estimates are extremely excessive No Agreement on Won. Before SDI's co s t can be accurately estimated r With the research conducted so far and the technological breakthroughs coming more rapidly than expected, however, enough is known about the capabilities of -2 strategic defense technologies to develop reasonably reliable c o st estimates for a near-term strategic defense. Such a system would employ 1) a space-based, kinetiekill vehicle system costing about $50 billion 2) a grouud-based guided missile component costing around $22.5 billion 3) a gded missile terminal defense co s ting about $13.5 billion; and 4) the sensors and radar needed for this three-tiered defense, costing about 32.5 billion The price of a longer-term defense is more difficult to estimate because research is only in the initial stages for many of the most po t entially useful technologies and costs may either increase or decrease depending upon future developments. Technolo cal breakthroughs and cost-saving modifications, such as both short- and long-term technologies. In any case, even the most careful estimat e will be approximate.

Enough is known, however, to conclude that the estimates by SDI opponents are very excessive. The way to settle the cost issue is for Reagan and the Pentagon to propose a specific near-term anti-ballistic missile program. Only when c rucial questions are answered regarding SDI's mission and structure can final cost calculations be made further miniaturization o P components, may reduce anticipated costs even more for THE SDI MISSION Assured Destruction doctrine, which for several deca des has based deterrence on the threat of massive retaliation and subsequent destruction of U.S. and Soviet societies.

From the outset, the mission of SDI has been to redefine deterrence by giving the U.S. the means to protect its population and assets fro m Soviet 'attack The Strategic Defense Initiative was devised as an alternative to the Mutual While the ultimate objective is a defensive system that can protect the U.S population directly, the more immediate goal, according to the PentagonkStrategic Def e nse Initiative Organization's (SDIO Chief Scientist, Allan Mense, is "to devalue Soviet offensive ballistic missiles in the mind of the Soviet offensive mission planners."l A strate 'c defense that is 90 percent effective, a level of effectiveness most SD I scientists P eel is attainable, will certainly cause this devaluation Deterring a First Strike. Even if the feasibility of a layered strategic defense of over 90 percent effectiveness can be documented, the value of such a system is doubted by those who a rgue that a level of protection less than 100 percent is inadequate and undesirable. An SDI system capable of destroying 90 percent of all Soviet intercontinetal ballistic missiles (ICBMs) that are launched, while not 1. Dr. Allan Mense, Acting Chief Scie n tist, Strategic Defense Initiative Organization, briefing on March 20, 1987. -3 leakproof would render the Soviet ICBMs "impotent and obsolete If Soviet war planners know that 90 percent of the Soviet missile force is going to be destroyed 111 flight, the i r confidence in the success of a Soviet first strike is substantially diminished. The Soviet ICBM force will have been so devalued that it is no longer able to accomplish its mission. Realizing this, the Soviets will not launch their ICBMs. Thus, a defens e may be much less than perfect and still provide total population protection by deterring a Soviet nuclear offensive.

SDI also could provide an equally valuable and unprecedented secondary protection. If the Soviets do launch a first strike, or if a limit ed number of ICBMs are launched accidentally or by an unstable smaller nuclear power, SDI could destroy almost all of these incoming missiles. At present the U.S. has no such capability. Any missile launched at the U.S. will hit it. The U.S. today is-nake d to any missile attack.

The SDI mission is deterrence through defense of populations and strategic forces. Perfect technology and 100 percent effectiveness are not required to accomplish this mission. Estimates of SDI's cost predicated on the requirement of constructing a "leakproof umbrella" add immensely and unnecessarily to the cost of strategic defense.

Many differing technologies have the potential to fulfill SDI's mission I requirements. Some, such as lasers and particle beams, show great promise, but also require ten to fifteen years of research and development before deployment.

Other technologies will be available in the next five years for construction and deployment of an effective SDI system at a reasonable cost. The mgnstays of such a near-te rm system will not be lasers or particle beams, but kinetic energy weapons KEW space and ground-launched anti-missile rockets.

A layered KEW strategic defense could destroy up to 90 percent of the warheads from a massive Soviet ICBM attack. This near-term system could be upgraded as new technology matured system as soon as ossible to offset the immense Soviet strategic arsenal and The U.S. should deploy a near-term KEW increasing Soviet e E orts in ballistic missile defense?

