The Declining Ballistic Missile Threat

Joseph Cirincione
Director Non-Proliferation Project
Carnegie Endowment for International Peace
Washington, D.C.

presented at American Association for the Advancement of Science
Annual Meeting and Science Innovation Exposition
Boston, Massachusetts
February 18, 2002


One of the most important proliferation debates of the past ten years has concerned the assessment of the ballistic missile threat and efforts to deploy missile defenses. When the end of the Cold War largely eliminated the likelihood (if not the capability) of a global thermonuclear war, policymakers turned their attention to the very real danger that weapons of mass destruction could be used in smaller, but still horrifically deadly, numbers. Ballistic missiles garnered the lion's share of attention, though they are only one--and perhaps the most difficult--method of delivery of these weapons.

The Proliferation Threat

Many experts and officials view ballistic missiles as a particularly menacing, difficult to detect, and rapidly proliferating technology. Several threat assessments and reports followed the lead of the 1998 study by the Commission to Assess the Ballistic Missile Threat to the United States (known as the Rumsfeld Commission for its chair, Donald Rumsfeld):

"With the external help now readily available, a nation with a well-developed, Scud-based ballistic missile infrastructure would be able to achieve first flight of a long range missile, up to and including intercontinental ballistic missile (ICBM) range (greater than 5,500 km), within about five years of deciding to do so. During several of those years the U.S. might not be aware that such a decision had been made."

The Commission identified two countries as particularly dangerous: North Korea and Iran. These nations apparently had made a decision to achieve first flight of an ICBM:

"The extraordinary level of resources North Korea and Iran are now devoting to developing their own ballistic missile capabilities poses a substantial and immediate danger to the U.S., its vital interests and its allies. While these nations' missile programs may presently be aimed primarily at regional adversaries, they inevitably and inescapably engage the vital interests of the U.S. as well. . . Each of these nations places a high priority on threatening U.S. territory, and each is even now pursuing advanced ballistic missile capabilities to pose a direct threat to U.S. territory."

The August 31, 1998 North Korean test of a Taepodong-1 missile/space launch vehicle (SLV) appeared to lend credence to these warnings. The Taepodong-1 failed in its attempt to launch a satellite into orbit and flew only 1320 km, but it had an enormous impact internationally due to the unexpected use of a third stage on the rocket. As a result of this test and the changing strategic environment, the 1999 National Intelligence Estimate (NIE) included countries other than Russia and China as ballistic missile threats to the United States for the first time.

Although neither the North Korean or Iranian missile programs appear to have made significant progress since 1998, the most recent NIE submitted in December 2001 and released in unclassified form in January 2002 concluded that before 2015 the United States

"…most likely will face ICBM threats from North Korea and Iran, and possibly Iraq - barring significant changes in their political orientations - in addition to the strategic forces of Russia and China. One agency assesses that the United States is unlikely to face an ICBM threat from Iran before 2015.

The threats to the US homeland, nevertheless, will consist of dramatically fewer warheads than today owing to significant reductions in Russian strategic forces. China has been modernizing its long-range strategic missile force since the 1980s. . .by 2015, the total number of Chinese strategic warheads will rise several-fold, though it will remain still well below the number of Russian or US forces"

The report states "The probability that a missile with a weapon of mass destruction will be used against US forces or interests in higher today than during most of the Cold War and it will continue to grow as the capabilities of potential adversaries mature." (emphasis in original) However, the assessment notes that

"U.S. territory is more likely to be attacked with [chemical, biological, radiological and nuclear] materials from nonmissile delivery means-most likely from terrorists-than by missiles, primarily because nonmissile delivery means are less costly, easier to acquire, and more reliable and accurate. They can also be used without attribution." (emphasis added)

The report also cautioned:

"Our assessments of future missile developments are inexact and subjective because they are based on often fragmentary information….States with emerging missile programs inevitably will run into problems that will delay and frustrate their desired development timelines. The impact of these problems increases with the lack of maturity of the program and depends on the level of foreign assistance. Most emerging missile states are highly dependent on foreign assistance at this stage of their development efforts and disturbance of the technology an information flow to their programs will have discernible short-term effects."

