Foreign Policy Research Institute A Nation Must Think Before it Acts U.S.-Japan-Taiwan Dialogue: Deterrence, Defense, and Trilateral Cooperation
U.S.-Japan-Taiwan Dialogue: Deterrence, Defense, and Trilateral Cooperation

U.S.-Japan-Taiwan Dialogue: Deterrence, Defense, and Trilateral Cooperation

 

DISCLAIMER: The dialogue discussed in this report was sponsored by the U.S. Defense Threat Reduction Agency and their Strategic Trends Research Initiative (STRI) program. The views expressed herein are those of the author(s) and do not necessarily reflect the official policy or position of the Defense Threat Reduction Agency, the U.S. Department of Defense, or the United States Government.

Introduction

China’s investments in military modernization have narrowed the capability gap between U.S. and Chinese forces, especially in the Western Pacific.  China’s military modernization has also greatly increased its nuclear capabilities.  Beijing has increased pressure on Taiwan (including through gray zone activities), expressed growing impatience with the status quo, and has been developing the means for invading or blockading Taiwan. 

The United States, Japan, and Taiwan share an interest in preventing armed conflict over Taiwan and elsewhere in the region, preventing China from forcing unilateral changes to the status quo, and preserving a rules-based international order.  This shared interest is multifaceted and rooted in security interests (geopolitical—in the Indo-Pacific—for the United States, regional and national for Japan, and existential for Taiwan), economic inter-dependence (including key global value chains), and shared democratic values. A Taiwan conflict carries high-risk of spillover into Japan’s waters and airspace and could escalate to a large-scale great power conflict between the United States and China, with significant human and economic costs, and threats to the survival of the current rules-based order. A Taiwan conflict also risks nuclear escalation, which would have serious consequences for all countries in the region and increases the importance of preventing conflict.

In unilateral moves and joint statements with the United States, Japan has signaled stronger support for Taiwan and defined peace and stability in the Taiwan Strait as vital to Japan’s security.  The United States has sharply increased its statements of commitment to opposing coerced unification of Taiwan by China, to the extent that its long-standing policy of “strategic ambiguity” is increasingly in doubt.  Japan is the key U.S. ally in a conflict over Taiwan and in deterring China from attacking Taiwan.

The United States, Japan, and Taiwan must cooperate to address these threats and determine how to manage them. To that end, the Foreign Policy Research Institute (FPRI) convened a U.S.-Japan-Taiwan Track 2 dialogue in July 2022 to discuss defense and deterrence issues.

Findings Summary

The dialogue yielded several key points of broad agreement among the participants:

  • The United States and Japan must develop better channels for military-to-military communication with Taiwan at the senior and middle levels. This will require Japan to reinterpret or remove official restrictions on contacts with Taiwanese personnel.
  • The United States, Japan, and Taiwan need to address domestic political foundations for resolve and resilience, primarily by communicating effectively to their respective publics the national security consequences of failing to deter conflict or defend Taiwan (as well as the costs that must be borne to deter China or in a conflict with China).
  • The United States, Taiwan, Japan, and other like-minded countries should pay greater attention to the assurances side of deterrence. Possible examples include signaling to China that Taiwan will not pursue de jure independence and the United States and others will refrain from recognizing Taiwan as an independent state, so long as China upholds its obligation to refrain from use of force and coercion.
  • As China’s nuclear forces modernize, the United States must prepare to address the “stability-instability” paradox: China’s leadership may become increasingly likely to initiate a Taiwan conflict and use conventional force as its confidence in its nuclear deterrent grows.
  • Because a Taiwan conflict poses grave threats to U.S., Japanese, and Taiwanese national security, all three countries should urgently increase investments in deterrence and defense readiness.

This report addresses these and other conclusions in more detail in the Analysis and Key Findings sections.

Methodology

FPRI convened a U.S.-Japan-Taiwan Track 2 dialogue in Washington, DC, on July 13-15, 2022, to address challenges of defense and deterrence in responding to the threats posed by China, primarily in Taiwan-focused scenarios, and the possible role of trilateral (U.S.-Japan-Taiwan) cooperation. Participants included retired military leaders, former government officials, and non-government subject matter experts from the United States, Japan, and Taiwan. Approximately ten participants attended from each country.

