"No one can solve this problem alone, but together we can change things for the better." 

– Setsuko Thurlow
Hiroshima Survivor
June 6, 2016
New START: Good News for Arms Control

April 2, 2010

A version of this article will appear in the May issue of Arms Control Today.

By Steven Pifer

U.S. and Russian negotiators have concluded a new strategic arms agreement to replace the 1991 Strategic Arms Reduction Treaty (START I), which expired in December. Presidents Barack Obama and Dmitry Medvedev are to meet in Prague on April 8 to sign New START.[1] If the Senate and the Russian Duma ratify the treaty, the United States and Russia each will be limited to 1,550 deployed strategic warheads on no more than 800 deployed and non-deployed strategic nuclear delivery vehicles—a steep cut from the levels of START I, which permitted each side 6,000 warheads on 1,600 strategic nuclear delivery vehicles (SNDVs) or launchers.[2] The New START limit on deployed strategic nuclear warheads would be 30 percent below the 2,200 target set by the 2002 Strategic Offensive Reductions Treaty.

The new treaty is supported by a set of effective verification measures, which will provide high confidence that any militarily significant violation would be detected in a timely manner. The verification regime, however, differs from the 1991 agreement. For example, New START is a simpler agreement in several ways, requiring less demanding monitoring measures.

The new treaty is good news. It will reduce Russian and U.S. strategic forces while allowing the United States to maintain a robust nuclear deterrent. It will provide transparency and predictability regarding Russian strategic nuclear forces. Its conclusion demonstrates that Washington and Moscow are fulfilling their nuclear Nonproliferation Treaty (NPT) commitments on the eve of the May NPT review conference; that will strengthen the administration’s hand in seeking to tighten the nonproliferation regime. The treaty should give a boost to the overall U.S.-Russian relationship. Finally, it provides a framework for further reductions in strategic nuclear forces.

This article describes New START, its principal numerical limits, and its monitoring measures. The article also discusses how the treaty will advance U.S. national security interests. The texts of the treaty and protocol have not yet been released; this article draws on the July 2009 Joint Understanding between Obama and Medvedev, U.S. government fact sheets and official briefings, and conversations with knowledgeable U.S. officials.

The Treaty’s Numerical Limits
Under New START, the United States and Russia each must reduce its strategic forces to no more than 1,550 deployed warheads. This limit covers the warheads on deployed intercontinental ballistic missiles (ICBMs) and submarine-launched ballistic missiles (SLBMs), as well as counting each deployed heavy bomber equipped for nuclear armaments as carrying one warhead. The reductions are to be implemented within seven years of the treaty’s entry into force. The treaty has a10-year duration, with a provision for extension of up to five years.

The 1,550 deployed-warhead limit represents a significant reduction for both sides. According to the last data exchange conducted under the START I in July 2009, the United States was attributed with more than 5,900 warheads, though it actually deployed about 2,200. The Russians’ START-attributed number at that time was almost 3,900. Most analysts believe the actual number of deployed Russian strategic warheads is 2,600 to 2,700.[3]

New START will limit each side to 800 “deployed” and “non-deployed” ICBM launchers, SLBM launchers, and heavy bombers equipped for nuclear armaments. Within that 800 limit, each side will be allowed no more than 700 deployed ballistic missile launchers and heavy bombers equipped for nuclear armaments. “Non-deployed” systems include training and test launchers and bombers. Launchers without missiles will count as non-deployed. For example, a Trident ballistic-missile submarine that goes into long-term overhaul has its 24 SLBMs first removed. Under the new treaty, those 24 missile tubes will count as “non-deployed” launchers until the overhaul is finished and the SLBM tubes are reloaded.

These limits accommodate the planned U.S. force structure. Washington intends to implement its reductions largely by downloading warheads from its ICBMs and SLBMs. The United States will be able to reach the 1,550 warhead limit with a relatively modest reduction in its force of ICBMs, SLBMs, and heavy bombers.

What’s Different and Why
In its approach to counting warheads, New START differs from START I. The 1991 treaty used a type-attribution counting rule that assigned each ballistic missile type a number of warheads, regardless of the number of warheads actually carried. For example, the Trident D-5 SLBM counted as carrying eight warheads, even though today most—if not all—carry fewer. The new treaty uses an actual-load counting rule, monitored by on-site inspections. This provision allows the sides to deploy different numbers of warheads on missiles of the same type. (START I permitted some limited downloading; the sides could—and in the U.S. case did—do more downloading, but the type-attribution rule then resulted in an over-count of warhead numbers.)