These cautions and caveats are often brushed aside in the political discussions and program decisions concerning the ballistic missile threat. For example, the Quadrennial Defense Review presented to Congress from the Department of Defense on October 1, 2001, argues that "In particular, the pace and scale of recent ballistic missile proliferation has exceeded earlier intelligence estimates and suggests these challenges may grow at a faster pace than previously expected."

Director of Central Intelligence George Tenet went beyond the official intelligence assessment and told the Senate select Committee on Intelligence on February 6, "The proliferation of ICBM and cruise missile designs and technology has raised the threat to the US from WMD delivery systems to a critical threshold."

Whether one thinks the ballistic missile threat is greater than the terrorist threat or not, and somewhat independent of one's views on the feasibility of ballistic missile defenses, there is a general sense that the threat is increasing. But is this true? More precisely, is the risk to U.S. cities from ballistic missiles attack greater now than in the past and will it get worse?

Changes from Earlier Assessments

Ten years ago, the consensus was that the threat was not increasing. The 1993 NIE ("Prospects for the Worldwide Development of Ballistic Missile Threats to the Continental United States," NIE 93-17) concluded:

"Only China and the CIS [Commonwealth of Independent States] strategic forces in several states of the former Soviet Union currently have the capability to strike the continental United States (CONUS) with land-based ballistic missiles. Analysis of available information shows the probability is low that any other country will acquire this capability during the next 15 years."

The 1995 NIE ("Emerging Missile Threats to North America during the Next 15 Years," NIE 95-19), as summarized publicly by Richard Cooper, Chairman of the National Intelligence Council, found:

"Nearly a dozen countries other than Russia and China have ballistic missile development programs. In the view of the Intelligence Community, these programs are to serve regional goals. Making the change from a short or medium range missile-that may pose a threat to US troops located abroad-to a long range ICBM capable of threatening our citizens at home, is a major technological leap…. The Intelligence Community judges that in the next 15 years no country other than the major declared nuclear powers will develop a ballistic missile that could threaten the continuous 48 states or Canada."

Several leading members of Congress harshly attacked the 1995 and 1993 estimates. In December 1996, a congressionally mandated panel headed by former Bush administration CIA Director Robert Gates reviewed the 1995 NIE. They agreed that the continental United States was unlikely to face an ICBM threat from a third world country before 2010 "even taking into account the acquisition of foreign hardware and technical assistance, and that case is even stronger than was presented in the estimate."

Congress then mandated another panel that finally gave them a different answer. Ever since the 1998 report from the Rumsfeld Commission asserted, somewhat hysterically, that a new nation could plausibly field an ICBM "with little or no warning," intelligence analysts have struggled to cover all possibilities, while still preserving some predictive net assessment. This conflict is evident in the introduction to the 1999 NIE, which notes a dissenting opinion from one of the intelligence agencies involved in producing the consensus report:

"Some analysts believe that the prominence given to missiles countries 'could' develop gives more credence than is warranted to developments that may prove implausible."

This "could" issue is perhaps the most striking difference between the new NIEs and those published in 1993 and 1995. "Could" is a highly ambiguous word. For some it means "remotely possible;" for others it means "will."