Participation by leading non-government subject-matter experts and retired officials with intimate knowledge of their home country’s political and military initiatives and relations with the other two countries facilitated informed and candid discussions about prioritizing challenges and taking constructive and feasible actions.

For several Japanese and Taiwanese participants, this dialogue was their first opportunity to travel to the United States and interface directly with their counterparts in several years. For some, this also was their first trilateral discussion of defense and security issues with former officials from the United States, Japan, and Taiwan.

The dialogue was comprised of 90-minute sessions on the following topics: (1) threats posed by China; (2) nuclear weapon threats; (3) foreseeable threat scenarios; (4) strategies to deter a People’s Liberation Army (PLA) invasion of Taiwan; (5) opportunities to improve interoperability and defense; (6) deterrence signaling; (7) conventional escalation and spillover risks; (8) nuclear escalation risks; and (9) de-escalation strategies.

A participant from each country began each session with a brief presentation. The session then shifted to a moderated open discussion. The dialogue operated under the Chatham House Rule, which allows participants to use information from the dialogue but does not permit identifying the individual who made particular comments.

By design, this dialogue covered a broad range of topics, partly to identify questions that require further examination and to lay the groundwork for future U.S.-Japan-Taiwan dialogues.

Analysis

The dialogue addressed six major issues: deterring conflict, developing trilateral defense cooperation, improving deterrence signaling, reinforcing the political foundations of deterrence, assessing potential spillover risks, addressing emerging nuclear weapon challenges, and anticipating de-escalation challenges.

Deterring Conflict

Participants overwhelmingly agreed that the risk of a China-initiated conflict focused on Taiwan is real and rising.  China has been investing heavily in developing the power projection, sealift, and other capabilities required to attempt large-scale military action against Taiwan, including invasion and blockade.  China has greatly increased gray zone activities targeting Taiwan and Japan, including military exercises that also build Chinese capabilities for an armed attack. Signs of growing intent have accompanied China’s increasing capacity. China’s leadership has linked unification of Taiwan to the goal of “national rejuvenation” it plans to accomplish by mid-century.  Xi Jinping has declared that the Taiwan issue “cannot be passed down from generation to generation,” suggesting a possible timeline of achieving unification during Xi’s tenure.[1]  In 2005, China adopted a Taiwan-targeting Anti-Secession Law that reiterated its commitment to unifying Taiwan and using force if necessary to do so.  Meanwhile, achieving unification by peaceful means is growing less likely, even more so after China’s crackdown in Hong Kong demonstrated to Taiwanese voters the potential consequences of accepting a “one country, two systems” model.

Events after the dialogue reaffirmed these concerns. China launched large-scale military exercises near Taiwan, including blockade-simulating actions, and issued a third White Paper on the Taiwan issue that reiterated past positions and took a somewhat tougher tone, including in its formulation of Beijing’s “one country, two systems” model for post-unification Taiwan.[2] China’s increasing assertiveness toward Taiwan aligns with a broader trend of China shifting toward more assertive and militarized foreign policy tactics (which have also targeted Japan as seen in Chinese gray zone actions in the East China Sea). This trend demonstrates senior Chinese leaders’ increasing willingness to use its military and bear higher risks and costs.

Participants identified at least seven conditions that could prompt China’s leadership to initiate a major use of force:  

