Richard C Bush is Director of the Center for Northeast Asian Policy Studies at the Brookings Institution. Richard's paper exploring US extended deterrence in East Asia can be found here.

Extended deterrence is complicated, because it is all about the credibility of threats.
Take the security problems posed by North Korea as an example. If the DPRK is considering the use of force to unify the Peninsula, it is less likely to act on those plans if it is certain the US will carry out its threat to retaliate. Pyongyang is more likely to attack (and less likely to be deterred) if it concludes that the threats are idle — either because Washington does not signal clearly or because North Korea underestimates US resolve due to its faulty perception.
South Korea must have a high degree of confidence in Washington's defense pledge as well. Otherwise it faces a stark choice between appeasement and an independent defence policy.
Extended nuclear deterrence is even more complicated, because credibility is but a gossamer thread. The Republic of Korea's insistence that the United States pledge to defend it with nuclear weapons, if necessary, imposes a serious obligation on Washington. If North Korea acquires its own nuclear weapons and the ability to deliver them to US territory, that increases the risk Washington would face in threatening to retaliate against the DPRK.
That is, an adversary's ability to retaliate on the American homeland can reduce the credibility of the US nuclear umbrella for both allies and adversaries alike — which, I am sure, is one reason China acquired nuclear weapons and the means to deliver them and why North Korea is feverishly trying to do so.
Deterrence on the Korean Peninsula is even more complicated because it works in different ways at different levels. The full spectrum of US and ROK power deters a full-scale invasion by North Korea. But the DPRK's forward-deployed conventional forces and their capacity to wreak extensive damage in the densely populated Seoul metropolitan area have generally led Seoul and Washington to be very cautious about using conventional force against the North. Pyongyang, therefore, has seen only modest risk in undertaking provocative military actions against the South, such as the sinking of the ROK Navy's Cheonan in March 2010 and the shelling of Yeonpyeong Island eight months later. (For an excellent discussion of these issues, see this essay by Admiral (Ret.) Mike McDevitt).
One reason Pyongyang has taken these reckless actions, it seems, is a belief that China is willing to tolerate them. At both levels, the situation is now taking a serious turn. In order to promote both deterrence and domestic solidarity, Seoul has decided that it will retaliate 'kinetically' against future DPRK conventional provocations. That, of course, raises the danger that North Korea will not simply absorb the punishment but retaliate itself. Controlling escalation would become a serious challenge. The chances increase that the US and China — the backers of the two Koreas — would become parties to a widening conflict. What is their stomach for a fight? What is their stomach for not fighting and bearing the costs of backing down?
At the strategic level, it appears to be only a matter of time before the DPRK is able to hit the continental US with a nuclear weapon. In January, in Beijing, Defense Secretary Robert Gates said, 'With the North Koreans' continuing development of nuclear weapons, and their development of intercontinental ballistic missiles, North Korea is becoming a direct threat to the United States'.
Which raises the question, would the US be willing to lose San Francisco as a price for protecting Seoul? Would the government and people of South Korea and the regime in Pyongyang believe Washington would make that sacrifice? Based on what facts and logic would they make that judgment? If they are not convinced, will they opt for their own nuclear deterrent? (A recent poll suggested that the South Korean public is split down the middle on a nuclear option.)
To strengthen deterrence, Washington and Seoul are employing a 'bank shot'. They seek to persuade China that it is not in its interests to allow North Korea to define the security environment in Northeast Asia, either at the level of conventional provocations (with the danger of escalation) or at the strategic level (with the prospect of regional destabilisation). Beijing, of course, is not able to dictate Pyongyang's security policy, but it does have points of influence: the degree of its support for the succession to Kim Jong-il, whether it constrains vital North Korean trade that flows through China, and so on. Moreover, it cannot idly ignore the consequences for China's national security as a result of the measures that the ROK and the US take to strengthen deterrence.
That is the leverage that Seoul and Washington now seek to exercise, with the hope that China will restrain its North Korean partner, not as a favour to them but to better secure China. If successful, this effort will help guarantee that American extended deterrence pledges will never have to be honoured.
The Nuclear Reactions column is supported by the Nuclear Security Project of the Nuclear Threat Initiative, as part of a wider partnership between the NSP and the Lowy Institute.
Photo, of a North Korean guard looking south, courtesy of the US Army Korea IMCOM