Chinese force developments
China has a very modest capacity to mount nuclear attacks on the US. It has a relatively small number of nuclear weapons. Estimates differ from as few as 200 warheads  to around 400. Of these there are only about 20 intercontinental-range missiles capable of reaching targets in the continental US, and one ballistic missile submarine of doubtful operational value.  Most of its current ICBMs are liquid-fuelled, which means they take many hours to prepare for launch, and are tied to elaborate launch sites. All this makes them easy to destroy in a surprise first-strike attack. China has however evidently been satisfied that secrecy about the precise number of missiles it has, and the uncertainty that afflicts any military operation, has meant that the US would be deterred from trying to disarm China for fear that one or two missiles would survive, still able to inflict massive casualties and damage to American cities.
China started to modernize its rather crude intercontinental nuclear forces over 20 years ago. It has developed a new generation of solid-fuelled intercontinental-range missiles, the DF 31 and DF 31A, but has been slow to field them. The first of these new missiles may only have become operational over the last year or so.  They are more accurate, more mobile, can be launched much more quickly, and hence are more 'survivable'. China is also developing a new and better submarine-launched missile, raising the possibility that it will eventually be able to deploy operationally effective submarine-based forces which would be much harder to find and attack than any land-based missile. Finally China has probably developed but not deployed the capacity to put several warheads on each missile, thus increasing the range of targets it can hit and complicating missile defense efforts.
Nonetheless these developments do not give China much ground for confidence about the future of its minimum deterrence posture in the light of the evolution of US nuclear forces. Many scholars - including Chinese observers - have noted that even with their new missiles, the combination of America's highly accurate offensive forces and expanding national missile defenses will see it lose the minimum deterrent capability that it believes it has enjoyed hitherto.  Beijing's concerns probably focus most clearly on the way US nuclear primacy over China would affect the dynamics of a crisis over Taiwan. They may fear that the US, confident that it could defeat any Chinese counterattack, could credibly threaten a nuclear strike on China to force China to desist from conventional military operations against Taiwan. This would neutralize both China's most powerful sanction against a Taiwanese declaration of independence, and its growing conventional capacity to limit US conventional military options in support of Taiwan. For China there is a clear historical precedent for such action by America, from the era before China acquired its own nuclear weapons. In the confrontation over Quemoy and Matsu in the Taiwan Strait in the 1954-5, the US used explicit threats of nuclear attack against China to deter Chinese conventional operations to seize these disputed islands.  So for China the stakes are very high. It seems most unlikely that China will simply allow the US to achieve nuclear primacy and to neutralize China's nuclear forces without vigorous efforts to counter American measures. It has been clear for some time that China has several options to respond to American nuclear force developments. The first is to accelerate production and deployment of its new, more survivable missiles, and arm them with multiple warheads. What counts for China is numbers. The more missiles and warheads it has, the less confident the US can be that it will be able to destroy them all with the combination of its offensive strike and defensive shield. The arithmetic may be especially tough on US defensive systems: it could be easier for China to double the number of warheads it can launch towards the US than for the US to double the capacity of its defenses.  Other sensible responses to American nuclear-force developments would include the expansion of China's fledgling submarine-based missile capability, the development of nuclear cruise-missile attack options, perhaps from covert platforms like commercial shipping, and the covert delivery of nuclear weapons direct into US ports aboard ships. China might also move from its traditional posture of only using its nuclear forces in retaliation for a nuclear attack to a more risky 'launch on warning' or 'launch under attack' posture. No doubt China is looking at all these options.
The problem of course is that any of these measures would attract a US response in turn. There is in fact a lively debate in the US about whether China's current nuclear-force developments do not suggest that it is already moving away from a minimum deterrence posture towards something more ambitious. Debates in China's strategic community about nuclear strategy have become more active, and some US scholars have queried whether China's evolving nuclear capabilities might be intended to underpin a shift away from minimum deterrence to a more active posture, more threatening to US interests and to America itself. 
There is a clear risk that China and the US will be drawn into an escalating and mutually reinforcing cycle of responses to one another's strategic nuclear developments. To preserve minimum deterrence in the face of US offensive and defensive developments, China is likely to expand the number, sophistication and variety of its offensive forces, and modify its nuclear doctrine. The US in turn may well interpret those measures as an attempt by China to challenge US strategic primacy, and to strengthen a nuclear deterrent screen behind which it can apply conventional military pressure in regional contingencies like Taiwan. It is therefore likely that the US would in turn respond by further enhancing its offensive forces and missile defenses to stay ahead of China's countermeasures. As US forces grew, China would do more to counteract them, and so both sides would seem likely to be drawn into an arms race. This poses very significant risks.