The shift to the "could standard" represents one of the three major changes made to the assessment methodology from previous assessments. The other two shifts are:

  • substantially reducing the range of missiles considered serious threats by shifting from threats to the 48 continental states to threats to any part of the land mass of the 50 states; and,
  • changing the timeline from when a country would first deploy a long-range missile to when a country could first test a long-range missile. Significantly, the new NIEs do not require that is be a successful test

The shift of potential US targets represents a range change of some 5,000 kilometers (the distance from Seattle to the western-most tip of the Aleutian Island chain in Alaska). It essentially means that a medium-range ballistic missile, such as the Taepodong-1, could be considered the same threat as an intercontinental-range missile. The Taepodong-1 tested on August 31, 1998, impacted 1320 kilometers from its launch point, and tried but failed to put a small satellite into orbit. This missile does not have the range to strike any part of the United States with a large payload (for example, a nuclear warhead), though it might be able to strike the western most parts of Alaska and Hawaii with a very small payload. The Taepodong-2 is theoretically judged to have a range of 4,000 to 6000 kilometers, allowing it to strike parts of Alaska and Hawaii. A three-stage Taepodong-2 could have a longer range.

The timeline shift represents a difference of five years (what previous estimates said was the difference between first test and likely deployment). "With shorter flight test programs-perhaps only one test-and potentially simple deployment schemes, the 1999 NIE concluded, the time between the initial flight test and the availability of a missile for military use is likely to be shortened." The Indian experience with the Agni missile provides some indication that the original standard may be the more accurate. The Agni program began in the mid-1980s. An Agni-1 missile was flight tested in February 1994 and a medium-range, 2,000 -km version, the Agni-2, was tested in April 1999. Despite Indian declarations of intent to deploy and substantial financial and scientific resources devoted to the program, the missile has yet to enter production.

These three changes account for almost all of the differences between the 1999 and 2001 NIEs and earlier estimates. Thus, these new estimates, rather than representing some new, dramatic development in the ballistic missile threat, represents a lowering of the standards for judging the threat. These NIE may lead some observers to conclude that there has been a significant technological leap forward in Third World missile systems, when, in fact there has been only incremental developments in programs well known to analysts for years.

To compare more completely today's ballistic missile threats to those of the past and to perform a net assessment of the global ballistic missile threat, it is useful to evaluate the threat in its component parts.

Global Ballistic Missile Arsenals

The blurring of short, medium, intermediate, and intercontinental ranges for the world's missile inventory often results in the misinterpretation of the oft-quoted assessment that over 25 nations possess ballistic missiles. This statement is true, but only the United States, China, and Russia possess the ability to launch nuclear warheads on land-based intercontinental missiles. This has not changed since Russia and China deployed their first ICBMs in 1959 and 1981 respectively.

  • Analysis of global ballistic missile arsenals shows that there are far fewer ICBMs, and long-range submarine-launched ballistic missiles (SLBMs) in the world today than there were during the Cold War.
  • The number of intermediate-range ballistic missiles (IRBMs) has decreased in the past 15 years by an order of magnitude.
  • The overall number of medium-range ballistic missiles (MRBMs) has also decreased, though five new countries have developed or acquired MRBMs since 1989.
  • The number of countries trying to develop ballistic missiles has also decreased and the nations still attempting to do so are poorer and less technologically advanced than were the nations 15 years ago.
  • The number of countries with short-range ballistic missiles (SRBMs) has remained fairly static over the past 20 years and is now decreasing as aging inventories are retired.
  • There are also the same nations potentially hostile to the United States trying to develop MRBMs as there were 15 years ago (1980s: China, Iraq, Libya, Soviet Union; 2002: China, Iran, Iraq, North Korea).
  • The damage from a ballistic missile attack on the United States today with one or two warheads is also lower by orders of magnitude than fifteen years ago when thousands of warheads would have destroyed the country, even the planet.

Thus, the most accurate way to summarize existing global ballistic missile capabilities is: (1) There is a widespread capability to launch short-range missiles. (2) There is a slowly growing, but still limited, capability to launch medium-range missiles. (3) Most importantly, there is a decreasing number of long-range missiles from the levels of the Cold War and this number will continue to decline dramatically over the next fifteen years. (4) There is some possibility that one or two new nations could acquire a very limited capability to launch long-range missiles over the next two decades.

Long-Range Ballistic Missiles

Force reductions in U.S. and Russian arsenals have dramatically decreased the number of long-range ballistic missiles (missiles with a range of greater than 5,500 km) in the world from their Cold War levels.

