Dennis M. Gormley and K. Scott McMahon
*This chapter is adapted fromthe authors' monograph Controlling the Spread of Land-Attack Cruise Missiles (Marina del Rey, Calif.: American Institute for Strategic Cooperation, January 1995). It has been modified and updated in light of current developments.
Although deploying national missile defenses may not enjoy a political consensus in the US Congress, robust theater missile defenses find strong bipartisan support. Propelled by Saddam Hussein’s use of modified Scuds during the Gulf War of 1991, America has earmarked most of its investment in theater missile defenses for defeating currently deployed and future ballistic missiles. But ballistic missiles are not the only means by which rogue states could threaten the prompt and secure projection of US forces in regional contingencies. Cruise missiles, particularly those capable of land-attack roles, show signs of quickly becoming at least as threatening as ballistic missiles.
According the Central Intelligence Agency, at least a dozen countries now have land-attack cruise missiles under development.1 Several appear willing to export complete systems, including ones with low-observable features, as well as component technologies and development expertise. Moreover, the widespread availability of cheap guidance, navigation, and digital-mapping technologies throughout the developing world augurs the conversion of widely proliferated antiship cruise missiles (ASCM) and unmanned aerial vehicles (UAV, including remotely piloted vehicles [RPV]) to land-attack roles. Today more than 70 countries possess ASCMs, while UAVs for various and sundry missions are witnessing explosive growth.2
The quickness with which the cruise missile threat will emerge has serious implications for Western security planning. According to the 1994 Defense Science Board Summer Study on Cruise Missile Defense, the United States “is not in good shape” to defend against low-flying cruise missiles with small radar cross sections (RCS)—despite huge investments in conventional air defenses.3 Even so-called poor man’s cruise missiles, with large RCSs,4 could greatly complicate existing air defenses. Consider, for example, how Iraq’s use of crudely manufactured cruise missiles might have increased friendly-fire casualties during the Gulf War. As evidenced by the inadvertent shootdown in 1994 of two US Army Black Hawk helicopters over northern Iraq by friendly aircraft, the military services have yet to solve the problem of friendly-fire casualties. Facing both cruise and ballistic missiles in a far more complicated wartime setting, coalition defenses would have been acutely tested by the need to distinguish friendly aircraft from Iraqi cruise missiles.5 Should such land-attack cruise missiles emerge as quickly as the aforementioned study by the Defense Science Board suggests (within five to 10 years), America’s capacity to deter and defend against threats to its regional interests could be severely challenged.
This chapter examines prospects for the proliferation of cruise missiles and considers their implications for the formulation of export policy. It first addresses those factors that condition the pace and scope of this new proliferation challenge—namely, adversary motivations to acquire cruise missiles and routes to (and challenges associated with) acquiring complete systems or the necessary enabling technologies. We then turn to analyzing the Missile Technology Control Regime (MTCR), the international community’s principal export-control mechanism for slowing the spread of both ballistic and cruise missiles. The chapter concludes with a set of recommendations bearing on both general policy matters as well as specific measures to slow the spread of land-attack cruise missiles.
To appreciate fully just how existing export controls affect the prospects for cruise missile proliferation, one first must distinguish the differences between ballistic and cruise missile systems—particularly the close relationship between unmanned cruise missiles and manned aircraft. Unlike ballistic missiles, cruise missiles fly through the air in powered flight for the duration of their trip. They fall into the category of aerodynamic missiles. Ballistic missiles, by contrast, shed their rocket motors once the missiles are propelled outside the atmosphere, after which they pursue an unpowered ballistic course to the target.6 Jane’s Aerospace Dictionary defines cruise missiles as aerodynamic vehicles that are “wing supported.” A more restricted definition of cruise missiles would relegate them to the category of aerodynamic missiles employing air-breathing propulsion to achieve extended ranges (e.g., the US Tomahawk and the Russian AS-15 cruise missiles).
The first aerodynamic missiles were adapted from drones or manned aircraft reduced in size or range to achieve the desired range-payload objective. Designed with two wings and three surface tails (not until the 1960s did four-wing, four-tail cruciform designs come along), they used standard, liquid-fueled aircraft engines and autopilots for guidance and control. Increasingly more sophisticated guidance schemes replaced these original designs, including command updates, terminal guidance having passive or active radar, and passive infrared (IR) seekers. Television and IR imaging systems came along about the same time that inertial navigation systems (INS) replaced autopilots. Liquid fuels eventually were replaced by solid propellants, and air-breathing engines (turbojets and turbofans) finally came along to extend missile range. When higher specific energies were desired for increased speed or range, ramjets were employed.
Aside from the German V-1 cruise missile, most aerodynamic missiles were produced to attack ships and airplanes or to defend coastal areas. Later, some were adapted to attack land targets. Aerodynamic missiles can be launched from the ground, aircraft, ships, or submarines. Most, to date, have been relatively short-range systems such as the greatly proliferated ASCMs, which are now in at least 40 third world military arsenals.
Understanding what motivated the third world to acquire and develop ASCMs is important because it may shed light on what may occur in the 1990s and beyond as regards land-attack cruise missiles. Perceived military utility appears to have been a compelling factor in explaining the rapid proliferation of ASCMs throughout the third world. Moreover, despite their great expense (a typical ASCM costs about $800,000), ASCMs promise high payoff for third world nations that lack the prestige and operational flexibility of large military establishments. ASCMs offer these countries the ability to defeat a major naval combatant in a superpower’s navy. Despite the vast differences in gross national product and military capability between third world nations and the industrialized powers, one accurately placed ASCM launched from a third world patrol boat or offshore launcher is capable of achieving strategic results. Argentina’s use of Exocet ASCMs in the Falklands War against the British Royal Navy furnishes perhaps the best example of both how effective ASCMs can be and how close a third world power came to achieving strategic results with just one weapon system.7
The US has become the most prolific exporter of cruise missile systems in the form of the Harpoon ASCM. This cruise missile is a second-generation system having four clipped-tip triangular wings at midbody and four smaller wings as moving control fins at the rear—a more sophisticated design compared with the first-generation airplane design. It can be launched from ships, submarines, and aircraft; uses a turbojet engine for propulsion; and has an active radar seeker for terminal guidance.8 The Harpoon-1C has a range of 100 to 120 kilometers. Overall, the US has transferred Harpoons to 23 nations, including North Atlantic Treaty Organization allies, the Middle East (Iran included), the Far East, and South America. Taiwan has reverse-engineered the Harpoon into the Hsiung Feng-2 (HF-2), which is reportedly for sale.
Harpoons in particular—and ASCMs generally—are relevant to the proliferation of land-attack cruise missiles for at least two reasons: (1) they are so widely proliferated within the third world and (2) they are potentially adaptable to land-attack missions. In the case of the Harpoon, its land-attack version is the US Navy’s standoff land attack missile (SLAM), which gained prominence in the Gulf War of 1991. Thus, it is safe to assume that countries that have acquired the Harpoon at least have an important building block for expansion into the land-attack area, however short-range that might be. The key to extending the range of cruise missiles lies in engine, guidance, and navigation technology.
Because cruise missiles for land attack—especially longer-range missions—require sophisticated guidance and complicated support infrastructures to map terrain, they have been relegated largely to superpower arsenals. However, both technology push and doctrinal drive are creating compelling incentives for third world nations to acquire land-attack cruise missiles capable of precise delivery of both conventional payloads and nuclear, biological, and chemical (NBC) weapons.
Technology push stems from numerous factors, the most important of which is the widespread availability of the navigation-and-guidance technology of commercial satellites, together with a variety of increasingly sophisticated mission-planning tools and commercially available satellite imagery. Combined, these technologies and products stand as the major missing elements in helping explain why more third world nations have not already developed or procured land-attack cruise missiles in militarily significant numbers. Worldwide technology diffusion also is prompted by increased motivation on the part of the developed world to sell sophisticated technology and systems to the third world as the developed world’s needs shrink in the aftermath of the cold war.9
With the demise of the bipolar world, technology push interacts strongly with doctrinal need. Regional powers now have even greater incentive to seek regional self-sufficiency and security from potential adversaries. Perhaps the clearest example of international system change interacting with technology proliferation is reflected in Russia’s arms sales. As heir to the former Soviet Union’s foreign policy, Russia has chosen not to continue furnishing the far-flung security guarantees that her predecessor state so generously distributed around the globe during the height of the cold war. Nevertheless, while formal security guarantees may have evaporated, the collapse of the Soviet empire has led to a virtual fire sale of high technology, weapon systems, and scientific talent to many of her former allies—and virtually anyone else with sufficient capital.