  • China might perceive that Taiwan is approaching or has crossed Beijing’s red line on Taiwan independence. 
  • China’s “deadline” for achieving unification (possibly by 2049) might arrive with unification still unachieved. 
  • Xi might near the end of his tenure and decide that unifying Taiwan is a vital “legacy” issue that must be resolved. 
  • Xi—or the party—might face an internal threat to its hold on power (related or unrelated to the issue of Taiwan) and may take some action (likely short of invasion) toward Taiwan in order to stir nationalist sentiments, or show progress toward “national rejuvenation” and the “China Dream.” 
  • China might determine that its window of opportunity is closing—as sentiment in Taiwan against unification and/or for independence deepens and the United States increases its military capabilities in the Indo-Pacific. 
  • China may perceive a fleeting moment of high opportunity to take Taiwan (perhaps due to a perception of U.S. distraction or a weak government in Taiwan). 
  • Unintended escalation, vertically from a Taiwan crisis or horizontally from another regional conflict, is also possible. In many cases, the opacity of Chinese perceptions and decision-making will make it difficult for the United States, Japan, and Taiwan to accurately predict escalation and prepare effective responses.

Participants broadly agreed that it would be difficult and costly to deter China in a Taiwan scenario, but that the risks and costs of failing to do so would be dire for the United States, Japan, and Taiwan.  Recent failures of deterrence (for example, in Ukraine) underscore the challenge.  States sometimes fight wars when it is irrational to do so, and China’s leadership appears willing to incur high risks and costs to accomplish its goals.  Some participants warned that, if China’s leadership perceives it necessary to use force toward Taiwan, China may do so even if it concludes it runs a significant risk of failure, knows it would face U.S. intervention (supported by Japan), and expected to suffer severe harm to China’s Xi-era goals to “national rejuvenation” and economic prosperity.

A core element of a U.S. and allied deterrence strategy is having adequate military capabilities to defeat China in a conflict scenario or, failing that, being prepared to impose decision-altering collateral costs on China. U.S. and Japanese participants generally saw Japan’s principal deterrence role as having the capacity to bear some of the burden of defending the first island chain and providing the United States with adequately defended or defensible bases and logistical and other support.  U.S. and Taiwanese participants saw Taiwan’s principal role as pursuing a “porcupine strategy,” acquiring a “large number of small things” in service of an asymmetrical defense strategy to be supplemented by popular resistance that would slow down, make costly, and thereby deter a Chinese attack.  The Ukraine conflict has underscored the importance of prepositioning dispersed stockpiles of resources before a possible Chinese blockade or attack.  Enhancing Taiwan’s resilience, including of civilian physical and governmental infrastructure, also serves this end.

A Japanese participant expressed concern that the United States did not have sufficient assets deployed in the Western Pacific to deter China in a Taiwan scenario.  Other participants noted that the broader context of great power conflict means that U.S. assets not specifically committed to the theater would matter for deterrence, however the Japanese participant remained concerned that some assets could not move into place soon enough to deter China in the event of a crisis or a sudden Chinese determination to act. Another participant from Japan also raised concern that the United States may need to commit its assets elsewhere to address another conflict or crisis that falls within the ambit of U.S. global commitments and responsibilities. Concerns, shared by some U.S. participants, focused on whether the United States had sufficient forces in the region to hold Chinese forces within the first island chain, and whether adequate U.S. assets in Japan, Guam, and elsewhere would survive an initial Chinese attack. Those participants did not specify what U.S. military presence would be needed to address their concerns. Further discussions with subject-matter experts on that specific topic would be beneficial.

Deterring China is also likely to require the United States, along with allies and like-minded partners (including Japan and Taiwan) to inflict major damage on China’s abilities to achieve its other high priority goals, including economic prosperity and international stature.  Non-military instruments should play a principal role in this campaign, including severe collaborative sanctions and norm-invoking lawfare.

Trilateral Defense Cooperation

Although the governments of the United States, Japan, and Taiwan now appear to recognize that China’s threat to Taiwan is a threat to all three and that a Chinese attack on Taiwan would likely draw all three into the conflict, challenges remain.  The three countries lack a clear and common definition of strategic goals in addressing threats from China.  Trilateral coordination and cooperation, especially in intelligence and military spheres, and the ability to allocate defense roles effectively, remain limited in the absence of formal alliance or diplomatic relationships with Taiwan.