Kinetic energy wea pons are the most promising systems for the immediate future because the technologies they draw on are the most mature. Attempts at cost estimates thus should focus on near-term technologies, the kinetic energy weapons and the associated systems required t o weave them into an effective defensive system. Yet SDI 'critics largely ignore these promising and relatively inexpensive technologies. Instead they base their estimates on far more expensive exobc proposals such as advanced laser technologies, which wi l l not be available for years 2. The CIA estimates that the USSR has spent $150 billion on strategic defense (fifteen times the U.S. expenditures) over the last ten years. Shvrtegic Defense, December 4, 1986, p. 1 4 In addition to failing to accept SDI's d e terrent mission and ignoring relatively inexpensive near-term technologies, many of the excessive cost estimates rely on imprecise methodology. Typical of this is developing estimates based on the relationship between past research programs and the system s that developed from them. This type of estimate can produce only the vaguest picture of a new system's cost. In the case of SDI, much of the hardware for a near-term kinetic-kill system either already exists, or is so similar to other current systems'tha t relatively reliable specific cost estimates can be made. Wherever possible, cost estimates should be based on careful analysis of the specific pieces of hardware needed to carry out the assigned mission.

One of these generic estimates, produced by the Co uncil on Economic Priorities,'places the cost of 'SDI at from $400 billion to $800 billion.3 The figures were reached by merely multiplying SDI's research and development budget by an arbitrary percentage that was selected by comparing the relationships b e tween research and development costs and final prices of past defense systems. The development programs, which are focused and directed at one final outcome, SDI research and development programs are exploratory and encompass a, broad range of very differ e nt technologies. Such an exploratory research program is inevitably more expensive than a more directed effort. A study by Barry Blechman and Victor Utgoff calculates the cost of a new-term, three-tiered SDI system at*$630 billion to 770 billi~n.~ This co s t estimate of a "comprehensive" defense system charges SDI with the cost of "upgrading" the space shuttle ($33-$47 billion) and adds $159-$209 billion in "operation costs In addition to these inflationary factors, Blechman and eatly exaggerate the number o f defensive missiles and satellites needed to utgoff accomp P ish SDI's mission Primitive, Arbitrary Formula. The origin of the most often quoted estimate of SDI cost, $1 trillion, is hard to trace. This widely circulated figure seem to have originated wi t h former Secretaries of Defense Harold Brown and James Schlesinger both SDI opponents.5 Though Schlesinger will not admit any responsibility for the $1 trillion figure, Brown says he arrived at that mark by using a "rule of thumb that involves multiplying the research and development expenses of the. program by ten to determine its final cost of this arbitrary formula is further guaranteed in this case by the unique nature inaccura% o the SDI research and development phase. Unlike most research and Such a p rimitive and arbitrary formula cannot produce an accurate or reliable estimate. Brown erred twice more: he repeated the Council on Economic Priorities' error by failing to consider the unique nature of the SDI research and 3. Hartung and Nmroody What Pric e Strategic Defense Council on Economic Priorities newsletter January 1985 4. Barry Blechman and Victor Utgoff, "The Macroeconomics of Strategic Defense International Securily, Vol. 11, No. 3 (winter 1986-83, p. 49 5. John Collins, Senior Defense Specialis t , Library of Congress, testimony before the Senate Armed Services Committee, January 20, 1987. -5 development program; and he started with an incorrect figure of $100 billion for SDI research and development costs. Had Brown plugged the correct numbers in to his formula he would have arrived at a total SDI cost of $260 billion.6 Brown moreover,.did not indicate how many years the program would run.

Had Brown focused his estimate on near-term systems, instead of on long range high-tech laser and particle bea m-based alternatives, he would have come up with a number somewhere in the neighborhood of $130 billion--a fairly reasonable cost. But he did not As result, the fictitious $1 trillion estimate continues to be used by scientists, commentators, and journali s ts unaware of its shaky foundation. In the hands of budget-conscious Wisconsin Senator William Proxmire, the Brown Schlesinger figure soars to "several trillion dollar It is impossible to fix SDI costs over the next decade with precision, just as it rogra m s as food stamps and welfare. Yet SDI research has progressed far enough is impossible to fix with precision the decade-long cost of such other federal for competent scientists to make intelligent cost estimates. These more realistic estimates take into a c count SDI's near-term requirements and properly assess the costs of available technology In 1982, High Frontier, an organization that. studies near-term strategic defense conducted an in-de th study that estimated the total cost of a near-term. strategic d efense system at P 40 billion.8 High Frontier's study based its conclusions on the assumption of a very modest mission for SDI and the system they evaluated relied too heavily on using cheaper "off-the-shelf' technology Though High Frontier's resultant es t imate is thought to be somewhat low by most experts, it was determined by a fundamentally sound method. High Frontier analyzed SDI's near-term requirements and suggested particular types of hardware to meet them. It estimated the cost of each component an d the number of each the system would require.