One major consequence of the above trends is that the most sophisticated versions of the industrial world’s land-attack cruise missiles may be transferred to third world recipients. For a glimpse of possible future transfers, one need only consider Russia’s offering, at the Abu Dhabi Defense Exhibition of February 1993, of a shorter-range version of the 3,000-kilometer-range AS-15 cruise missile or the French Apache stealth cruise missile, which was on display for export at air shows in Paris (June 1993 and 1995) and Singapore (February 1994).10 Direct transfers of advanced-technology systems such as these could accelerate indigenously based development efforts as well as directly threaten regional and Western interests—particularly if they fall into the hands of rogue states or states with reckless transfer practices. Thus, the extent to which existing export controls preclude or constrain such transfers is a topic of significant importance.
To what extent the third world will react to the availability of new guidance-and-control technology and acquire land-attack cruise missiles depends on several factors, not the least important of which is the effectiveness of voluntary controls on the part of the industrial world. Third world nations also must make difficult choices about the level of investment in development of domestic infrastructure relative to national defense programs. Within national defense programs, priorities inevitably compete for finite resources.
Because prestige is frequently an important factor in the third world’s acquisition of a weapon system, operational issues are just as often less critical in motivating a country to acquire a particular weapon system. This is especially true with respect to the way many countries view ballistic missiles. However, Tomahawk’s performance in the Gulf War has improved—if not equalized—the prestige value of cruise missiles relative to ballistic missiles.11
If, on the other hand, the degree of survivability against a Western power’s air force is the principal criterion for judging the relative importance of major weapon systems, cruise missiles might not become an alternative to ballistic missiles but to manned aircraft. Third world aircraft are especially vulnerable to preemptive attacks, particularly with the advent of stealth aircraft and low-observable cruise missiles. Tied as they are to vulnerable airfields, huge investments in aircraft may not make as much sense as a more balanced approach that includes far more survivable and ground-mobile cruise and/or ballistic missiles.
Another useful way to look at investment in land-attack weaponry is to compare the relative cost and operational advantages and disadvantages of cruise and ballistic missiles. On the issue of relative cost, cruise missiles clearly are less costly to design, develop, procure, maintain, and operate. Although the relative costs are much closer than they once were, it is insightful to compare the relative costs of the German V-1 cruise missile and V-2 ballistic missile programs. Put simply, the costs of the two programs reflected the difference in complexity between the simple V-1 design and the far more elaborate V-2 design. V-1s were procured under a contract with German industry for the equivalent of $500 per unit in 1943 dollars. By contrast, each V-2 cost roughly 500 times more than a V-1 cruise missile.12
In today’s combat environment, cruise missiles possess certain notable advantages over ballistic missiles. Perhaps the most important one lies in the area of accuracy. The aerodynamic stability of the cruise missile permits the use of less sophisticated and therefore less costly guidance-and-control methods than is the case for ballistic missiles, which must undergo the stresses of reentry and high speed. New commercially available guidance-and-navigation technology offers delivery accuracies at costs substantially lower than far more complex ballistic-missile guidance systems. Such accuracy is possible because cruise missiles can receive satellite-navigation corrections all the way to the target from the US global positioning system (GPS) or Russia’s global navigation satellite system (GLONASS).
Most of the ballistic missiles currently deployed in third world arsenals possess circular error of probability (CEP) in the range of 1,000 to 2,000 meters.13 Third world ballistic missiles can potentially receive satellite navigation corrections only until main-engine cutoff, which occurs early in their flight sequence. Assuming satellite navigation corrections before main-engine cutoff, third world ballistic missiles will be relegated to CEPs of no better than 200 to 300 meters for the foreseeable future. For example, China is developing the M-9 missile with a reported CEP of 300 meters. Despite the drawbacks of command guidance, the Indian Prithvi missile employs such guidance in combination with an INS to achieve a CEP in the neighborhood of 250 meters. Better accuracies are theoretically possible for third world ballistic missiles with the addition of map-matching guidance schemes integrated into maneuvering reentry or postboost vehicles for the terminal-delivery phase.14 The latter improvements, however, are both costly and subject to some export controls. In sum, the relative inaccuracy of ballistic missiles when compared with cruise missiles proscribes the effectiveness and utility of the former when they are equipped with conventional payloads. Cruise missiles, by contrast, offer the third world the capacity to attack military targets effectively without resort to NBC weapons.
Cruise missiles also possess other appealing operational features when compared with ballistic missiles. The fact that they can be placed in canisters makes them particularly easy to maintain and operate in harsh environments. Their relatively compact size offers more flexible launch options, more mobility for ground-launched versions, and a smaller logistics burden, which could reduce their battlefield vulnerability to detection—and thus improve their prelaunch survivability. Moreover, cruise missiles dictate no special launchpad stability requirements and can be launched from commercial ships and airplanes, as well as ground launchers. Finally, the cruise missile’s aerodynamic stability, which makes it an inherently easier and cheaper platform from which to achieve precise delivery of conventional payloads, also makes it a better platform for effective dispersal of chemical and biological agents.
Exhaust plumes from cruise missiles are not generally detected by launch-warning systems and, unlike the flight paths of ballistic missiles, those of cruise missiles are not predictable. Most importantly, however, cruise missiles can fly low and thereby pose severe detection challenges—even for airborne radars—due to ground clutter. As higher-quality terrain-elevation data become available through the commercial marketplace, future third world cruise missiles will place stress on the most capable of existing air defenses through very low flight profiles. Reductions in RCSs, which are generally easier to accomplish in more streamlined cruise missile designs than for manned aircraft, will further exacerbate the challenge to air defenses.
Perhaps the most demanding problem for defense against cruise missiles stems from their low cost. The US Army estimates that for a given investment of $50 million, a third world nation could acquire at least 100 cruise missiles. An equal investment for ballistic missiles would purchase only 15 tactical ballistic missiles and three transporter-erector-launchers.15 Thus, while the individual penetration survivability of a cruise missile with a large RCS may not compare favorably with that of a tactical ballistic missile, saturation attacks with low-cost cruise missiles could more than compensate for this deficiency—especially in light of the cruise missile’s better accuracy and resulting higher lethality.
The design requirements for the original cruise missile entailed some form of simple midcourse guidance (preprogrammed autopilot or remote/command guidance), a conventional airframe (metal skin structure with conventional aerodynamic flight controls), conventional propulsion (jet propulsion or use of liquid rocket motors), and terminal guidance (either passive radio-frequency homing, radar, or passive IR for terminal homing). Such designs possessed severe limitations. Midcourse guidance had limited autonomy and accuracy, while propulsion systems produced limited ranges due to poor fuel efficiency (typically 300 kilometers or less). Terminal guidance systems required a “cooperative target,” in that the ability to acquire targets at operating ranges beyond 150 kilometers was severely limited by uncertainties in midcourse guidance.
The two critical enabling technologies that promise to create major incentives for the third world to acquire cruise missiles include precise navigation and guidance technology (GPS and GLONASS) and higher efficiency, lower-volume engine technology.
Navigation-and-Guidance Technology and Systems
Satellite navigation and guidance offer a straightforward solution to the challenges of midcourse and terminal guidance enumerated above. By using very accurate satellite-navigation updates together with even a rudimentary INS, a modern cruise missile can achieve autonomous midcourse guidance and deliver a payload to within a few meters of its intended target.
