START Follow-on Negotiations: Problems and Progress

By Pavel Podvig, Stanford, California

Russian Analytical Digest, No. 66: US-Russian Relations, 20 Oct 2009 (PDF)


The 1991 Strategic Arms Reduction Treaty (START) has remained in force much longer than anyone expected as attempts to iron out a successor treaty failed. Now the presidents of the US and Russia are committed to reducing their forces to the level of 500–1,100 strategic launchers and 1,500–1,650 warheads. Success depends on whether the two sides can agree on counting rules. Observers also fear a gap in verification measures after the START treaty expires in December and before the new one is ratified.

A Hard Act to Follow

It is highly unlikely that anyone present at the signing of the Strategic Arms Reductions Treaty (START) in Moscow in 1991 expected it to stay in force for the full 15-year term specified in the treaty. The term was made long enough to give the United States and the Soviet Union a chance to negotiate a new agreement, which was supposed to supersede START and commit the two nuclear superpowers to deeper reductions of their nuclear forces. The process of negotiating these reductions, however, turned out to be quite difficult, since it raised a host of questions about the nature of the relationship between the two countries, the role of nuclear weapons in that relationship and in national security in general, as well as about the importance of missile defense and the balance of conventional forces. All of these issues have been at the center of the debate about international security and U.S.–Russian relationships in the past twenty years and all of them are in some form present in the current round of arms control talks.

In the first decade after the breakup of the Soviet Union, arms control was hardly the most urgent task of the new Russian leadership, which had to deal with the economic and social cost of the transition to a market economy. The United States also did not assign arms control a high priority, concentrating instead on what appeared to be a rising threat from third countries. Attempts to ratify the START II Treaty, which was signed by the United States and Russia in 1993, were unsuccessful, mostly because of Russia’s concerns about its growing disparity with the United States. U.S. pursuit of national missile defense and Russia’s economic problems only exacerbated the situation. The ratification attempts were finally abandoned in 2002, when the Bush administration withdrew from the ABM Treaty and adopted a policy that emphasized unilateral reductions in nuclear forces and generally rejected the value of arms control treaties.

To replace START II, in May 2002 the United States and Russia signed the Strategic Offensive Reduction Treaty (SORT or Moscow treaty), which ostensibly committed them to further reductions, but in reality was never meant to go beyond what the two sides were planning to do unilaterally. In addition, the Moscow treaty provided no legal framework of its own, relying instead on the one created by START. As a result, the START Treaty is still the only substantive strategic arms control that exists today and when it expires in December 2009, the United States and Russia will have no bilateral arms control and disarmament obligations that would cover their strategic nuclear arsenals.

Getting Serious about Replacing START

The first attempts to negotiate an agreement that would replace START were undertaken during the last years of the Bush administration. However, it is only after the change of administration in Washington that the U.S. and Russian presidents made, at their first meeting in April 2009, a strong commitment to resuming the process of “verifiable reductions in strategic offensive arsenals”, which would begin with negotiating a new arms control treaty, normally known as START follow-on. It is worth noting that at this point the scope of the arms control process is defined fairly narrowly – it is supposed to cover only strategic forces, leaving tactical nuclear weapons outside of the talks, and it will not formally include issues of missile defense.

A more detailed outline of the future treaty was provided at the U.S.–Russian summit held in Moscow in July 2009. The treaty is expected to commit the two countries to reducing their forces to the level of 500– 1,100 strategic launchers and 1,500–1,650 warheads. These ranges reflect the current disagreements and are expected to narrow substantially in the final text. The treaty will not have separate ceilings on components of the strategic triad, so each side would be free to make its own decisions about the structure of its nuclear force. The projected reductions seem to represent substantial progress when compared to the START agreement, which limited the number of strategic launchers and warheads by 1,600 and 6,000 respectively. It also appears to go further than the Moscow treaty, which set a limit of 1,700–2,200 strategic warheads (the treaty did not have a separate limit on launchers). In reality, however, the reductions will be much more modest, especially on the U.S. side, since the difference in numbers reflects a change of definitions rather than actual reductions.

A Numbers Game

In the START treaty, strategic launchers and warheads are counted by a set of rules designed to ensure that neither side has the capability of quickly reconstituting its strategic potential. These rules reflected the degree of distrust that existed between the United States and the Soviet Union at the time, but they did assure that the numbers in the treaty closely corresponded to the maximum number of warheads that a country could deploy. On the other hand, these rules make deep reductions difficult, for they, in most cases, require physical elimination of delivery systems. After achieving the START limits, the United States and Russia continued reducing their arsenals – the process that was codified in the Moscow treaty – but they handled it differently, creating a disparity in what is known as “upload potential,” the ability to reconstitute the force that the START Treaty sought to limit.