Decreases. In 1987, the Soviet Union deployed 2,380 long-range missiles in its combined ICBM and SLBM arsenals. The United States deployed 1,640 long-range missiles. At the end of 2001, Russia had 1,022 long-range missiles and the U.S. had 983 long-range missiles.

Increases. France has reduced its nuclear arsenal overall, but now has 48 long-range SLBMs that it did not have in 1987. Similarly, the United Kingdom has reduced its arsenal but now fields 58 long-range Trident SLBMs that it did not have in 1987.

Status Quo. During this period China has maintained a force of about 20 Dong Feng-5 ICBMs. No other country has developed an ICBM or long-range SLBM during this time period.

Net Decrease. By the beginning of 2002, the total number of long-range ballistic missiles in the world (including those of the United States, the United Kingdom and France) had decreased 47 per cent to 2,131 from the 4,040 deployed in 1987. More significantly the total number of long-range missile potentially threatening the United States had declined from 2400 fielded by the Soviet Union and China in 1987 to 1042 fielded today by Russia and China. This is a decrease of almost 57 percent.

Intermediate- Range Ballistic Missiles

Intermediate Range Ballistic Missile arsenals have undergone even more dramatic reductions. The Intermediate-Range Nuclear Forces (INF) Treaty eliminated this entire class of missiles (with ranges from 3,000 to 5,500 km) from the Soviet/Russian arsenal over a three-year period.

Decreases. Final INF inspections took place on May 31, 2001, verifying the destruction of 660 intermediate-range Soviet ballistic missiles. France deactivated its limited arsenal of 18 IRBMs in 1996 and has since destroyed them. The United States did not then and does not now field IRBMs.

Status Quo. China has maintained about 20 DF-4 missiles of this range. No other nation has developed an IRBM.

Net Decrease. The decrease from 680 IRBMS potentially threatening the United States, its forces or allies in the 1980s to 20 today, represents a 97 per cent reduction from Cold War levels.

Medium-Range Ballistic Missiles

The broad scope of the INF Treaty also covered medium-range ballistic missiles (MRBMs). Thus, the treaty resulted in the elimination of this class of missiles (with a ranges from1000 to 3000 km) from Soviet/Russian and U.S. ballistic missile arsenals.

Decreases. A total of 149 Russian SS-4 and 234 U.S. Pershing II missiles were destroyed under the INF treaty.

Increases Geographically. The most significant proliferation threat comes from the slow but steady increase in the number of states possessing medium-range ballistic missiles, even as Russia and the United States eliminated their arsenals. This development has attracted a great amount of attention and is often cited as evidence of a larger proliferation threat. China, India, Iran, Israel, Pakistan, North Korea, and Saudi Arabia now possess MRBMs. China also possesses a medium-range SLBM capability, though its operational status is in question. Only India, Iran, Pakistan, and North Korea have developed or obtained their missiles since 1989, and Iran and Pakistan base their missiles primarily on assistance or technology received from North Korea.

Net Decrease Numerically. Numerically speaking, even though MRBMs are now in the hands of more countries than in 1987, the total number of MRBMs in existence in 2002 is smaller than the 483 MRBMs in the combined U.S., Russian, and Chinese forces in 1987. Since then, Israel is believed to have deployed 50 operational Jericho II MRBMs while Saudi Arabia has 40 CSS-2 MRBMs that it purchased from China. North Korea is believed to have deployed 10 Nodong MRBMs but has possibly produced up to 100 missiles of this type. MRBMs in India, Iran, Pakistan and North Korea's Taepodong are still in operational testing. Assuming each of these countries could deploy one to five missiles in a crisis during the next five years, the global total of MRBMs today is no more than 310 and likely as low as 200. This represents a 36 and 58 per cent decrease, respectively, in global MRBM arsenals from the 1987 level. In terms of missiles potentially threatening the United States, the threat has gone from 249 Chinese and Soviet missiles in 1987 to 100 Chinese, plus as many as 100 North Korean Nodongs and one or two Taepodong missiles and a few Iranian Shahab-3 missiles today or a 20 percent decrease in threat systems.