First we need to consider whether increasing strategic nuclear competition between the US and China raises the danger that a regional clash between them would escalate into a nuclear exchange. Taiwan seems the most plausible spark for a US-China conflict, but other scenarios are also possible, including clashes between China and Japan or between Japan and Korea. It is clearly possible that such a conflict could escalate to a nuclear exchange,  and nuclear strategic competition between them may make that more likely. One scary scenario is that China, under sustained conventional attack from US forces based in Guam, including, possibly, conventional precision-strike operations directed against China's nuclear forces, might decide to risk a nuclear attack on Guam. Beijing might fear that it risked losing its nuclear forces to US conventional strikes, and might reason that because Guam is so remote from other US territory, and is primarily a military facility, a strike on Guam might not attract a US nuclear counter-strike on China's territory. A recent statement by a Chinese general seemed to hint at such reasoning. Another risk is that the US might fear that China might think this way, and thus consider either conventional or nuclear pre-emptive strikes against Chinese nuclear forces. 
Fortunately these scenarios remain somewhat improbable. A second and more significant risk is that strategic nuclear competition between Washington and Beijing will amplify the already strong elements of competition in their wider relationship, and thus make it harder for the two giants of the Asian Century to negotiate a lasting and harmonious modus vivendi. The more the US is seen to be striving to neutralize Beijing's deterrent and achieve nuclear primacy over China, the more likely that China will conclude that America's ultimate intentions towards it are hostile. The more that Americans see China striving to preserve the capacity to overcome US nuclear defenses, the more likely they are to see China as threatening, and the more reluctant they will be to seek accommodations with its rising power. This matters because Asia will only remain peaceful in coming decades if the US and China can avoid being drawn into a strategically adversarial relationship. America, it seems, is determined to retain its position as the leading power in Asia. China evidently expects and intends to exercise increasing regional leadership as its power grows. The adjustment of these potentially incompatible objectives to produce a mutually acceptable power-sharing arrangement which also finds space for Japan, India and others is no easy task, and success cannot be taken for granted. The tension and suspicion generated by an accelerating and increasingly overt nuclear arms race would make it all the harder. Failure would be a disaster for all of us.
Purists would say the best solution to the problem we have identified would be a global agreement to reduce and eliminate all nuclear weapons. But the specific risks that arise from escalating strategic competition between the US and China seem too great and too urgent for us to wait on the slender hope that this ideal solution can one day be achieved. Clearly there is a pressing need for the US and China to discuss frankly their perspectives and objectives on nuclear strategic issues, and proposals for such dialogue have been made fairly regularly.  A more ambitious and effective aim however would be the negotiation of a bilateral arms control agreement between them that would stabilize their respective nuclear and missile defensive forces at or near the current levels. The basic structure of such a deal is not very complex. Its core would be mutual agreement to levels of nuclear offensive and defensive forces on both sides. The agreement would need to allow the US to build missile defenses sufficient to protect against the small 'rogue' nuclear attacks against which their NMD is ostensibly directed, and sufficient offensive forces to preserve its own deterrent. The agreement would need to allow China sufficient offensive forces to ensure that enough of them would survive both a US disarming strike and the agreed level of US defenses to give Beijing confidence that it could successfully attack a small number of high-value US targets, and hence preserve their minimum deterrent posture.
No doubt the detailed negotiation of a deal along these lines would be extremely complex. The old Cold War bilateral agreements between the US and the Soviet Union provide a deep reservoir of ideas and approaches that could be brought to bear on this new problem, but there will also be new issues, such as how to take account of US conventional silo-busting weapons. However, now seems a good time to look for solutions, because none of these problems will get easier with the passage of time. China for the next decade or two is obviously at a big disadvantage in any sustained nuclear arms race with the US, and has much to gain from a deal which preserves its minimum nuclear deterrent without requiring it to spend immense sums on an endless spiral of bigger and more sophisticated offensive forces. The US for now enjoys clear advantages of resources and technology, but Washington may realize that this may be a declining asset as China grows stronger and its technology base improves over coming decades. Better to make a deal with China now when it can negotiate from a position of strength than wait for China to close the gap and lose its present advantages. Indeed the next few years may provide a unique moment of opportunity: the point in the rise of China at which the present relative levels of power provide both sides with the maximum incentive to negotiate. Today, China has an incentive to negotiate because it knows America will retain an advantage for years to come, and America has an incentive to negotiate because it knows that its advantage will be steadily eroded. Within a few years, as China closes the gap, such calculations may start to change, making the idea of a deal less attractive to both sides.