The US GPS system known as navigation satellite timing and ranging (NAVSTAR) consists of 21 satellites with three spares. Cruising in polar orbit, each satellite has a clock and transmits a signal, enabling a ground receiver with a similar clock to determine its exact position on the earth. A ground station maintains accuracy by introducing minute corrections into the system. One needs signals from three satellites to achieve a precise two-dimensional position and from four satellites for a three-dimensional fix. Receipt of signals from more than four satellites only increases the accuracy of the fix.16
Each satellite transmits two signals with slightly different frequencies.17 Coarse/acquisition (C/A)-code signals are available to all users and furnish an accuracy of roughly 30 meters. The precision (P)-code signals, which are encrypted, are intended only for military users; they deliver an accuracy of roughly 15 meters. Because the Department of Defense (DOD) fears that C/A-code accuracy is sufficient to threaten US security interests, it has introduced a feature—called selective availability (SA)—that intentionally degrades the C/A code signal to produce an accuracy of 100 meters in latitude and longitude and 140 meters in altitude.18
However, one can correct SA by employing differential techniques (DGPS), consisting of a receiver and broadcast station on a geodetically referenced site, which applies a correction to the GPS signal and rebroadcasts that correction to portable units within a radius of around a few hundred kilometers.19 The application of DGPS to cruise missile guidance and navigation is illustrated in figure 6.1. The introduction of wide-area DGPS is overcoming the inherent range limits for local-area DGPS service by collecting local-area differential corrections and transmitting them to a central facility, which then sends them to a satellite for broadcast. Reports indicate that using DGPS techniques can improve accuracy by a factor of 10 for the C/A-code signal; for the military P-code, one estimate suggests the attainment of accuracies of between 75 centimeters and five meters.20 The emergence of DGPS, combined with the explosive growth of commercial GPS users, may force DOD to abandon SA altogether.21
Importantly, one can incorporate differential GPS data not only into weapon systems but also into the making of very accurate map products for both mission planning and terrain-contour matching. Commercial DGPS systems are readily available on the open market throughout the world, with prices dropping in accord with price reductions in the general electronics marketplace.
Russia’s GLONASS is slated for completion during 1995–96. Because GPS is fully deployed, GLONASS may be marketed less as an independent source of satellite-navigation information than as a complement to GPS by virtue of the fact that joint use ensures the reliability of GPS and actually improves its accuracy.
As with the US GPS system, GLONASS will deploy 21 satellites (with three spares). Technically, it is similar in principle to GPS, although its coordinate system and the orbital planes of the satellites are somewhat different. Like GPS, GLONASS has C/A-code and P-code equivalents with roughly the same accuracy as GPS’s respective codes. GPS- and GLONASS-integrated receivers already have been developed and tested by Honeywell and Northwest Airlines for airline applications, with accuracies reportedly below 20 meters.22 GLONASS- and GPS-integrated receivers furnish an attractive option for third world users who fear any degradation of GPS signal quality and accuracy.
Integrating GPS and INS
A major constraint in third world missile performance relates to the relative quality of the INS. By using accelerometers and gyroscopes that detect motion and calculate needed changes in relative position, INSs furnish guidance and control for both aerodynamic and ballistic missile systems. Unfortunately, INSs accumulate inaccuracies as a function of time. Given the slow speed of long-range cruise missiles, INSs alone do not provide sufficient accuracy for conventional missions.
The advent of GPS has changed the INS picture in revolu- tionary terms—and in just a decade. Consider that in the early 1980s, third world countries had marginal navigation systems, such as the attitude-heading reference system for the MiG-21, MiG-23, and MiG-25 aircraft, and poor INS for their ballistic missile systems—mostly Soviet-furnished Scuds. A decade later, the developing world is just beginning to gain access to radically improved navigation and guidance by coupling GPS receivers with cheap and relatively inaccurate INS systems, which are widely available for commercial aircraft. Hybrid technology (INS plus embedded GPS) is now widely available.23 Overall, a quantum jump in capability (roughly a 15-year jump) has occurred; this capability will become increasingly available for military applications in the decade ahead at prices that continue to drop.24
Evidence exists that several countries are actively engaged in exploiting GPS, possibly for missile-guidance purposes. Pakistan, China, Burma, Israel, Iran, Russia, France, and Germany have all shown interest in the integration of GPS into missiles and unmanned air vehicles. Several countries (Pakistan, China, India, Indonesia, and Thailand) appear headed toward seeking DGPS to improve the quality of their photogrammetric techniques.25 For its part, China plans by about 1998 to deploy its own satellite-navigation system—dubbed Twin Star—with 20-meter accuracy.26
Mission Planning for Cruise Missile Applications
The advent of GPS technology also brings within the third world’s reach all the necessary tools for sophisticated mission planning and possibly even terminal-guidance schemes employing terrain-matching techniques. Although GPS as a guidance technique for cruise missiles obviates the need for detailed digital map making, some countries may nevertheless desire developing accurate digital maps to improve the penetration and survivability of their cruise missiles. Flying cruise missiles at very low altitudes dictates the need for accurate terrain-elevation data, which can be preprogrammed into the cruise missile, thereby avoiding the need for an expensive terrain-avoidance radar system.
The products for such mission planning are readily available today. Conventional wisdom has it that civilian space programs have little military utility. In fact, commercial products of SPOT and LANDSAT satellites were used extensively in Operations Desert Shield and Desert Storm for broad-area search and mission planning. Moreover, the recent US government decision to permit the sale of sophisticated spy-satellite technology and products (namely, imagery that depicts objects one meter in diameter) to commercial customers has generated concern that militarily relevant imagery will become available to potentially hostile powers, despite safeguards for controlling its spread.27
Geographic Information Systems (GIS), comprising personal-computer hardware and very sophisticated software (e.g., Auto-Cad), now permit users to make very accurate digital maps with GPS data inputs. One can use such hardware and software capabilities for more than just preprogramming the route of a cruise missile. Better maps and commercially available satellite imagery allow third world states to develop better targeting by improved photogrammetric techniques. For example, the Center for Mapping at Ohio State University blended imagery with DGPS data to archive data on highways and land features. The center used an eight-channel GPS receiver, stereoptic cameras, and standard GIS technology costing roughly $850,000 in order to map several states. Their output permitted vans traveling at 50 to 60 miles per hour to achieve accuracies of approximately two and one-half meters.28 In other words, the technology is commercially available today to permit proliferating states to digitize satellite imagery generated by SPOT and LANDSAT, add position information taken from differential GPS, and employ it together with a radar altimeter to create a terrain-contour-matching (TERCOM)-like guidance system for intermediate and terminal homing. The challenge is one of integrating these components into a weapon system—a difficult challenge indeed for any third world country. Yet, in a decade or so, it is safe to say that such targeting systems will probably be available in third world cruise missiles.29
Third world countries are already exploiting the benefits of this technology. India’s first cruise missile, the Sagarika, will reportedly employ a terrain-matching system for guidance and low-level flight.30 The US government approved the sale of GIS technology to Iraq in March 1987 for the stated purpose of remote sensing and photo interpretation, according to Iraq’s Remote Sensing Center in Baghdad. After using the center’s new capabilities to support its war against Iran, Iraq started taking a strong interest in imagery of Kuwait and Saudi Arabia. According to the chairman of SPOT Image Corporation, between 1988 and 1990, his firm delivered to Iraq 20 images of the area—including overlapping ones. SPOT denied another Iraqi request after its invasion of Kuwait in August 1990.31
The simple fact is that the US and its allies no longer have a monopoly on space technology. Spending by foreign govern- ments on space is increasing, and cooperative and cost-sharing agreements have reduced each country’s burden.32 Moreover, widespread availability of low-cost, dual-use space technologies (such as charge-coupled devices [CCD]) means that the prospects for enhanced imagery support to third world users will inevitably increase.
Third world countries are not likely to develop the indigenous capacity to produce efficient turbofan engines for small, long-range cruise missiles by the end of this decade. But that does not mean that one cannot acquire turbofan engines through the international marketplace. Turbofan engine technology—like that reflected in the Williams F-107 used for the air launched cruise missile (ALCM) and Tomahawk long-range cruise missile—is available in Russian systems such as the AS-15 and SS-N-21 long-range cruise missiles. As already noted, derivatives of the AS-15 cruise missile outfitted with turbofan engines have been advertised for sale at international air shows. Moreover, US commercial sales to China of turbofan engines for jet trainer aircraft illustrate the challenge associated with controlling cruise missile proliferation at a time when there are far fewer limits on manned aircraft—commercial and military alike.33
Turbojet engines are available from a variety of industrial and third world manufacturers.34 Several countries, including Russia, China, France, and the United Kingdom, produce turbojet engines suitable for cruise missile applications. Given past practices, Chinese and Russian sales to the third world are quite likely in the future; French and British sales have already occurred. Moreover, US turbojet engines are widely proliferated with the Harpoon ASCM. Also involved in the manufacture and sale of small turbojet engines for supersonic aircraft are India, Israel, South Africa, and Taiwan. Depending on payload weight, such turbojet technology in a small-engine configuration ought to be able to support cruise missiles capable of ranges out to 1,000 kilometers.