If the new treaty is to limit the strategic forces at the level of 1,500–1,675 warheads agreed upon in July 2009, it will have to relax the strict START counting rules and rely instead on some version of the U.S. definition of “operationally deployed nuclear warheads” that was used in the context of the Moscow treaty (Russia has not formally accepted that definition yet). According to the January 2009 START data exchange, the United States had 5,576 strategic nuclear warheads associated with 1,198 delivery vehicles. The actual number of warheads that were operationally deployed was substantially lower – it was estimated to be around 2,200 in the beginning of 2009. For Russia, the difference is smaller, but it exists nevertheless – it reported having 3,909 warheads associated with 814 launchers, although the number of operationally deployed warheads is believed to be about 2,800. More importantly, in the United States, most of the difference between the START count and the operationally deployed warhead count is due to easily reversible measures, such as removal of some warheads from ballistic missiles. For example, most Minuteman III ICBMs, which are capable of carrying three warheads, are currently deployed with only one; Trident II SLBMs are deployed with four warheads, although they are capable of carrying eight. Moreover, the United States as a matter of policy keeps reserve warheads specifically to have that reconstitution option. In Russia, the difference was created primarily by the slow pace of eliminating old delivery systems, so it has no reconstitution capability to speak of.

While it is unclear if the U.S. “upload potential” has any practical significance, Russia has been raising this issue at negotiations for many years, insisting that any new treaty should include measures that would limit the U.S. reconstitution capability. One way of dealing with it would be to preserve the START treaty counting rules and requirements. However, as noted above, since the treaty is expected to set a limit of about 1,500 warheads, it will definitely limit only operationally deployed warheads, meaning that the START counting rules would have to be abandoned. As a way of satisfying Russia’s demand for provisions that would limit upload potential, the new treaty will include a separate limit on the number of strategic launchers.

Although the concept of a limit on launchers is now accepted by both sides, in order for this limit to be meaningful, it would have to be set relatively low. This is the reason why the disagreement about the number of launchers, as reflected in the July presidential statement, is especially strong. Russia insisted on setting that limit at the lower end of the range – at about 500 launchers, while the United States would like to keep that number at about 1,100. Each side understandably tried to have a limit that would correspond to its existing plans – Russia has about 600 operational launchers and its modernization program envisages a force of about 400–450 delivery vehicles in a decade or so. The U.S. force currently includes almost exactly 1,100 deployed launchers, which the United States would certainly try to preserve, if only to have an option to convert them for non-nuclear missions (some of them have been converted already).

It is most likely that the number in the treaty will be a compromise that would set the limit lower than 1,100, but would allow the United States to exclude some of the launchers from the treaty count by demonstrating that they have been converted and no longer have nuclear missions. This might allow the United States to exclude 56 B-1 bombers that are still counted as deployed in START, about 50 B-52 bombers, 96 SLBM launchers on Trident submarines, and maybe some other systems as well. It is possible that the final treaty will set a limit of 800–900 launchers on each side. While this would probably not fully address all concerns about the “upload potential”, this limit is likely to be accepted by Russia.

Another issue that has been discussed during this round of negotiations is closely linked to the possible conversion of strategic launchers from nuclear to conventional missions. Russia has expressed two separate sets of concerns related to this. One is that the use of strategic launchers with conventional payloads, e.g. as planned in the U.S. Conventional Trident Missile program, could lead to a misunderstanding and an accident that could prompt a nuclear strike. Another, more long-term concern often expressed by the Russian military is that the U.S. high-precision conventional strike capability could at some point pose a threat to Russia’s strategic forces. Accordingly, Russia would want to place some limits on the U.S. ability to convert the existing strategic delivery systems for conventional missions. It highly unlikely that these issues could be adequately addressed in the strategic arms control negotiations, but the treaty will probably include provisions that would allow some additional transparency measures to apply to former strategic delivery systems.


Although Russia and the United States are seeking to relax the START treaty counting rules, in general they seem to be committed to preserving most of its transparency and verification provisions. This may well be the most important element of the future agreement, for it would maintain the legal and institutional framework established by the START treaty. These arrangements allowed the two countries to preserve an important communication channel during the last two decades and greatly reduced the chances for misunderstandings similar to those that happened in the areas which did not have a similar supporting infrastructure – tactical nuclear weapons and nuclear testing among them. Given the political support that the negotiations have received from the presidents and the progress that has been made so far, there is little doubt that the new treaty will be signed before the START Treaty expires on December 5, 2009. However, it almost certainly will not enter into force by then, so the United States and Russia would have to find a way of dealing with the resulting gap in the arms control and disarmament regime. One possible option that is being considered by the negotiators is a joint commitment that would be made by the presidents not to take actions that would undermine the goal of the treaty. Since both countries have enough flexibility in their nuclear planning, they will have no difficulty fulfilling their obligations regarding reductions to their arsenals. It is not clear, however, if an executive agreement would be sufficient to ensure continuity in transparency and verification regimes – data exchanges and inspections may prove impossible without a formal treaty. This may not be a problem if the new treaty quickly enters into force, but the ratification process may take a significant amount of time – as long as one year.

A better alternative to the executive agreement would be an extension of the START treaty for five years, which is allowed by Article XVII of the treaty. This extension would be relatively simple to make since it would not require ratification by the legislature of either country. At the same time, the new treaty would automatically supersede START as soon as its ratification is complete. This course of action would provide the best way of avoiding the gap between two arms control agreements and preserving the structure of transparency and accountability established by START.

About the Author

Pavel Podvig is a researcher at the Center for International Security and Cooperation at Stanford University.