234. Memorandum From the Special Advisor to the President and the Secretary of State on Arms Control Matters (Nitze) to Secretary of State Shultz1
SUBJECT
- Projected Effectiveness of SDI
On May 21, I spoke confidentially with Bob Perret, a program leader in the Planning and Evaluation Division at Livermore Lab, about a study he is conducting on the projected effectiveness of SDI. Max was also present at the meeting.
[Page 863]The study examines the projected near-term capability and cost ratios of strategic defenses against current and projected Soviet threats. It concludes that Soviet force reconfiguration, based on current Soviet technical capability, may well outpace near-term U.S. defense deployments, preventing a capable defense from being established. It concludes also that the total cost of defense exceeds the cost of upgrading the offense, modestly by 1994, but by a factor of two by 1996, and by a factor of three by 1998. Marginal cost ratios favor a modernized offense, including countermeasures, by an even greater factor, since the defense is found to be ineffective against the final assumed threat.
The study assumes an SDIO candidate architecture for near-term deployment, including SBKKVs with Boost Surveillance and Tracking System (BSTS) sensors, and ERIS and HEDI ground-based interceptors. First generation deployments would occur from 1993–96, second generation from 1996–98. The penetration of current Soviet forces, assuming no modernization, against these U.S. defenses would decrease from a 100% leakage rate in 1992 to 40% in 1994 to 5% in 1996.
Based on intelligence estimates, the study postulates five types of near-term technology responses, which are within the current technical capability of the Soviet Union: simultaneous deployment of warheads and decoys from the post-boost vehicle (PBV), as opposed to currently-employed sequential deployment of RVs from the buses during an extended post-boost phase; fast burn boosters; ASAT weapons; proliferation of warheads and ICBMs; and other countermeasures to confuse U.S. space-based sensors.
All of these countermeasures are judged to be “likely” responses during the early 1990s, perhaps the most effective of which would be fast-burn boosters and simultaneous release PBVs. These countermeasures would reduce significantly the time of the boost and post-boost phases of the missile’s trajectory. During these phases, offensive targets are more vulnerable and more valuable, since these missiles and PBVs carry many RVs.
These countermeasures would be particularly effective against KKVs. First, a more significant midcourse threat, consisting of proliferated warheads and decoys, would be deployed earlier in the missile’s flight. Enhancing the offensive threat would require a corresponding increase in available defenses. The key question here, therefore, is whether the defense meets our cost-effectiveness criterion. The study concludes that the first two generations of defenses would not. Second, reducing the duration of the boost and post-boost phases reduces the fraction of the defenses that can engage targets in these phases of their flight trajectory. Consequently, the Soviets would be able to attack these fewer defensive systems at less cost and perhaps with greater certainty.
The study concludes that, when anticipated Soviet countermeasures are factored in, the expected penetration of Soviet warheads will [Page 864] decrease from 100% in 1992 to 80% in 1994, where it will remain about constant through 1996 and finally increase again to 90% by 1998.
I believe that this study presents a credible, although perhaps not conclusive, case that there remain significant uncertainties whether the early generations of defensive deployments would achieve U.S. strategic objectives. This uncertainty implies that an “early deployment” decision would be imprudent. First and second generation architectures appear to be unable to meet our deployment criteria if the offense is modernized with advanced countermeasures. This implies that we must maintain a broad technology program in SDI and arrest the current tendency to reallocate funding from longer-term technologies to the near-term technologies.
I have learned that, on June 30, the Defense Acquisition Board is due to make a “DAB Milestone 1” decision, which would establish the case that defenses have military utility and that current technology is sufficiently mature to meet projected military requirements.
Several forces are at work here. Those who are committed to early deployment see this as a first step in the acquisition process. Others are pushing this decision as a vehicle to force the Administration to clarify once and for all how and whether ABM Treaty limits will apply to SDI in the future. Finally, General Abrahamson hopes to win the support of the JCS for a positive DAB Milestone 1 decision, and thus establish a firm JCS commitment to SDI acquisition and deployment.
- Source: Department of State, Ambassador Nitze’s Personal Files 1953, 1972–1989, Lot 90D397, 1987. Secret; Sensitive.↩