Some Japanese and Taiwanese participants argued for a clear division of labor among the United States, Japan, and Taiwan, which would lessen the need for fuller interoperability and could more generally improve the effectiveness of defense efforts.  Again, participants agreed that Taiwan should focus on a “porcupine strategy” and an asymmetrical defense to impede or slow a Chinese attack.  Some participants noted that this did not mean entirely foregoing weapons or military roles with a more offensive capacity to engage Chinese air force and navy assets before they reach Taiwan.  Japan’s principal mission would be to defend its own territories and near seas, contain China’s forces at the northern portion of the first island chain, and provide bases and extended logistical support for U.S. forces.  Other missions in a conflict with China, including the southern end of the first island chain would fall primarily and inevitably to the United States.  However, Japan’s roles might expand, as a U.S. participant suggested, so that Japan would eventually aim to take the lead on defending all Japanese territory (including the southern islands), with the United States assisting Japanese forces with the defense of Japan’s territory where needed but primarily focusing on defending Taiwan’s territory and potentially engaging Chinese forces in other theaters. The exact division of labor between the United States and Japan may still require further discussion.

Participants identified impediments to closer defense cooperation, including some aspects of interoperability.  These problems are unevenly distributed; the least serious are between the United States and Japan and the most significant are between Japan and Taiwan.  Impediments include the absence of a full-fledged mutual security arrangement between the United States and Taiwan and between Japan and Taiwan, Japan’s limitations on government-to-government and military-to-military contacts with Taiwan, language barriers, and a lack of experience in jointly conducted combat operations.  The presence of U.S. forces in Japan and regular interactions among U.S. and Japan military counterparts greatly enhances U.S.-Japan cooperation.

Cooperation and interoperability vary considerably across levels.  Hardware compatibility is mostly good although the United States and Taiwan diverge in some key procurement priorities (with Taiwan still seeking expensive and vulnerable platforms that the U.S. views as ill-suited to asymmetric defense or a porcupine strategy).  Japan and the United States have well established Operational Control (OPCON) and staff communications, but they are weaker between the United States and Taiwan and weaker still between Japan and Taiwan.  This creates problems in a crisis or conflict scenario when Operations Center staff do not have channels to their counterparts.  Encrypted communication is constrained because it requires compatible systems and sharing of sensitive encryption technology and keys.  Physical infrastructure for communications and exchange of information is likely vulnerable in a Taiwan scenario because China will target Taiwan’s relevant infrastructure.  Cooperation and information sharing at the operational planning level is also limited.

Intelligence sharing is especially valuable and difficult.  Participants noted that the lack of deep and real-time intelligence sharing (including signals intelligence) are impediments to effective defense cooperation and interoperability.  But obstacles to reform are multiple and formidable: resistance to sharing intelligence information across agencies within each government; domestic laws that need to be made compatible with other states’ requirements for intelligence sharing; and the complexity of extending fundamentally bilateral intelligence sharing agreements to trilateral or multilateral contexts. 

A U.S. participant pointed especially to a more fundamental issue in the U.S.-Japan-Taiwan context. Taiwan and Japan seek greater intelligence cooperation from the United States, but the United States has concerns: the benefits of such sharing is highly asymmetrical and the United States lacks the requisite trust that its information, operations, and methods will not be compromised.  This last concern is especially great in the case of Taiwan, which China has heavily targeted in its espionage efforts.  For these reasons, the high level of cooperation among the Five Eyes cannot be replicated among the United States, Japan, and Taiwan.

Participants pointed to several areas where cooperation exists or is relatively feasible, and could serve as building blocks to more effective cooperation and interoperability.  The United States and Japan should reduce impediments to military-to-military contacts with Taiwan.  The United States, Japan, and Taiwan should make concerted efforts to connect action officers to their foreign counterparts. Formal contacts could include Professional Military Education (PME) exchanges, observing or participating in joint training exercises (including Coast Guard, Non-Combatant Evacuation Operation, or Humanitarian Assistance and Disaster Relief exercises). Informal channels could include sending active-duty personnel to serve as visiting fellows at the same research institutions, observers to the same Track 2 or Track 1.5 dialogues, or attendees at the same conferences or ceremonies.  The Global Cooperation Training Framework (GCTF) provides a rare, existing forum that fully includes Taiwan that could provide a ready basis for enhancing broadly security-related contacts and cooperation. These mechanisms can foster cooperation in contexts where deep interoperability is not necessary (for example, in the East China Sea).

