At the start of the new millennium, the forces of electoral politics are moving the nation toward a decision to deploy a national missile defense, or NMD. The decision is scheduled for the summer of 2000--a few months before a presidential election. The Republicans have succeeded in painting opposition to missile defense as synonymous with being soft on defense. Sensitive to this charge, the politically minded Clinton administration--despite some recent waffling--probably will decide to deploy the system. Thus, the debate should not be about whether to deploy a missile-defense system but about what kind of a system to deploy.
Although the Clinton administration has had to be dragged to the brink of supporting NMD, the administration correctly favors developing a limited land-based missile-defense system. Zealous proponents of NMD have oversold missile defense, have oversold sea-based missile defense and have a hidden agenda.
The zealots consistently have overstated the importance of missile defense in protecting the nation against attacks using weapons of mass destruction. They ignore that NMD may not kill all ballistic missiles fired at the Untied States and offers little protection against the more likely threats of shorter-range ballistic or cruise missiles fired from a commercial ship off the U.S. coast or a smuggled device delivered by a truck or by a freighter into a U.S. port. Because the United States has infrared satellites that can pinpoint the location of a ballistic missile launched from the territory of a rogue nation, any nation wishing to attack the United States and not be counterattacked by the most potent nuclear arsenal on the planet would probably choose one of those alternative methods of delivery. If any of those alternatives were used, the United States would have more difficulty determining the origin of an attack and retaliating. However, the fact that the U.S. government has difficulty defending against all threats to the nation from weapons of mass destruction does not mean that it should not do what it can to defend against some of the threats.
But because NMD is a backup system (the robust U.S. offensive nuclear arsenal still should deter most nations from attacking the United States directly or indirectly) arrayed against a threat of low probability, the funds devoted to building it should be commensurate with its purpose. A limited land-based system satisfies the criterion that any system should be cost-effective. The sea-based system--either standing alone or supplementing a land-based system--most likely would not meet that test, would destroy the antiballistic-missile treaty outright and would poison relations with Russia and China.
The administration estimates that an NMD system of 20 land-based interceptors would cost $13 billion to build. Some early supporters of the sea-based system claimed that the system could be built for as little as $2 billion to $4 billion. But in 1998 the president of Heritage Foundation one of the primary advocates of the sea-based approach--came in with a much higher estimate--$1.5 billion to $2 billion a year for five years for a total of $7.5 billion to $10 billion. And in early 1999, a study by the Pentagon came up with an even higher estimate--$16 billion to $19 billion.
Many proponents of the sea-based system maintain that Navy air defense ships with Aegis radars could be modified quickly and cheaply to knock down incoming ballistic missiles instead of aircraft. However, the Aegis radar most likely would need to be replaced with an expensive new radar and the ships standard missiles would need to be modified for the new mission. Both improvements would be costly. In addition, the Navy certainly would ask for more ships to carry out the added mission, and Aegis ships are very expense to build and operate.
Cost is an important factor in deciding which NMD system to buy, but it is not the most important factor. In reality, Pentagon estimates for land- and sea-based missile defense probably understate the cost of both systems.
A major problem with the sea-based approach--and the hidden reason its proponents are pushing it--is that it will destroy the ABM treaty outright. The ABM treaty prohibits sea-based and space-based ABM systems but allows land-based systems with severe restrictions. The treaty--amended in 1974--permits each country only one land-based ABM site that protects only a limited area (for example, a capital city or offensive missile field) rather than the whole nation. To build a system with ground-based interceptors that protect all 50 states would require renegotiating the treaty but would not destroy it.
A sea-based system using Aegis ships deployed along the coasts of the United States would have difficulty providing protection to all 50 states--especially those in the midsection of the country. In addition, sufficient warning of an imminent missile attack would be needed to allow a ship to deploy within a few hundred kilometers of the area to be protected. If instead the Aegis ships were deployed overseas, sufficient warning of an attack also would be needed so that a ship could deploy to within a few hundred kilometers of the location of the launch. If the ship was off the coast of North Korea and the missile was launched from Iran, for example, no defense could be provided. The only way to protect the entire nation using a sea-based approach also would be to put missile-defense weapons in space (or use the cost-prohibitive approach of maintaining the modified Aegis ships on constant patrol in all potential launch areas). Such a deployment drastically would increase the cost of NMD.