Short-Range Ballistic Missiles

Aging Scud Arsenals. In addition to the five recognized nuclear-weapon states, there are 29 nations with ballistic missiles. Of these 29 nations, the vast majority have only missiles with a range under 1000 km. Twenty of the 29 nations only have Scud-B or similar missiles with an approximate range of 300 km or less. These missiles, many of which are quite old and have not been well maintained, are declining in military utility.

Number of Countries with Ballistic Missile Programs

Another factor by which proliferation can be measured is the number of states with missile development programs. The number of countries with ballistic missile development programs has also decreased from the number of countries pursuing missile programs during the Cold War. In addition to the five recognized nuclear-weapon states, countries such as Israel, India, Brazil, Argentina, Egypt, South Africa, Iraq and Libya had programs to develop long-range or medium-range missiles in 1987. By 2002, Brazil, Argentina, Egypt and South Africa had abandoned their programs, and Libya's remains largely defunct.

Today, the nations pursuing long-range missile development programs today are also smaller, poorer and less technologically advanced than were the nations with missile programs 15 years ago. U.S. threat assessments such as recent National Intelligence Assessments on the Ballistic Missile Threat (NIEs) note that Iran and North Korea currently possess active programs. With UN inspections (now suspended) and sanctions against it, Iraq faces many hurdles to reviving its program. Syria and South Korea have active short-range ballistic missile programs, but have not yet demonstrated interest in or the capability to produce MRBMs. Egypt and Libya may be pursuing purchases of MRBMs (from North Korea) but do not appear to have active development programs. Thus, even with the inclusion of India and Pakistan (countries not generally included in U.S. threat assessments because of friendly diplomatic relations with the United States) the recent NIEs highlight the limited nature of the missile proliferation threat, one that is confined to a few countries whose political evolution will be a determining factor in whether they remain threats to global security.

Nor have these programs advanced as quickly as predicted by the worst-case assessments that came to dominate U.S. policy on missile proliferation and missile defense. The 1998 Rumsfeld Commission report asserted that "Scud-based ballistic missile infrastructures would be able to achieve first flight of a long-range missile, up to and including intercontinental ballistic missile (ICBM) ranges, within about five years of deciding to do so." The report concluded that Iran and North Korea had decided to do so: The extraordinary level of resources North Korea and Iran are now devoting to developing their on ballistic missile capabilities poses a substantial and immediate danger to the U.S." The commissioner said, "Each of these nations places a high priority on threatening U.S. territory, and each is even now pursuing advanced ballistic missile capabilities to pose a direct threat to U.S. territory."


However, four years later, neither country has achieved first flight of an ICBM. Iran's program appears to be in disarray after its "Shahab-3" (a missile largely based on and perhaps nothing more than a North Korean Nodong missile) blew up in two of its three tests in 1998 and 2000. The 2001 NIE notes "All agencies agree that Iran could attempt to launch an ICBM/SLV about mid-decade, although most agencies believe Iran is likely to take until the last half of the decade to dos so. One agency further judges that Iran is unlikely to achieve a successful test of an ICBM before 2015."

North Korea

North Korea has had only two missile tests in the past ten years, one of a Nodong in 1993 and one of a Taepadong-1 in 1998. North Korea is observing a self-declared moratorium on missile tests through 2003.

North Korea is the most serious case of a potential new threat. It may be able to test a Taepodong-2 missile that could approach ICBM ranges, but it would require a third stage to be able to deliver a payload to the continental United States. The capability, reliability, and payload of such a missile are highly speculative.

Unclassified photos of the North Korean test facilities revealed what many analysts have long concluded: the missile program is primitive by world standards, not capable of sustaining multiple launches of missiles, and of limited military utility. North Korea, increasingly eager to open normal trade relations with the West, seems be willing to suspend a dubious program for real material gain.