In evaluating the developments discussed thus far, we conclude that to date the problem of cruise missile proliferation has centered on antiship—not land-attack—systems. Still uncertain—though evidence of strong third world interest is growing—is just how aggressively regional adversaries of the US will exploit the revolution in guidance and navigation that now makes land-attack cruise missiles appear so attractive as an alternative or complement to ballistic missiles and attack aircraft.
It is also fair to say that the cruise missile threat has been both understated and exaggerated—though understatement greatly dominates exaggeration. We judge third world incentives to acquire land-attack cruise missiles to be sufficiently compelling to suggest a threat of some considerable magnitude emerging by the end of this decade and growing significantly more prominent thereafter. To the extent that virtually no one has considered this prospect, the cruise missile threat has been understated. In stark contrast, “a virtual blizzard of books, scholarly articles and now official analyses” on ballistic missile proliferation has offered just about everything that can be said about that subject—or so notes Janne Nolan in the journal Survival.35 With a few notable exceptions, that is not the case for cruise missile proliferation.36
In part, the relative levels of attention are a function of the recent emergence of the enabling technologies for land-attack cruise missiles. Ballistic missile proliferation, by contrast, came into prominence as an important security issue in the mid-1980s. In addition, political controversy in the US and Western Europe surrounding active ballistic missile defenses has fixated the analytical and policy communities on the issue of ballistic missile proliferation—at the expense of a broader consideration of other, perhaps equally serious, proliferation trends.37
Exaggeration of the cruise missile threat is reflected in the general tendency to focus on the individual components of land-attack cruise missile capability—particularly the implica- tions and impact of the availability of GPS for cruise missile guidance—without giving sufficient attention to the challenges facing the third world in systems integration. What separates the industrial from the developing world is the former’s ca- pacity to integrate technology components into complex systems that produce repeatable results according to desired specifications. When we approach cruise missile proliferation purely from the standpoint of individual technology components, it is easy to conclude that the spread of cruise missiles represents a more significant threat than ballistic missile proliferation.
Whatever the reasons for the imbalance, the prospects for cruise missile proliferation undoubtedly are great. Whether militarily significant threats emerge within five or 10 years naturally depends on a number of difficult variables, not the least important of which is the effectiveness of existing export controls.
In light of the alternative paths available to the third world for acquiring land-attack cruise missiles (namely, upgrading ASCMs or UAVs for land-attack missions; developing an indigenous manufacturing capability; or purchasing directly from the industrial world), clearly the relative effectiveness of existing export controls will significantly shape the pace and scope of the future proliferation of cruise missiles. The principal international-policy mechanism for controlling exports of missiles capable of delivering weapons of mass destruction is the Missile Technology Control Regime.
The MTCR was announced in 1987 as a voluntary accord (i.e., not a legally binding international treaty) aimed at limiting “the risks of nuclear proliferation” by controlling transfers that could contribute to the development of “nuclear weapons delivery systems other than manned aircraft” (emphasis added).38 The regime had seven original members: the United States, Canada, West Germany, France, Italy, Japan, and the United Kingdom.
In 1993 MTCR member governments agreed to extend the regime’s purview to cover missile-delivery systems for chemical weapons (CW) and biological weapons (BW). As of 1995, 25 countries had joined the MTCR as full partners.39 Russia was invited to join in June 1995. China has apparently agreed to abide by the MTCR guidelines but has not joined as a full partner.40
How the MTCR Works
The MTCR seeks to accomplish its purpose through member adherence to an agreed set of export-policy guidelines, which are applied to an extensive list of items contained in the MTCR’s equipment and technology annex (appendix F). The annex itself is divided into two sections: category 1 contains complete missile systems and highly sensitive missile-related equipment; category 2 lists dual-use items. The MTCR offers general export guidance applicable to the entire technical annex, as well as specific guidance tailored to each annex category.
The MTCR’s general guidance directs members to make a “strong presumption to deny” transfers of any annex item or any missile (regardless of its inclusion in the technical annex) that the member believes is “intended” for the delivery of NBC weapons.41 In evaluating the recipient government’s end-use intentions, MTCR member states are directed to undertake, inter alia, an assessment of the capabilities and objectives of the recipient’s missile and space programs, as well as an evaluation of the significance of the transfers in terms of their potential to “contribute” to the development of delivery systems (other than manned aircraft) for NBC weapons.42
Category 1 items are, for all intents and purposes, automatically considered able to “contribute” to the development of NBC missiles. Within category 1, item 1 includes complete rocket systems (including ballistic missile systems, space-launch vehicles, and sounding rockets) and UAV systems (including cruise missile systems, target drones, and reconnaissance drones) capable of delivering 500-kilogram payloads to ranges of 300 kilometers or more. Item 2 includes certain major subsystems usable in rockets and UAVs meeting the 300-kilometer/500-kilogram threshold, as well as specially designed production facilities and production equipment for 300-kilometer/500-kilogram missiles and their major sub-systems.43 The category 1 guidelines are supplemented by language in the technical annex of 1993, which directs MTCR members to assess whether recipient states could modify missiles or components via range-payload trade-offs to develop missiles meeting the 300-kilometer/500-kilogram threshold.44
Because category 1 items are inherently usable as—or in the development of—missiles for NBC delivery, MTCR members should make “a strong presumption to deny” category 1 transfers, regardless of the recipient’s “intended” end use. In the unlikely circumstance that a member government does decide to export a category 1 item, it should obtain “binding government-to-government” assurances and take “all steps necessary to ensure” that the item is put only to its stated end use. Members are advised further that the export of category 1 production facilities is flatly prohibited.45
MTCR category 2 lists a variety of subsystems, components, machinery, and technologies usable in the development of missiles and other military systems, as well as commercial systems. Major classes of items include, inter alia, propulsion components, propellants, missile structural composites, flight-control systems, missile computers, reduced-observables technology, launch equipment, and test facilities.46 In a reflection of the MTCR’s expanded mandate to cover CW- and BW-capable missiles, item 19 was added in 1993. This category 2 item includes complete rocket or UAV systems capable of “a maximum range equal or superior to 300 kilometers,” regardless of payload.47 Moreover, since item 19 is covered by the annex language on range-payload trade-offs, even shorter-range systems such as ASCMs might be covered if they could be modified through payload reductions to achieve a 300-kilometer range.
An MTCR member government may export category 2 items and associated production facilities at its own discretion, but only after it has determined that the items are not usable in a missile for NBC delivery, or in one captured by the 300-kilometer/500-kilogram threshold of MTCR category 1. If the internal finding is positive for either application, then the MTCR member is obligated to obtain assurances from the recipient state that the items will not be put to these end uses.48 However, end-use assurances are not required for a variety of UAV-relevant items if they are “exported as part of a manned aircraft or in quantities appropriate for replacement parts for manned aircraft.”49
The MTCR export guidelines are implemented according to national legislation. Licensing and enforcement activities, therefore, vary among member states. The accord makes no provision for penalizing countries that violate its guidelines, but individual members can—and do—impose sanctions on violators unilaterally. MTCR members meet at least once a year to discuss enhancements to the regime as well as intelligence information on missile projects of concern. A primary strength of the regime is member agreement that an export denial by one member state will be upheld by all.
Analyzing the MTCR’s Effectiveness
As the only active regime aimed specifically at stemming the diffusion of missile systems to the third world, the MTCR represents a constraining mechanism of considerable importance. In addition to helping derail some surface-to-surface ballistic missile programs, enforcement of the regime’s provisions has slowed the emergence of new states wielding ballistic missiles. The list of suppliers also has shrunk. Notably, the former Soviet Union no longer dispenses Scuds to client states. Argentina, Brazil, South Africa, South Korea, and Iraq apparently have terminated indigenous ballistic missile programs, leaving North Korea as the main supplier of MTCR-restricted ballistic missiles.50
That said, the regime does suffer from weaknesses regarding cruise missile proliferation, stemming largely from the fact that the MTCR and its key supporters have yet to establish a firm consensus against the spread of cruise missiles. This fundamental shortfall ensures an additional weakness. Specifically, the MTCR’s controls on critical enabling technologies for UAVs and complete cruise missile systems are not stringent enough to impede significantly the spread of advanced cruise missiles.
As to recognition of the emerging cruise missile threat, one must recognize first and foremost that, relative to NBC weapons, the consensus against missile proliferation in general has yet to become firmly established. Indeed, a stronger consensus appears to exist—even among MTCR members—for restricting ballistic rather than cruise missile or UAV systems. This conclusion is supported by the fact that key MTCR members have demonstrated a greater willingness to export cruise missiles and other UAVs than ballistic missiles.