This provision is of particular interest to the United States. While the U.S. Air Force plans to download its Minuteman III ICBMs to carry one warhead each (as opposed to the Minuteman III’s capacity of three warheads), the U.S. Navy now loads—and plans in the future to load—different numbers of warheads on its Trident D-5 SLBMs. The actual load-counting rule is preferable to a type-attribution rule, because the latter would over-count the number of warheads on the Trident force.
The new treaty maintains something of an attribution rule for heavy bombers equipped for nuclear armaments: each will count as carrying one warhead. An actual-load counting rule would not work for nuclear bombers. Currently it is neither U.S. nor Russian operational practice to maintain nuclear weapons on board bombers—in contrast to the warheads deployed on ICBMs and SLBMs. The New START negotiators thus agreed to a somewhat arbitrary rule of one warhead per heavy bomber equipped for nuclear armaments.

Interestingly, although a different logic led to this counting rule for bombers, the rule is similar to the approach in the 1991 START Treaty. That treaty attributed heavy bombers equipped to carry air-launched cruise missiles (ALCMs) with fewer ALCMs than they could actually carry, and it attributed bombers not equipped to carry ALCMs as carrying only one warhead. That reflected a U.S. push in the negotiations to discount weapons-loads on bombers compared to ballistic missiles, as bombers are slower than missiles, can be recalled, and would have to contend with unconstrained air defenses. This push for preferential treatment for bombers in the belief that they are the least destabilizing leg of the strategic Triad has been a central feature of U.S. arms control policy for some 40 years.

In contrast to START I, the new treaty’s limits will not apply to heavy bombers that are not equipped for nuclear armaments—i.e., bombers that have been converted to solely conventional roles.[4] Likewise, the missile tubes on the four Trident submarines that have been modified to carry canisters with conventional cruise missiles rather than Trident D-5 SLBMs will not count under the launcher limits. These bombers and submarines will not be subject to the new treaty’s numerical limits, though there will be inspection provisions to ensure that they no longer have a nuclear role. That will allow the U.S. Air Force and Navy greater flexibility to deploy conventional systems than would be the case were these launchers to count as SNDVs under the treaty.

These limits—1,550 warheads, 800 total launchers, and 700 deployed launchers—are the new treaty’s sole numerical limits on strategic nuclear systems, reflecting its simplified nature compared to START I. The 1991 agreement complemented its overall ceilings (1,600 launchers capable of carrying up to 6,000 warheads) with a series of nested sublimits for both launchers and warheads. For example, under START I, the Russians were allowed to deploy no more than 154 SS-18 heavy ICBMs, and the sides could deploy no more than 4,900 warheads on ballistic missiles, no more than 1,540 warheads on heavy ICBMs, and no more than 1,100 warheads on mobile ICBMs.

In contrast, New START gives each side complete “freedom to mix,” that is, the ability to choose its own force structure within the overall limits. Several reasons account for this shift. First, given the strategic and political changes of the past 20 years, including the end of the Cold War and collapse of the Soviet Union, neither side felt the need to have a say in determining the shape of the other’s force structure.

Second, as the Russian SS-18 ICBM force continues to age and is retired, Russia will get out of the heavy-ICBM business. The plant that built the SS-18s is located in Dnipropetrovsk, which today is in Ukraine. U.S. negotiators thus had no reason to expend negotiating capital to limit heavy ICBMs as they did 20 years before.

Third, START I’s limit on mobile-ICBM warheads was driven by a concern that Russia would produce and deploy hundreds of these missiles. In fact, although the Russians see mobile ICBMs as an important part of their strategic force, over the past two decades they have produced those ICBMs at a relatively modest rate.

Streamlining Verification
The purpose of verification measures is to give each side high confidence that it can detect a militarily significant violation of the agreement in a timely manner, that is, rapidly enough to respond before the violation jeopardizes its security interests. The new treaty has a full set of verification measures but, as the presidents agreed last July, the negotiators have streamlined the monitoring provisions where possible. The negotiators took START I’s verification measures as a starting point but did away with monitoring provisions that were not required to verify the new treaty. This will be welcomed by both the U.S. and Russian militaries, which have to adjust operational practices to accommodate inspections and other verification measures.