The recent NIEs and the Rumsfeld Commission assume an optimistic and fairly straightforward path for North Korea to scale up their existing missiles to true intercontinental range. Only the United States, Russia and China have been able to build missile in this range thus far. One cannot completely rule out the possibility that North Korea could develop a missile with enough range to reach the continental United States within ten years. However, the obstacles are formidable. As previous intelligence estimates have reported, the Taepodong -2, -3 or -4, would have to make remarkable progress in propulsion, guidance and reentry vehicle technology. Moreover, as the size of the missile increases, it requires a difficult manufacturing and engineering shift from the steel bodies employed by Scuds to low-weight, high-strength alloys. Finally, North Korea would have to manufacture a nuclear warhead small enough and sturdy enough to fit on the tip of the missile. There is no evidence that North Korea has mastered these techniques, only speculation that it might be possible.


Iraq is the least likely state to develop a long-range missile over the next 15 years. United Nations inspectors destroyed almost all of Iraq's existing inventory of Scuds and it is prohibited from producing or developing missile with ranges greater than 150 Km. Iraq has been able to maintain the infrastructure and expertise necessary to develop longer range missiles, however. The 2001 NIE concludes that the UN-related sanctions and prohibitions "plausibly will constrain Iraq during the entire period of this Estimate." Further:

"For the first several years after relief from UN prohibitions, Iraq probably will strive to reestablish its SRBM inventory to pre-Gulf war numbers, continue developing and deploying solid-propellant systems, and pursue MRBMs to keep pace with its neighbors. Once it regional security concerns are being addressed, Iraq may pursue a first-generation ICBM/SLV.

"Although Iraq could attempt before 2015 to test a rudimentary long-range missile based on its failed Al-Abid SLV, such a missile almost certainly would fail. Iraq is unlikely to make such an attempt."

As illustrated by the above assessment of Iraq's motivations, missile proliferation remains primarily a regional problem, though with global implications. In South Asia and the Middle East, strategic interest and political dynamics have fueled continued development of ballistic missile technology as both a means of gaining international prestige as well as obtaining a strategic advantage vis-à-vis regional rivals. Though relatively limited, this proliferation and the transfer of ballistic missile technology originating in North Korea and China does continue to destabilize regional, and therefore global, security.It is worth noting that the Joint Chiefs of Staff rejected the conclusions of the Rumsfeld Commission in 1998. Chairman of the Joint Chiefs General Henry Shelton wrote:

"While the Chiefs and I, along with the Intelligence Community, agree with many of the Commission's findings, we have some different perspectives on the likely developmental timelines and associated warning times.

"After carefully considering the portions of the report available to us, we remain confident that the Intelligence Community can provide the necessary warning of the indigenous development and deployment by a rogue state of an ICBM threat to the United States.

"For example:

  • We believe that North Korea continues moving closer to the initiation of a Taepo-Dong I Medium Range Ballistic Missile (MRBM) testing program. That program has been predicted and considered in the current examination.
  • The Commission points out that through unconventional, high-risk development programs and foreign assistance, rogue nations could acquire an ICBM capability in a short time, and that the Intelligence Community may not detect it. We view this as an unlikely development.
  • I would also point out that these rogue nations currently pose a threat to the United States, including a threat by weapons of mass destruction, through unconventional, terrorist-style delivery means. The Chiefs and I believe all these threats must be addressed consistent with a balanced judgement of risks and resources.

"Based on these considerations, we reaffirm our support for the current NMD policy and deployment readiness program. Our program represents an unprecedented level of effort to address the likely emergence of a rogue ICBM threat."

The Chiefs' judgments were overturned by political decisions, but in hindsight their assessment and the intelligence estimates provided in 1993 and 1995 have proven more sound than the assessments subsequently produced by the intelligence agencies.

In short, the ballistic missile threat is confined, limited and changing relatively slowly.