The US has transferred short-range ballistic missiles to just three third world countries.51 In contrast, it has sold ASCMs to more than a dozen and has sold reconnaissance drones worldwide, including the 2,250-kilometer-range Scarab (fig. 6.2) to Egypt. France is reported to have sold ballistic missiles to just one third world customer;52 Britain and Italy have not transferred any. Yet, France has sold ASCMs to a combined total of nearly 30 developing countries. Britain has sold its turbojet-powered, 110-kilometer-range Sea Eagle ASCM to at least three third world nations. Italy has widely exported its Mirach family of UAVs, including a 900-kilometer-range model to Iraq, Libya, and Argentina.53
The US and allied exports cited above demonstrate the MTCR members’ unwillingness to restrict key enabling technologies for cruise missiles. The exported ASCMs and UAVs failed to meet the MTCR’s category 1 threshold—missiles carrying 500-kilogram payloads to ranges of 300 kilometers—and thus escaped its most restrictive export guidelines. But the category 1 threshold is better suited to impeding ballistic rather than cruise missile proliferation. As noted above, shorter-range ASCMs and RPVs can be adapted for land-attack missions.54 From an engineering standpoint, it is relatively easier to “scale-up” the range of an existing cruise missile system than a ballistic missile.55 Indeed, the technology required to produce a 1,000-kilometer-range cruise missile is not fundamentally different from that needed for very short-range systems.56 Hence, UAVs and UAV technologies falling clearly below the MTCR’s range-payload threshold can be exported and applied to the development of long-range cruise missiles.
The fact that the MTCR does not restrict manned-aircraft exports also eases the determined proliferator’s task. This exemption represents a direct way to work around MTCR restrictions on UAVs because the relationship between manned aircraft and UAVs is strong. The structures, propulsion, autopilots, and navigation systems used in manned aircraft are essentially interchangeable with those of cruise missiles and other UAV variants; the same is true for production facilities and equipment for UAVs and manned aircraft.
Hence, to impede the spread of cruise missile production capabilities, the MTCR would have to restrict the sale of aircraft-related technologies. But such restrictions appear no more realistic today than they did when the MTCR was developed in the mid-1980s. In fact, global competition to export aircraft and UAVs, their related technologies, and production facilities is increasing.
The major powers are expected to begin selling off their cold war arsenals of military aircraft.57 They are becoming increasingly dependent on manned-aircraft exports to preserve their defense-industrial bases as domestic military budgets decline in the post-cold-war era.58 Industry analysts predict that, not counting US purchases, the global market for jet trainers during 1995–2000 could total $4 billion. The market for military UAVs could be higher still, possibly reaching $5 billion, and that figure likely will be outstripped by orders for commercial UAVs. In addition, air forces worldwide are expected to begin a rash of fighter upgrades, which will lead to a major trade in aircraft engines, advanced electronics, radar, and other aerospace subsystems.59
Developing countries increasingly are taking advantage of the “buyers market” in aerospace to demand offsets providing indigenous aircraft maintenance—and even production—capabilities. The willingness of former Eastern bloc aircraft producers to undercut the prices of their Western competitors is likely to further accelerate the diffusion of production capabilities related to cruise missiles. Thus, the link between cruise missiles and manned aircraft represents a major challenge to MTCR effectiveness in controlling the spread of enabling technologies and production capabilities for land-attack cruise missiles.
Of course, the quickest way for the developing world to obtain land-attack cruise missiles is to purchase them directly from an industrial-world supplier. Possible confusion over the extent to which MTCR provisions apply to ALCMs may account for France’s apparent willingness to consider exporting its Apache cruise missile (fig. 6.3).
Under development since 1989, the Apache is modular in design and is intended (at present) to come in three different versions:60
The Apache family of cruise missiles was designed from the outset with high terminal effectiveness in mind: a stealthy aerodynamic shape with low RCS and IR signatures and very low-level, terrain-following flight characteristics. Although the Apache’s prime contractor, Matra Missiles and Space, boasts of the missile’s stealthiness in its marketing, it has become increasingly apparent that the French government wants to reserve the most effective low-observable features for the Apache Scalp.64
All versions of the Apache are powered by the same turbojet engine. Variations in fuel loading account for differences in range among the three versions.
The MTCR provides the only established basis to object to Apache exports. French industry officials have argued that the Apache AP is designed to fall under the MTCR’s range-payload threshold (300 kilometers/500 kilograms), given its 140-kilometer range and 520-kilogram submunition package.65 But the MTCR cautions members to “take account of the ability to trade off range and payload.”66 At the very least, Apache AP appears readily adaptable to fly to at least 300 kilometers through payload reductions, which would subject it to the MTCR’s less stringent category 2 (item 19) restrictions. But it is also quite plausible to believe that the Apache AP could fly to at least 300 kilometers with its 520-kilogram payload package if launched from a sufficiently high (yet not operationally implausible) altitude. That capability would place the Apache AP squarely under the MTCR’s category 1 restrictions, requiring the French government to exercise a “strong presumption to deny” any exports. Responding to a reporter’s question about MTCR and the Apache in January 1995, a senior French Defense Ministry official stated that the shorter-range version of Apache does not fall under MTCR guidelines and therefore is not subject to multilateral controls.67
Russian activities are also worrisome. The sharp decline in Russian defense spending has reportedly forced Russian cruise missile builders to search for foreign customers. According to one source, Moscow has sold its SSC-1 Sepal cruise missile (1,000-kilogram payload to 450 kilometers) to Syria.68 Also of concern is the flow of systems and technologies to China.69 Russian technology transfers could facilitate China’s development of advanced cruise missile weapons, and one has reason to question whether China can be persuaded to forgo exporting them, the MTCR notwithstanding.70
Beyond reported transfers, Russia has marketed a variety of cruise missile systems at arms shows around the globe. In addition to the AS-15, among the more troubling systems is an export version of the AS-16, which can carry a 150-kilogram payload to a range of 150 kilometers.71 The effectiveness of the country’s export-control system is open to question, as is its professed commitment to the MTCR.72
Several other MTCR members and adhering states are developing or considering the export of cruise missiles. Israel (an MTCR adherent) is reportedly transforming its Delilah UAV into a 400-kilometer-range ALCM with the aid of Chinese funding.73 The Spanish company CASA announced recently that it wants to build a land-attack cruise missile. The company will keep the missile’s per-unit costs low by using commercial, off-the-shelf technologies. The CASA missile looks similar in design to the French Apache, and its competitive price suggests that Spain may go after the Apache export market.74 For its part, the United Kingdom has withheld an export license for GEC Marconi in connection with that company’s intended transfer of the El Hakim land-attack cruise missile to the United Arab Emirates, which directly funded research on this program.75 In the end, the way Apache transfers are handled is likely to greatly affect the behavior of these other aspiring cruise missile exporters.
Although the problem of cruise missile proliferation is just beginning to manifest itself, the findings presented herein suggest that constraining the spread of cruise missiles may be much more difficult than constraining the spread of ballistic missiles. Hence, a need exists for immediate action while there is still time to constrain rapid advances in the cruise missile threat. In this regard, a critical first step is acknowledgment that the challenge of cruise missile proliferation exists, followed by placing cruise and ballistic missile nonproliferation efforts on an equal footing.
We examined the Clinton administration’s treatment of cruise missile proliferation in congressional testimony, major foreign-policy speeches, and policy proclamations on export controls and counterproliferation initiatives.76 Not one of these key addresses or documents specifically mentioned cruise missiles as an important element in the overall problem of missile proliferation. Each focused instead on the proliferation of ballistic missiles and space-launch vehicles. In view of the prospects for cruise missile proliferation, it will become increasingly important to draw specific attention to the cruise missile dimension of the missile proliferation threat—particularly in light of the export-control challenges detailed above.
Any new MTCR initiatives must be firmly grounded in reality, which dictates that member states recognize that they no longer monopolize aerospace expertise or industrial capabilities. Some developing countries are already producing relatively unsophisticated cruise missiles, and they might exploit satellite navigation systems to build longer-range cruise missiles over time. Moreover, a latent cruise missile production capability exists in many regions because of the globalization of the manned aircraft and UAV industries. Hence, although MTCR members should not abandon technology-denial efforts aimed at unsophisticated cruise missiles, neither should they expect them to have a major impact.