Thus, missile-defense enthusiasts really are using the sea-based system as a Trojan horse to bust the ABM treaty and put weapons into space. Those zealots may be trying to bring back the grandiose Star Wars program designed to neutralize the Russian strategic arsenal or at least to resurrect the more modest Global Protection Against Limited Strikes, or GPALS, system of the Bush administration. Both of those expensive systems included space-based interceptors and really were international rather than national missile-defense systems because they were designed to protect clients and allies as well as the United States (a sea-based system standing alone also fits this category). If the goal is--as it should be--to provide a backup system in case the potent U.S. nuclear deterrent fails to dissuade a radical or crazy rogue state leader from taking the unlikely step of launching a missile from his home territory at the United States, such expensive systems are not cost-effective against such a narrow threat. The Congressional Budget Office estimated that a GPALS-like system that included 300 ground-based interceptors, 500 space-based interceptors and 20 space-based lasers would cost $60 billion to acquire ($140 billion when the costs to operate and support the system are included).
Of course, it is doubtful whether the zealots agree with the legitimate limited goals of the Clinton administrations NMD program. Their hidden agenda seems to be protecting clients and allies or defending the United States against larger missile strikes (read a Russian attack). Spending the dollars of U.S. taxpayers to defend wealthy clients and allies against a missile attack is questionable.
Russias greatest fear is a much grander U.S. missile-defense system that includes space-based weapons and undermines the ability of the Russian nuclear arsenal to ride out a surprise American attack and have enough warheads left to inflict unacceptable damage on the United States. Although the Russians are strapped for cash, large U.S. defenses could cause them to take desperate actions that would destabilize the nuclear balance--retaining missiles with multiple warheads that they are supposed to eliminate under the as-yet-unratified START II treaty or converting their new single-warhead Topol-M missile into one with multiple warheads. (Missiles with multiple warheads are destabilizing because an adversary might be tempted to launch a preemptive attack to wipe out several warheads in each silo or to launch its weapons on warning of a preemptive attack on its own missiles with multiple warheads.)
Smaller U.S. defenses do not pose as big a threat to the Russian nuclear arsenal and would be less likely to engender those responses. But the Russians believe that any smaller U.S. defense against rogue states could be a stepping stone to a larger defense that would threaten Russia. That concern is an important reason why the United States should work to negotiate with the Russians to amend the ABM treaty rather than simply abrogate it outright. Although the Russians currently oppose U.S. initiatives to amend the treaty, the Bush administration made substantial progress in renegotiating the treaty to allow the GPALS system, which was more ambitious than the Clinton administration program. U.S. negotiating leverage also would be enhanced by stronger statements that the United States is prepared to abrogate the treaty unilaterally if the negotiations fail. (In addition, the United States could offer deep cuts in offensive warheads--below START III levels [2,000-2,500 warheads]--to lessen the threat to Russias decaying nuclear arsenal.) When facing the choice between an unrestrained U.S. defense program and a restrained one under the treaty, the financially strapped Russians probably will see the light.
Given the needless kicks in the pants that the United States recently has delivered to Russia (NATO expansion and a war against Russias Serbian ally) and potential Russian retaliation if the United States abrogates the ABM treaty (stopping cooperation on securing Russian weapons of mass destruction or actually selling such weapons to rogue states or terrorists), unilaterally abrogating the treaty before attempting to renegotiate it is a bad idea. Building a sea-based NMD essentially would require abrogation of the treaty merely to obtain a system that is likely to be more costly and less effective in protecting the United States than a land-based system. Renegotiating the ABM treaty to permit the deployment of a limited land-based system is a better alternative.