New START includes provisions that will prohibit each side from interfering with the other side’s national technical means of verification (NTM), such as imagery satellites. It also will require the sides to exchange data on their strategic forces and to notify each other of certain changes regarding its forces. While START I used warhead inspections to confirm that missiles did not carry more than the number with which they were attributed, the new treaty, with its actual-load counting rule, will place greater weight on on-site inspections to verify the number of warheads on ballistic missiles and will permit a larger number of on-site inspections for that purpose.

In some areas, the treaty provides for less intrusive measures than did START I. The 1991 agreement required that each side not encrypt the telemetry[5] from its ballistic missile tests (with limited exceptions). The new treaty provides that each side will not encrypt telemetry on up to five ballistic missile tests per year, with the specific tests to be agreed. The telemetry from other tests can be encrypted.

There are several reasons for the change. First, NTM have improved significantly over the past 20 years. While the United States rarely discusses the capabilities of its satellites and other NTM, they can provide more information relevant to monitoring arms control treaties than they could in the past.

As regards telemetry, New START does not contain the kinds of START I provisions that required telemetry access for monitoring purposes. For example, the new treaty has no “new-type” provision specifying what constitutes a new type of missile as opposed to an existing type. START I incorporated that provision to prevent a side from putting multiple warheads on an existing single-warhead ballistic missile. The effect of the provision was to require a side to build a new type of missile to carry multiple warheads. Telemetry access helped to verify whether a missile test was of an existing or new type. The new treaty does not contain such a new type-provision, and its on-site inspection provisions will permit monitoring of warhead numbers.

Senior administration officials, including Secretary of Defense Robert Gates, have said the United States does not need any telemetry access to monitor the new treaty. The administration negotiated some telemetry access, however, bearing in mind the importance attached to telemetry by some in the Senate.

The Russians resisted providing telemetry, largely because the Russian military does not want to provide information that it believes is not needed for monitoring the treaty’s limits. The Russians argued, moreover, that providing all telemetry was unfair, in that Russia is now developing two new missiles—the Bulava SLBM and Topol-M ICBM—while the United States intends to deploy its existing Minuteman III and Trident D-5 missiles for another 20 to 30 years. The Russians felt the United States would learn a lot from the telemetry of their two new missiles, while they would learn relatively little from telemetry from Minuteman and Trident D-5 tests, as they have monitored those tests many times. This reluctance to share information not needed for treaty verification is not just a Russian proclivity; the U.S. military encrypts and intends to continue encrypting the telemetry on missile defense tests.

Another streamlining of the START I verification regime is the end of permanent monitoring at the Votkinsk missile production plant, something that the Bush administration had already agreed to forgo. Mobile ICBMs have always represented a verification challenge. On the one hand, a side needs to be able to count the number of the other side’s mobiles for arms control purposes. But the advantage of mobile ICBMs is their mobility: moving around makes them difficult to locate and increases their survivability, thereby reducing motivations for launching a first strike in a crisis. The solution encapsulated in START I was to allow U.S. inspectors to monitor mobile ICBMs as they left the production facility at Votkinsk. The Obama administration believes that the other measures in the new treaty will give the United States confidence that the number of mobile ICBMs can be monitored without having a presence at Votkinsk. Having observed Russian practice with mobile ICBMs over the past 20 years, the United States now has a better understanding of where they will fit in the overall Russian strategic force. Moreover, Russia likely will deploy fewer launchers than allowed under the new treaty. The Russians in 2009 proposed a launcher limit of just 500, which provides an indication of their planned post-treaty force structure. The new treaty would permit the Russians to deploy many more launchers without exceeding the new treaty’s limits.

Finally, the United States will have a hedge against Russian cheating. The U.S. military will reduce to its 1,550 warhead limit primarily by downloading ballistic missiles and, at least initially, will store many of the extra warheads. It will thus have a significant upload potential.[6] Indeed, in the unlikely event of Russian cheating that the United States deemed a significant threat, the U.S. military could within a matter of months increase the number of its deployed warheads, to perhaps more than 3,000.