We recommend that the MTCR focus its attention on slowing the spread of relatively advanced systems, such as stealthy cruise missiles capable of high speed and/or long range. The critical enabling technologies needed to acquire advanced cruise missiles—including stealth and advanced propulsion systems—are produced almost exclusively by MTCR members or by states that might be persuaded to support tighter controls. Because stealth and advanced propulsion systems are covered under the dual-use section of the MTCR, member governments can export them at their discretion. Given the particular sensitivity of stealth-technology transfers, however, MTCR members should consider enhancements to the regime.
Low-observable or stealth technologies are covered by item 17 in category 2 of the MTCR’s equipment and technology annex. MTCR members should therefore export stealth technologies in accordance with the regime’s category 2 guidelines, but the restriction applies only if the technologies are usable in the systems described in category 1, items 1 and 2. These items include, respectively, complete missiles capable of carrying 500-kilogram payloads to ranges of 300 kilometers or more and certain major subsystems and equipment for 300-kilometer/500-kilogram missiles. Hence, to better restrict the spread of low-observable cruise missiles, the least controversial modification would be to extend item 17 applications to cover not just category 1 missiles, but also those missiles described in category 2, item 19 (i.e., missiles capable of a maximum range of 300 kilometers or more, regardless of payload). The fact that this change did not occur at the time item 19 was fashioned appears to have been an oversight, but, in any case, subjecting stealth technologies to category 2 export guidelines would still leave transfers at the discretion of individual supplier states.
The second and potentially more effective option would be to make stealth technologies subject to category 1 controls (i.e., “a strong presumption to deny” exports). In its simplest form, one could effect this change by transferring the stealth technologies described in item 17 to category 1, item 2. Some people will argue that considering low observables as a major subsystem has nothing to do with the MTCR’s original intent (i.e., controlling delivery systems for weapons of mass destruction). Yet, the addition of stealth to a cruise missile essentially furnishes it with the same characteristics of ballistic missiles that gave impetus to the MTCR’s creation: difficulty of defense, short warning time, and shock effect. Moreover, the fact that cruise missiles represent an even more effective means of BW and CW delivery than do ballistic missiles gives further weight to the merits of such a change.
Our discussion of whether or not Apache is subject to MTCR guidelines points to another potential improvement in the regime’s technical annex: clarifying the way trade-offs between range and payload for cruise missiles are calculated. It appears that the MTCR’s reference to range and payload was written with surface-to-surface ballistic missiles in mind. Calculating such trade-offs for ALCMs is far more difficult, given the various flight profiles such missiles might employ. A discussion of how one determines “payload” in view of the modularity of modern cruise missiles also would seem sensible.
Beyond advocating modifications to the MTCR’s technical annex, the US should take the lead in a more general effort aimed at raising MTCR members’ awareness of the emerging cruise missile threat. Members should be sensitized to the fact that, with the predicted worldwide expansion of the aircraft upgrade and UAV markets, export-control authorities can expect export-license applications for advanced subsystems usable in cruise missiles. MTCR governments should take such applications as a warning signal. Thereafter, member states should thoroughly investigate the end-use intentions of recipient states, especially when the recipient does not have current, acceptable aerospace systems employing such technologies. Members should prohibit exports of stealth and advanced propulsion systems or proceed only with utmost caution if available evidence suggests that the recipient government is interested in acquiring cruise missiles. If the export is permitted, end-use monitoring would be advisable, even in cases in which end uses involving manned aircraft seem certain. Monitoring might deter—although it cannot prevent—diversions of end items and production equipment from acceptable aerospace projects to cruise missile applications.
Even the most perfectly crafted export-control strategy would be limited in what it could achieve, which is to slow the pace of—not stop—cruise missile proliferation. Yet, slowing the pace can raise the costs and risks that proliferators must incur to acquire advanced cruise missiles. It also furnishes the United States and other affected states with time to develop effective defenses against emerging threats. Demonstrating that effective cruise missile defenses are being developed apace with the emerging cruise missile threat could have a strong deterrent effect on third world acquisition plans for such missiles.
l. “The Weapons Proliferation Threat” (Washington, D.C.: Central Intelligence Agency, Nonproliferation Center, March 1995), 3.
2. See, for example, Kenneth Munson, “The Unmanned Air Vehicle Comes of Age,” Jane’s Defence Weekly 24, no. 3 (22 July 1995): 21.
3. As cited in Bryan Bender, “Defense Science Board Report Brands Cruise Missiles Increasing Threat,” Inside the Army 7 (30 January 1995): 1.
4. RCS is a standard measure defining how visible a target is to a radar and therefore indicating at what range a given radar can detect and track the target. For a tutorial on the importance of RCS in aircraft and cruise missile design, see Bill Sweetman, Stealth Aircraft: Secrets of Future Airpower (Osceola, Wis.: Motorbooks International, 1986), especially chap. 3.
5. Without any great fear of Iraqi aircraft threats, US Patriot air defense batteries could focus their radars on high-angle ballistic missile threats, thereby avoiding the fratricide problem. See also Dennis Gormley, “Cruise Missile Threat Quietly Rises,” Defense News 10 (27 March–2 April 1995): 27–28.
6. 0f course, one can configure a reentry bus to undertake terminal maneuvers to avoid active defenses. For more on the differences between ballistic and cruise missiles, see System Planning Corporation, Ballistic Missile Proliferation: An Emerging Threat, 1992 (Arlington, Va.: System Planning Corporation, 1992), passim.
7. For the best appraisal of the Falklands conflict and the impact of Exocet cruise missile attacks on British naval operations, see Max Hastings and Simon Jenkins, The Battle for the Falklands (New York: W. W. Norton, 1983), 153–54, 316–20.
8. The sea-skimming version of the Harpoon employs a radar altimeter to get the missile to the target area; another version employs a climb-and-dive approach, necessitating an inertial navigation scheme in the high-altitude mode.
9. The US Air Force and US Navy—like the French air force over the last two decades or more—may not be able to go into large-scale production for a future fighter until sufficient foreign sales are made to bring down per-unit costs. In an effort to preserve national industrial bases, nations may err on the side of transferring technology by reducing the number of production lines (and accompanying overhead and production costs) to perhaps just the frontline model. As a consequence, prospective buyers have a rare opportunity to purchase the best the West is producing. With offsets included, the third world recipient is receiving not just aircraft but technological infra- structure as well.
10. Information on these displays is derived from interviews with attendees of the Abu Dhabi Air Show and from attendance and discussions with company representatives at the Paris and Singapore exhibitions.
11. For example, see Patrick J. Garrity, Why the Gulf War (Still) Matters: Foreign Perspectives on the War and the Future of International Security (Los Alamos, N.Mex.: Center for National Security Studies, 1993), passim.
12. David Israel, “History Repeats?” February 1992 (unpublished paper).
13. The most widely proliferated longer-range ballistic missile in the third world is the Soviet-designed Scud B. Declassified US Department of Defense estimates assert that Soviet forces could achieve Scud B CEPs of approximately 600 to 900 meters. Third world forces have demonstrated significantly less proficiency in their conduct of ballistic missile operations. It seems unlikely that third world Scud operators could even match the upper bound in accuracy achieved by their Soviet counterparts. Iraq, for instance, achieved CEPs of roughly two kilometers with its Scud-derived, 650-kilometer Al Hussein missiles during the Gulf War of 1991. For details and source materials on Scud B accuracies, see Dennis M. Gormley, Double Zero and Soviet Military Strategy: Implications for Western Security (London: Jane’s Publishing Co., 1988), 75–77; for details on Al Hussein accuracies, see Gregory S. Jones, The Iraqi Ballistic Missile Program: The Gulf War and the Future of the Missile Threat (Marina del Rey, Calif.: American Institute for Strategic Cooperation, Summer 1992), 31–32.
14. For a useful treatment of missile accuracy, see Jones, especially 42–43.
15. Briefing charts, Department of the Army, Office of the Deputy Chief of Staff for Operations and Plans—Force Development, Concepts, Doctrine, and Policy Division, subject: Army Theater Missile Defense.
16. The ideal situation is for a receiver to have access to five signals from five satellites at any one time.
17. For technical details, see Department of Commerce, Federal Radio Navigation Plan, 1990, PB-91-190868 (Washington, D.C.: US Department of Commerce, 1990); and J. J. Spilker, “GPS Signal Structure and Performance Characteristics,” in Global Positioning System, vol. 1 (Washington, D.C.: Institute of Navigation, 1980). The most useful layman’s guide is Jeff Hurn, GPS: A Guide to the Next Utility (Sunnyvale, Calif.: Trimble Navigation, 1989).