Several careful studies since 1982 have exceeded High Frontier's original estimate but most have used the same sound methodology High -1 of Protection. Estimates provided by Lt. Col. Simon P. Worden Senior Pol icy Analyst at the White House Office of Science and Technology&-and former assistant to the director of the Strategic Defense Initiative Organization, put the cost of a near-term strategic defense system at "less, perhaps much less, than 100 billion.lg A c cording to Worden, the space-based missiles and ground-based 6. According to the Department of Defense, SDI's total research and development costs are only about 26 billion--half of which has been devoted to near-term technologies 7. Senator William Proxm ire When you talk about 'star wars' you're talking trillions The Chtistiun Science Monitor, May 19, 1986, p. 16 8. Hi

Frontier: A New Nutionul Smegy (Washington, D.C High Frontier, Inc 1982 9. Simon Worden SDI: What Can We Do? When Can We Do It Nutionul R eview, December 31 1986, pp. 36-40. -6 sovereign-area-defense- interceptors that could be purchased and deployed for that amount would provide levels of protection high enough to make a successful Soviet first strike "impossible thereby deterring any atta c k. This type of system, explains Worden, could be available by the mid-1990s and could be updated to a "multi layered system" capable of "making obsolete the Soviets' trillion dollar offensive investment 10 A detailed study conducted by a panel of strateg ic defense experts at the George C. Marshall Institute comes to much the same conclusion, placing the end cost of a near-term, layered, kinetic-kill strategic defense system at $121 billion.

The Marshall Institute's study focuses on the most promising near -term SDI architecture: a three-tier system with space- and land-based components, capable of destroying Soviet ICBMs at various stages in their flights.ll THE OF A NEAR-TERM m3EGIC DEFENSE The most promising near-term system would employ layers consistin g of space-based kinetic-kill vehicles (SBKKVs) targeted at the incoming missile's boost phase and post-boost phase a ground-based component similar to Lockheed's Exoatmospheric Reentry Interceptor System (ERIS) to .shoot down missiles in the mid-course of their trajectories; and a terminal defense, erhaps similar to McDonnell Douglas' High Endoatmospheric Interceptor &EDI) to destroy those few missiles that get through the other two layers. A strategic defense system of this configuration could achieve eff e ctiveness levels higher than 90 percent' and: would cost about $118.5 billion in 1987 dollars, excluding operation and maintenance costs.12 Studies by the Department of Defense, relying on information obtained from contractors, indicate that space-based k i netic-kill vehicle interceptors would cost about $1.5 million each. Each SBKKV would weigh about 500 pounds and cost about $750,000 to launch. Another $2.25 million must be added to the cost of each interceptor for building and launchin8 the satellite tha t would cany the interceptor missiles. Defense Department data mdicate that 11,OOO SBKKVs would be needed to insure an adequate anti-ICBM force. The total cost for 11,000 SBKKVs,,and their launching platforms would be around $50 billion 10. lbid 11. De loy m ent of Missile Defense in the 19903 (Washington, D.C The George C. Marsh4 Institute 12. This estimate and the supporting figures in subsequent paragraphs were derived from discussions with officials at the Department of Defense, the Strategic Defense Init i ative Organization, Los Alamos National Laboratory, Lawrence Livermore Laboratory, Lockheed Aerospace Corporation, Rockwell International Corporation, McDonnell Douglas Cor ration, High Frontier, Inc. Many of the figures defense field to be the most caref u l and most accurate unclassified study to date Decem IE r 1986 used also appear in the Marshall Institute study, w E ch is considered by many experts in the strategic -7 ERIS Ekoatmospheric Reentry Interceptor Systems would cost about $1.3 million each an d about the same amount to launch. Again, about 10,000 would be needed. The total cost of these missiles and their launchers would be approximately 22.5 billion HEDI High Endoatmospheric Interceptor-type systems, mainly because of the increased speed neede d to counter ICBMs in the terminal stage of their flights would be larger and more expensive, at about 3 million per interceptor. HEDI launchers would run about 1.5 million each. Researchers estimate that about 3,000 HEDI-type interceptors would be needed to provide an adequate terminal' defense at a total cost of about 13.5 billion.

Added to the costs of the interceptors themselves must be the cost of the sensors needed for tracking and targeting Soviet ICBMs. Each of the three layers requires a different type of sensor to aim and guide its projectiles toward incoming targets. For the first layer, ten low-earth-orbit sensor satellites and four geosynchronous satellites (that maintain a position high in orbit over a particular point on the Earth's surface) w ould be needed to provide complete coverage of potential Soviet launch areas and missile paths. At $1 billion and $2 billion respectively, the total cost for the SBKKV layer's sensors would-be about $18 billion. The ERIS layer would require twenty airborn e optical system (AOS) sensors at about $500 million each for a total of $10 billion. And the HEDI layer would need about thirty ground-based radars at about $150 million each for a total of $4.5 billion. The total costs for the sensors and radars needed b y such a three-level defense would be about $32.5 billion. This estimate does not include operation and maintenance costs. In addition to being irrelevant to the costs of building SDI these costs are more difficult to figure accurately because of their hyp othetical nature.