Advantages for U.S. National Security
New START offers significant benefits for U.S. national security. First, it will reduce by more than 30 percent the number of strategic nuclear warheads that Russia could target at the United States. With or without the reductions, Americans need not spend a lot of time worrying about a Russian nuclear strike, but reducing the potential threat nonetheless makes America more secure.

Second, the treaty will provide transparency and predictability regarding Russian strategic forces. With New START, the United States will know far more about Russian strategic force deployments than would be the case without the agreement. For example, NTM cannot peer inside a Russian missile and reveal the number of warheads it carries. The treaty’s on-site inspection provisions will allow U.S. inspectors to check precisely that. The treaty will give Washington a good sense of what Russian strategic forces will look like over the coming decade. That kind of predictability will lead to better-informed decisions about the investment that the United States should make in its strategic nuclear forces as opposed to other kinds of military capabilities.

Third, while U.S. strategic forces will be reduced and capped by the treaty, the United States will still maintain a strong and effective nuclear deterrent. In fact, the United States will be able to maintain most of the current force structure: 400-450 Minuteman III ICBMs (each with a single warhead), up to 336 Trident SLBMs on 14 submarines (the missiles will be downloaded), and a small number of B-52 bombers. The B-2 force and some B-52s currently in the nuclear force will be converted to conventional-only roles and thus will fall outside of the treaty’s launcher limits. This force structure is survivable, robust, and agile. Moreover, the conventional capabilities of bombers assigned a conventional-only role and Trident submarines modified to carry conventional cruise missiles will be unconstrained.

Fourth, the new treaty will bolster the basic bargain of the NPT. Under the NPT, the nuclear-weapon states agreed to disarm while the non-nuclear-weapon states gained access to civil nuclear technology but agreed not to acquire nuclear arms. The new agreement demonstrates that the United States and Russia are living up to their part of the deal. That will strengthen Washington’s hand in pressing for a tighter nuclear nonproliferation regime, particularly at the May NPT review conference.7

Fifth, New START should contribute to the administration’s effort to “reset” relations with Russia. Over the past 40 years, when Washington and Moscow have made progress on strategic arms control issues, it has had a positive effect on the broader bilateral relationship. Indeed, one reason that the Obama administration sought at the beginning of 2009 to address certain Russian concerns regarding strategic arms control – such as the Bush administration’s refusal to limit launchers – was to shape a more positive relationship. The administration hopes that this will help secure Moscow’s cooperation on issues such as Afghanistan and Iran. The jury is still out, but the Russians have been more helpful on these questions over the past nine months than they were previously.

Finally, the new treaty sets the stage for further reductions. Obama has made clear that he sees this as the first step in a process. He wishes to continue negotiations with the Russians to lower the number of strategic nuclear weapons and, for the first time, address the issue of tactical (nonstrategic) nuclear weapons. The new treaty incorporates a framework that can accommodate further reductions in nuclear forces.

Possible Criticism

Even before New START was complete, critics voiced concerns over possible weaknesses. Those concerns can be grouped into three categories.

First, some worried that the treaty would limit missile defense. The treaty does acknowledge the interrelationship between offense and defense. That is stating an obvious point: there is a link between the number and type of strategic offensive forces and the number and type of strategic defenses, including missile defense. But the treaty does not limit any current or planned U.S. missile defenses. The Russians may make a unilateral statement to the effect that a significant change in missile defense could lead them to conclude that their supreme interests were jeopardized and thus lead to their withdrawal from the treaty.[8] The Russians have the right to say whatever they want in a unilateral statement, which will have no legal impact on the treaty. With or without such a statement, Russia will have the right to decide unilaterally on withdrawal from the treaty, just as the United States decided to exercise its right to withdraw from the Anti-Ballistic Missile Treaty during the Bush administration.