18. Accuracy for GPS is defined differently than missile CEP accuracy. Thus, a 100-meter GPS accuracy has a confidence of two dRMS, which means that at least 95 percent of the time, the position information reported is within 100 meters of its true position. By contrast, CEP has a 50 percent confidence level, making CEP four-tenths as large as two dRMS. In other words, a 100-meter GPS accuracy equates to a 40-meter CEP for a missile.
19. For technical details, see V. Ashkenazi et al., “Wide-Area Differential GPS: A Performance Study,” Navigation Journal of the Institute of Navigation 40, no. 3 (Fall 1993): 297–319.
20. W. Seth Carus, Cruise Missile Proliferation in the 1990s (Westport, Conn.: Praeger Publishers, 1992), 61.
21. Cheri Privor, “Panel Says GPS Security Efforts Are Outmoded,” Defense News 10 (5–11 June 1995): 6.
22. Steve Wooley, “Proliferation of Precision Navigation Technologies and Security Implications for the U.S.” (Alexandria, Va.: Institute for Defense Analyses, 1991), 8.
23. For an example of applications in integrated INS/GPS systems, see Mark Hewish, “Integrated INS/GPS Takes Off in the US,” International Defense Review 26 (February 1993): 172–74; and idem, “GPS Users Proliferate following Gulf War,” Defense Electronics & Computing, no. 4 (1992), editorial supplement to International Defense Review 25 (September 1992): 115–20.
24. According to Steve Wooley, stand-alone and relatively accurate INSs for Western commercial aircraft cost something in the neighborhood of $150,000. Cheaper, less-accurate systems—widely available from France, Germany, China, the United States, and the United Kingdom—cost roughly $50,000 but can be updated with GPS and GLONASS. The integration com- plexity varies, depending on the platform. See Wooley, 11. As far as GPS technology is concerned, Rockwell offers the NAVCORE V, a five-channel receiver in embedded-chip form, four by two-and-one-half inches in size, for $450 apiece or $250 in bulk.
25. Wooley, 14.
26. “Chinese ‘GPS’ Project Set,” Aviation Week & Space Technology 141, no. 16 (17 October 1994): 25.
27. Safeguards on misuse of such high-resolution imagery reportedly include the requirement that companies maintain a record of every job requested. Moreover, the government reserves the right to shut down services during “periods when national security and/or foreign policies may be compromised.” As tight as these safeguards may appear, they cannot eliminate the prospect that a hostile power might use an apparently legitimate company to purchase imagery useful for supporting the targeting of fixed military installations. That such services might be eliminated in a crisis only deals with constraining a hostile nation’s access to time-critical imagery; it would not preclude the acquisition in peacetime of militarily relevant imagery for targeting fixed installations. See Edmund L. Andrews, “U.S. to Allow Sale of the Technology for Spy Satellites,” New York Times, 11 March 1994, l[A].
28. Wooley, 19.
29. To build such highly accurate maps using DGPS, the developer must have access to en route navigation points, which should not be difficult to achieve. What’s more, TERCOM guidance is viewed in the US as a great challenge because of the extensive and costly mapping that is required to support TERCOM-guided cruise missiles—not the technology components of the guidance system itself. It should be noted, however, that the US must plan against a variety of worldwide military contingencies, which raises the cost of mission planning considerably. By comparison, a third world nation’s scope of mapping activity will be on a much smaller scale.
30. Rahul Roy-Chaudhury, “India Developing Sea-Based Missile System,” Inter Press Service, 29 September 1994 (LEXIS-NEXIS [electronic news service of Reed Elsevier, Inc., Dayton, Ohio]).
31. Michael Krepon, “Bush Ignored Warnings on Saddam,” Defense News 7 (1–7 June 1992): 19.
32. A study by Rockwell International states that a third world country could exploit commercially available technologies and launch services to procure and launch a two-and-one-half-meter-resolution reconnaissance satellite for less than $60 million. See briefing charts, James R. Howe, Rockwell International, Space Systems Division, Seal Beach, Calif., subject: Nth Country Satellite Threat Estimates, 10 November 1992.
33. According to an account in the Washington Post, AlliedSignal concluded a turbofan deal with the Chinese in 1987. Beijing claimed that the engines would be used in jet trainer aircraft. AlliedSignal officials reminded US authorities that similar turbofans were available from other manufacturers and used in business aircraft around the world. The Commerce Department thus approved the sale. But DOD opposed it, citing an intelligence-community finding that China could use the turbofans to upgrade its Silkworm ASCMs and create cruise missiles capable of carrying 450-kilogram payloads to ranges of about 600 kilometers. China’s proven willingness to sell missiles to the third world raised the possibility that rogue states would acquire the upgraded Silkworms and use them against US forces in the future. Even so, economic considerations ultimately won the day. The Clinton administration approved the half-billion-dollar sale in 1994. See Jack Anderson and Michael Binstein, “Worrisome Engine Sales to China,” Washington Post, 9 May 1994, 14[C]. For further ramifications of this sale, see Bill Gertz, “Russia Sells Rocket Motors to China: Moscow Ignores U.S. Objections,” Washington Times, 13 February 1995, 4[A].
34. See Carus, 76–79, for a useful overview.
35. Janne E. Nolan, review of Going Ballistic: The Build-Up of Missiles in the Middle East by Martin Navias, in Survival 36, no. 1 (Spring 1994): 177–79.
36. The exceptions are Carus and Henry D. Sokolski, “Nonapocalyptic Proliferation: A New Strategic Threat?” Washington Quarterly 17, no. 2 (Spring 1994): 115–27.
37. For an analysis of how politics affected analytical consideration of the threat of Soviet theater ballistic missiles in the 1980s, see Gormley, Double Zero, xi–xx and 174–90.
38. The White House, “Missile Technology Control Regime: Fact Sheet to Accompany Public Announcement” (Washington, D.C.: Office of the Assistant to the President for Press Relations, 16 April 1987). (See appendix E.)
39. The members are Argentina, Australia, Austria, Belgium, Canada, Denmark, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Japan, Luxembourg, the Netherlands, New Zealand, Norway, Portugal, Spain, Sweden, Switzerland, the United Kingdom of Great Britain and Northern Ireland, and the United States of America.
40. In 1990 the USSR first pledged its support for the MTCR’s “objectives.” Russia reaffirmed this commitment with written assurances in 1993 and was invited to join the regime two years later. Chinese officials initially pledged Beijing’s adherence to the regime in 1991, followed by the issuance of a US-Chinese summit statement in 1994 in which Beijing agreed not to export “ground-to-ground” missiles covered by the MTCR’s original guidelines of 1987. Diplomatic disputes later stalled US attempts to gain Chinese adherence to the MTCR’s enhanced guidelines of 1993. Theresa Hitchens, “U.S. Backs Russia in MTCR,” Defense News 10 (17–23 July 1995): 12; “U.S. and Russia Agree to Joint Space Station,” Arms Control Today 23, no. 8 (October 1993): 22; “Secretary’s Talks in China: A Summary of Results,” U.S. Department of State Dispatch 2, no. 47 (Washington, D.C.: US Department of State, Bureau of Public Affairs, 25 November 1991), 859; Department of State, “Fact Sheet: Joint United States–People’s Republic of China Statement on Missile Proliferation” (Washington, D.C.: US Department of State, Office of the Spokesman, 4 October 1994); and Steven Mufson, “China Halts Missile Talks with U.S.: Beijing Delays Visits in Taiwan Visa Feud,” Washington Post, 29 May 1995, 17[A].
41. Arms Control and Disarmament Agency, Office of Public Affairs, “Fact Sheet: The Missile Technology Control Regime (MTCR)” (Washington, D.C.: US Arms Control and Disarmament Agency, Office of Public Affairs, 17 May 1993), 3.
42. Ibid., 4.
43. Ibid., 1.
44. [Department of State,] “Missile Technology Control Regime (MTCR): Equipment and Technology Annex” ([Washington, D.C.: US Department of State, Office of Politico-Military Affairs,] 1 July 1993), introduction.
45. Arms Control and Disarmament Agency, “Fact Sheet: The Missile Technology Control Regime (MTCR),” 3–4.
46. [Department of State, Office of Politico-Military Affairs,] “Summary of the Equipment and Technology Annex” ([Washington, D.C.: US Department of State, Office of Politico-Military Affairs]).