However, this is not the most important reason to pursue a US-China nuclear arms control agreement urgently. The biggest risk of delay is that events and attitudes will overtake us. At present, US-China relations are in relatively good shape, thanks to sustained efforts and effective diplomacy by both sides. A US-China clash over Taiwan, territorial disputes with Japan or some other cause could provoke a freeze in US-China relations which would make the negotiation of any agreement much harder, if not impossible. Less dramatically, we cannot be sure that the overall temperature of the relationship will not fall as the US disengages from the Middle East and starts to recognize how much the strategic challenge from China has grown since 9/11. There may never be a better time than now.
Even so, the task would not be easy, because this kind of arms control agreement would be a huge step. First, it would require, especially for China, a new approach to nuclear strategy, relinquishing the benefits it believes it gains from secrecy, and opening up its nuclear forces to inspection and verification. More profoundly, a US-China arms control agreement would require something of a revolution in attitudes in both of the two countries themselves - in America's attitudes towards China and its growing strength, and in China's attitudes towards its place in the international system and its obligations as a great power For America, a decision to reach this kind of deal with China would mean moderating aspirations of unchallengeable primacy, and a step back from expectations that the unipolar moment can be extended indefinitely. For both Beijing and Washington, a bilateral nuclear deal would provide an opportunity and a framework to think deeply about where this most vital relationship is going, what each can realistically hope from it, and how it can best be managed. It would imply, and perhaps impel, recognition on both sides that the relationship, so full of promise, also carries serious risks that they have to find a way to manage together, cooperatively, through compromise and accommodation. That, in the end, would be the biggest benefit of a deal, and the biggest loss if the opportunity is allowed to pass.
* Lowy Institute for International Policy is an independent international policy think tank based in Sydney, Australia. The views expressed in this paper are entirely the author's own and not those of the Lowy Institute for International Policy.
13 Natural Resources Defence Council, Chinese nuclear forces, 2006. Bulletin of the Atomic Scientists May-June 2006 pp 60-63.
14 McDonough, Nuclear superiority p 77.
15 IISS Military Balance, but contra see Lieber and Press, The rise of US nuclear primacy p 49.
16 McDonough, Nuclear superiority pp 77-79; Lieber and Press, The rise of US nuclear primacy pp 42-54; Medeiros, Evolving nuclear doctrine p 55; Li Bin, The effects of NMD on Chinese strategy. Janes Intelligence Review 7 March 2001; Lindsay and O'Hanlon, Missile defense after the ABM treaty; Tian Jingmei, The Bush Administration's nuclear strategy and its implications for China's security. Center for International Security and Cooperation, Stanford University, March 2003 http://iis-db.stanford.edu/pubs/20188/tian.pdf; Brad Roberts and Shen Dingli, The nuclear equation in Asia, in Burkard Schmitt (ed.), Nuclear weapons: a new Great Debate. Chaillot Paper 48. Institute for Strategic Studies, Paris July 2001 pp 127-146. Official Chinese statements on these issues are rare, but see: Briefing by Ambassador Sha Zulang Director-General of the Department of Disarmament and Arms Control, Chinese Ministry of Foreign Affairs, Press Briefing on National Missile Defence, Beijing, 14 March 2001.
17 McDonough, Nuclear superiority p 18: Gaddis, The Cold War p 131.
18 Phillip C. Saunders and Jing-dongYuan, China's strategic force modernization: issues and implications for the United States, in Michael Barletta (ed.), Proliferation challenges and non-proliferation opportunities for new administrations. Occasional paper No 4, Monterey Institute of International Studies, September 2000 pp 40-46.
19 Medeiros, Evolving nuclear doctrine; Saunders and Yuan, China's strategic force modernization.
20 See for example Paul Dodge, China's naval strategy and nuclear weapons: the risks of intentional and inadvertent nuclear escalation. Comparative Strategy 24 2005 pp 415-430.
21 Danny Gittings, General Zhu goes ballistic. Wall Street Journal 18 July 2005 p A13; Dodge, China's naval strategy and nuclear weapons.
22 For a full analysis of some aspects of this problem, see Paul Dodge, China's naval strategy and nuclear weapons.
23 See for example Saunders and Yuan, China's strategic force modernization; Evan S. Medeiros (Rapporteur), Ballistic missile defense and Northeast Asian security: views from Washington, Beijing and Tokyo. The Stanley Foundation and Centre for Non-proliferation Studies, Monterey Institute for International Studies, April 2001.