Second, some may be concerned that the treaty limits conventional warheads on ballistic missiles. The treaty will treat all warheads on ICBMs and SLBMs as if they were nuclear; should the United States or Russia later decide to put some conventional warheads on its strategic ballistic missiles, those warheads would be counted under the 1,550 warhead limit. This is different from how the treaty handles Trident SLBM launchers and heavy bombers that have been converted or modified to a conventional-only role; such systems are not counted under the 800 and 700 launcher limits. Counting all warheads as nuclear obviates the need for a far more intrusive verification regime that would be necessary to distinguish between nuclear and conventional warheads on strategic ballistic missiles. Moreover, at this point, the United States does not even have firm plans to deploy conventional warheads on its Minuteman III or Trident D-5 missiles. Should it decide to do so in the future, the number of conventional warheads is likely to be very small, which would not cut deeply into the permitted overall level of warheads.[9]

Third, some are concerned that the verification regime has been weakened by reducing telemetry access and not allowing permanent monitoring at Votkinsk. The new treaty’s verification provisions will allow each side to have high confidence that the other side is complying with the treaty’s provisions. While it might be nice to have additional provisions – and the insights they would give into Russian strategic forces—such provisions are not necessary for monitoring the treaty’s limits. It is a bit ironic that some of those who assert that the new treaty falls short on verification supported the Strategic Offensive Reductions Treaty signed by Presidents George W. Bush and Vladimir Putin in 2002. That treaty had no verification provisions.

The administration should not pursue, and the Senate should not ratify, nuclear arms control agreements for the sake of arms control. The key questions are: Does the treaty in question advance U.S. national security interests? Does the treaty allow the United States to maintain a reliable and effective nuclear deterrent? Does the treaty contain sufficient verification measures such that any militarily significant violation would be detected in a timely manner so that an appropriate response could be taken?

New START differs significantly from START I. On each of these three key questions, however, the answer for the new agreement is “yes.” It will cut U.S. and Russian strategic nuclear forces to their lowest level in four decades in a stabilizing manner. When the Senate examines the treaty and considers these questions, it should conclude that the treaty serves U.S. national interests and merits ratification.

Steven Pifer is a senior fellow and director of the Arms Control Initiative at the Brookings Institution. A retired Foreign Service officer, he spent much of his career working on arms control and security issues.



1. The total package will include the treaty, protocol, and technical annexes, all of which will be legally binding. As of the end of March, negotiators were still working to complete the annexes. U.S. officials hope that they will be finished by the end of April.
2. START I limited intercontinental ballistic missile launchers (missile silos and transporter-erector-launchers—TELs—for mobile missiles), submarine-launched ballistic missile launchers (missile tubes), and heavy bombers. It limited silos, TELs, and missile tubes rather than the missiles themselves because it was easier to monitor the number of silos, TELs, and tubes, and the sides produced additional missiles for use in tests or as spares.
3. The discrepancy between the U.S. START count of more than 5,900 warheads and the actual count of some 2,200 warheads is explained by three factors. First, most U.S. strategic ballistic missiles have been downloaded so that they carry fewer warheads than the number with which they were attributed by START. Second, a number of U.S. SNDVs—for example, the B-1 bombers—have been converted to conventional-only roles, though they still were counted under START rules. Third, the START count included a number of “phantom” systems; for example, 50 Minuteman III silos sit empty but, because the silos (the launcher counted by START) remained intact, they still counted as if they contained 50 Minuteman III missiles, each attributed with three warheads. Similar factors led to the discrepancy on the Russian side.
4. The START II Treaty, signed in 1993 but which never entered into force, had a provision allowing each side to deploy up to 100 heavy bombers converted to conventional-only roles.
5. Telemetry is the information broadcast by a missile during its flight test that describes how the missile is performing. START I required that telemetry not be encrypted and that, following a flight test, the side conducting the test provide a copy of the telemetric information it had recorded to the other side. Each side thus gained a considerable amount of information about the other side’s ICBM and SLBM performance.
6. Upload refers to the ability to put downloaded warheads back on a ballistic missile. For example, a Trident D-5 downloaded to carry three warheads would still have five slots for additional warheads.
7. A tighter regime would place more obstacles in the path of a future nuclear weapons aspirant. An important example would be securing broader adherence to additional protocols to countries’ safeguards agreements with the International Atomic Energy Agency. The additional protocol gives the agency greater authority to conduct inspections to ensure that a country’s nuclear program complies with the NPT.
8. New START will contain a standard provision allowing each side to withdraw should it consider its supreme interests to be jeopardized by the treaty.
9. The Bush administration plan envisaged arming two Trident D-5 missiles on each of the 14 SLBM-carrying submarines with one conventional warhead each. That would have meant no more than 28 conventional warheads on the Trident force.