47. [Department of State,] “Missile Technology Control Regime (MTCR): Equipment and Technology Annex,” category 2, item 19.
48. Dr Richard Speier, Department of Defense, Office of the Undersecretary of Defense for Policy, Washington, D.C., interview with authors, 19 November 1993.
49. [Department of State,] “Missile Technology Control Regime (MTCR): Equipment and Technology Annex.” The quoted language—or similar language—is found in category 2, items 3, 9, 10, 11, and 13.
50. China also apparently continues to supply MTCR-restricted ballistic missiles, as well as missile technology and equipment, to a limited number of customers. Recent clients include Pakistan and Iran, the former purchasing complete M-11 ballistic missiles (500-kilogram payload to 300 kilometers) and the latter purchasing missile technology and equipment. For more on these transfers, see R. Jeffrey Smith and David B. Ottaway, “Spy Photos Suggest China Missile Trade,” Washington Post, 3 July 1995, 17[A]; and Barbara Opall, “U.S. Queries China on Iran: Fears Transfer of Missile Technology,” Defense News 10 (19–25 June 1995): 1. For a history of Chinese MTCR commitments and Beijing’s compliance, see Robert Shuey and Shirley A. Kan, “Chinese Missile and Nuclear Proliferation: Issues for Congress,” Congressional Research Service Issue Brief IB92056 (Washington, D.C.: US Library of Congress, 6 July 1995); and Tim McCarthy, “China’s Missile Sales—Few Changes for the Future,” Jane’s Intelligence Review 4 (December 1992): 559–63.
51. The US transferred the 37-kilometer-range Honest John to Taiwan and South Korea and the 130-kilometer Lance ballistic missile to Israel. No US transfers have occurred since the mid-1970s. W. Seth Carus, Ballistic Missiles in the Third World: Threat and Response (New York: Praeger Publishers, 1990), 16–17.
52. France reportedly transferred MD-660 ballistic missiles to Israel in 1968. The MD-660s are said to be the basis for Israel’s 500-kilometer-range Jericho I ballistic missiles. Ibid., 17.
53. System Planning Corporation, Ballistic Missile Proliferation, 83, 86–88; and Jeffrey M. Lenorovitz, “Italian RPV Wins $16-Million Bid for NATO Missile Range Service,” Aviation Week & Space Technology 126, no. 8 (23 February 1987): 52.
54. Some analysts have warned that third world countries will attempt to follow the US and Soviet examples and convert ASCMs to land-attack variants. South Africa was reported to be working on a conversion project in the late 1980s. RPVs can also be converted. India’s first cruise missile will apparently be derived from the country’s Lakshya target drone (200-kilogram payload to nearly 500 kilometers). Similarly, in the late 1980s, Argentina reportedly converted a Mirach 100 RPV into a 900-kilometer-range, multirole platform that is believed to be capable of performing land-attack missions. “India Is Ready to Put Its Unmanned Target Aircraft into Production,” BMD Monitor 9 (22 April 1994): 146; and Carus, Cruise Missile Proliferation, 72–73.
55. Israel’s 80-kilometer-range Popeye air-to-surface missile provides a good example. The missile’s contractor, Rafael, decided to enter the Popeye as its bid in the United Kingdom’s conventionally armed standoff missile procurement. To meet the British range requirement (250–600 kilometers), Rafael made several modifications, including exchanging the Popeye’s solid rocket motor for a turbofan engine. “Missiles Join Line-Up for UK Requirement,” Jane’s Defence Weekly 23, no. 25 (24 June 1995): 12.
56. Carus, Cruise Missile Proliferation, 93.
57. Theresa Hitchens and Barbara Opall, “Fighter Exports Will Leapfrog Domestic Buys,” Defense News 9 (21–27 November 1994): 6; and Barbara Opall, “Upgrade Work Could Top New Sales,” Defense News 8 (9–15 August 1993): 16.
58. Barbara Opall, “Politics Influence International Fighter Decisions,” Defense News 8 (9–15 August 1993): 8; Giovanni de Briganti, “Government Holds Key to Export Push,” Defense News 8 (27 September–3 October 1993): 16; Konstantin Sorokin, “Russia’s ‘New Look’ Arms Sales Strategy,” Arms Control Today 23, no. 8 (October 1993): 9; and US Congress, Office of Technology Assessment, Global Arms Trade: Commerce in Advanced Military Technology and Weapons OTA-ISC-460 (Washington, D.C.: Government Printing Office, June 1991), 21, 48.
59. Robert Holzer, “JPATS Rivals Target World Market for Trainers,” Defense News 8 (9–15 August 1993): 9; Frank Oliveri, “Officials Say U.S. Sparks Interest in UAVs,” Defense News 10 (24–30 July 1995): 10; and Opall, “Upgrade Work,” 16.
60. The following information on Apache is taken from Jean-Paul Philippe, “Matra to Develop APTGD Missile: A New ‘Stealth’ Cruise Missile for France,” Military Technology 19, no. 2 (February 1995): 60–62.
61. Antipiste or antirunway dispenser, which consists of 10 Kriss runway-cratering submunitions, each weighing 51 kilograms.
62. Anti-infrastructure or a penetrating unitary warhead known as Arcole.
63. This version was formerly known variously as the Super Apache, the Apache-C, and the Apache APTGD (i.e., arme de précision tirée à grande distance [long-range, high-accuracy weapon]).
64. More recent accounts of the Apache in technical journals make reference to the enhanced stealth characteristics of the Apache Scalp. See, for example, Philippe, 61.
65. Authors’ interviews (see note 10).
66. The full text of the MTCR’s equipment and technology annex can be found in K. Scott McMahon and Dennis M. Gormley, Controlling the Spread of Land-Attack Cruise Missiles (Marina del Rey, Calif.: American Institute for Strategic Cooperation, January 1995): 92–107. The reference to trading off range and payload is in the annex’s introduction. (See also appendix F.)
67. Giovanni de Briganti and Barbara Opall, “France Spurns Cruise Missile Proliferation Claim by U.S.,” Defense News 10 (16–22 January 1995): 1.
68. Duncan Lennox, ed., Jane’s Strategic Weapon Systems (Alexandria, Va.: Jane’s Information Group, September 1994).
69. Jim Mann, “Russia Is Boosting China’s Arsenal,” Los Angeles Times, 30 November 1992 (LEXIS-NEXIS); and Sorokin, 10.
70. See footnote 50 for details on recent Chinese missile-and-technology transfers and sources on China’s MTCR compliance record.
71. Ingo Raupach, “Russian Air-to-Surface Guided Weapons,” Military Technology 19, no. 5 (May 1995): 11.
72. K. Scott McMahon and Dennis M. Gormley, Russian Cruise Missiles: The Prospects for Control, PSR Report 2526 (Arlington, Va.: Pacific-Sierra Research Corp., May 1995), 26–37; and R. Jeffrey Smith, “U.S. Waives Objection to Russian Missile Technology Sale to Brazil,” Washington Post, 8 June 1995, 23[A].
73. See “China Provides Cash for Israeli Cruise Missile,” Flight International, 17–23 May 1995; and Munson, 21.
74. Craig Covault, “Spanish Ground Attack Missile Design Advances,” Aviation Week & Space Technology 141, no. 21 (21 November 1994): 103.
75. Christy Campbell, “Arab Missile Ban May Sink Tornado Sale,” Sunday Telegraph, 26 February 1995, 7.
76. President William J. Clinton, “Address by the President to the 48th Session of the United Nations General Assembly” (New York: Office of the White House Press Secretary, 27 September 1993); “White House Fact Sheet on Non-Proliferation and Export Control Policy” (Washington, D.C.: The White House, Office of the Press Secretary, 27 September 1993); R. James Woolsey, director of central intelligence, statement prepared for the House Foreign Affairs Committee, Subcommittee on International Security, International Organizations, and Human Rights, 28 July 1993, US Central Intelligence Agency, Washington, D.C.; Lynn E. Davis, undersecretary of state for international security affairs, statement prepared for the House Foreign Affairs Committee, 10 November 1993, US Department of State, Bureau of Public Affairs, Washington, D.C.; and Anthony Lake, assistant to the president for national security affairs, “From Containment to Enlargement,” speech before the Johns Hopkins University School of Advanced International Studies, Washington, D.C., 21 September 1993.