The costs of battle management computers and command, control communications, and intelligence systems have not yet been studied adequately.

Further work is needed in this area. Also, many of the components. of the. space based tier of t his system likely will require greater space-lift capacities than NASA currently has to put them into orbit. Even when the shuttle program is revived, a larger orbiter, or perhaps a return to large expendable boosters, will be needed to launch most of SDI ' s space-born components. NASA is currently working on, this problem and the cost of the new heavy payload booster will be borne mostly!by either the Air Force or NASA as part of the space station and other space projects kerall Sh0rt-T- Ikplm cost The tot a l cost to deploy the three-layered SBKKV-ERIS-HEDI system would be around $118.5 billion. Though a large amount of money, it is reasonable when spread out over a ten-year period. At an average yearly expenditure of $11.85 -8 billion, this would amount to l ess than .01185 percent of the yearly federal budget and only .Om8 percent of the annual GNP Given the cost of offensive systems, such as the B-1 bomber at $27.3 billion the MX ICBM at $22.3 billion, and battle-ready nuclear aircraft caqiers at $6 billion each, a ten-year $120 billioa expenditure on a strategic defense system, which would actually defend the United States, should not be regarded as too expensive.n Future systems will be based largely on the emerging technology of lasers particle beams, or o ther advanced systems. Their costs thus are impossible to estimate accurately. Among the most promising and most talked about is the space based laser. Estimated by SDI opponents to cost "trillions reliable estimates. on the cost of developing and deployi n g an effective space-based, anti-ballistic- missile laser place it as low as $200 million. Though this estimate by the Department of Defense and scientists at Los Alamos National Laboratory in New Mexico is the best now available, these experts acknowledg e that costs are likely to fluctuate greatly in the years before deployment of such a system becomes p0ssib1e.l The Strategic Defense 'Initiative Organization is working to minimize the cost of research and development by issuing competitive contracts. SDI planners are considering such cost-saving measures and concepts as the use of available, proved technology for the first phased deployments of the system and utilization of.smal1 combat cells" rather than a large centralized command, control, and communic ation apparatus. Unexpected cost-saving technological advances may result from current SDI research.

An effective SDI system will eventually decrease U.S. offensive strategic expenditures by reducing reliance on offensive systems. Furthermore, research con ducted for SDI will undoubtedly yield computer, sensor, and rocket technology useful in conventional and strategic military programs as well as in the space program and other industries.fi United States would be incalculably higher than the $100 billion t o $15.0. billion it would take to buy an effective strategic defense system. The cost of the dimage caused by even a couple of accidentally launched intercontinental ballistic missiles or submarine-launched ballistic missiles dwarfs any reasonable estimate of SDI's cost Finally, the cost in lives, land, and property of a Soviet nuclear attack on the 13. Based on statistics from the Air Force and Navy 14. Telephone conversation with Greg Canovan, Scientist, Los Alamos National Laboratory, May 1987 15. Lt. Ge n . James Abrahamson, Director of the Strategic Defense Initiative Organization, Boston Globe, January 14, 1987, p. 17, and Charles Bridge, Chief Scientist, Litton Industries, Inc, Fort Worth Star Telegram, January 14, 1987, p. 8B. -9 Until the Reagan Admin i stration and Congress 'agree on the SDI mission and settle on a specific SDI system, the cost of an effective strategic defense cannot be calculated with precision. Realistic estimates by experts, however, indicate that SDI's cost will be well within reas o n--far below the $1 trillion estimated by many of those determined to .stop SDI. The costs, moreover, become even more reasonable when spread over ten years. The several hundred billion dollars that it may take to deploy strategic defense in the next ten y ears pales when compared to the over 700 billion that the federal government will spend on Medicare and the $260 billion on farm subsidies over the next decade (and these figures even assume that spending on these programs will remain at current levels Th e potential cost of SDI is a legitimate national concern, but the lkost issue has become just another tactic by SDI opponents to derail the project. This focus on SDI's cost will continue to linger as long as SDI remains a vague1 defined issue by developin g a specific proposal for a near-term anti-ballistic missile program.

The Soviet ICBM threat must be clearly defined, SDI's mission must be precisely delineated, and the structure and hardware proposed must be appropriate to both. research project. The Pre sident and the Pentagon must address this L ndamental When the crucial questions regarding the SDI mission and structure are answered, accurate cost calculations can be made and SDI can be debated and judged on its merits. No doubt an effective defense ag ainst Soviet offensive ICBMs will be expensive. But when the benefits of SDI in terms of enhanced national security and increased U.S. foreign policy flexibility are added to the equation, SDI's benefits will outweigh its costs Grant Loebs Policy Analyst