Deterrence Signaling

The United States, Japan, and Taiwan have all taken steps recently that signal to China their resolve in resisting invasion and some other forms of coercion of Taiwan.  Taiwan has reiterated its commitment to maintaining the cross-Strait status quo and moved to adopt a more “porcupine”-like defense strategy.  Japan has made unprecedented declarations, sometimes at meetings with the United States, that it views peace and stability in the Taiwan Strait as important to its national security.[3]  The United States has repeatedly condemned China’s coercion of Taiwan, moved away from traditional positions associated with “strategic ambiguity” toward more robust commitments to defend Taiwan in the event of an attack by China, raised the level of contacts and visits by current and former officials to Taiwan, and maintained a heightened pace of arms sales to Taiwan.

Participants expressed concern that such signals may not be effective in deterring China, for several reasons.  A U.S. participant stressed the lack of a robust theory of how signaling works and the means for measuring whether it is effective.  Deterrence signaling is challenging because it requires signaling both capacity and will and thus presents challenges about how much to emphasize one or the other.  Moreover, effectively signaling capacity and/or will undermines concealment of abilities and intentions, which are themselves valuable assets in dealing with, and deterring, an adversary.

Deterring China from coercive action or force against Taiwan is especially challenging and success may be very hard to discern.  Unifying Taiwan—and avoiding the “loss” of Taiwan—are exceptionally high priorities for the Chinese leadership.  Under some conditions (addressed elsewhere in this report), China may invade or attack Taiwan even if it believes it is relatively likely to fail (or incur very high costs).  In such circumstances, efforts at deterrence signaling are unavailing.  Some participants expressed concern that China was drawing incorrect and deterrence-undermining lessons from the conflict in Ukraine, inferring that Russia’s nuclear threats had worked and that the United States is unlikely to engage in direct military conflict with China if China were to make similar nuclear threats when attacking Taiwan (or using lesser coercive measures).

Participants noted that analysts lack sufficient insight into the decision-making process and influences on China’s top leader(s), who will make a decision to act, or refrain from acting, in a Taiwan scenario.  China’s statements provide noisy signals.  China “cries wolf” (making ultimately empty assertions that it deems statements and actions from the United States, Taiwan, and others extremely provocative and requiring a harsh response) and unleashes “wolf warriors” (adopting a bellicose posture that, if taken at face value, call for stronger deterrence signaling in response).  Some participants pointed to China’s pledges to abide by international norms and rules, and then not doing so, as another challenge for deterrence signaling.

U.S. signals—especially of will—are problematic in several respects.  Some participants and many analysts see recent statements by the President as departing from traditional positions in expressing commitments to defend Taiwan with U.S. troops in the event of an attack by China and suggesting that Taiwan’s status (including independence) is a matter for Taiwan to decide.  While these are reconcilable with established U.S. positions of “strategic ambiguity” and “not supporting” Taiwan independence, they have sown confusion—especially when the administration has sought to partly walk them back—and have created room for China to charge that the United States is changing its long-standing policies and undermining the status quo.[4] The recent rise in prominence of congressional voices on U.S.-Taiwan policy has also muddled U.S. deterrence signaling. This includes proposed or implemented legislation and (after our event) the Speaker of the House’s visit to Taiwan. Some participants from all three parties agreed that the United States should restate or clarify its policy. Taiwan participants, in particular, urged that U.S. signals focus on substantive measures to enhance Taiwan’s security, rather than symbolic gestures that could provoke China. 

Some experts, including U.S. participants in this dialogue, have urged that U.S. policy pay greater attention to the “assurance” side of deterrence—that is, to signal to China that the U.S. will not seek to damage China’s vital interests if China refrains from attacking (or perhaps otherwise severely coercing) Taiwan.  Although participants did not agree on what form assurances should take, possibilities (suggested by some U.S. participants) include [5]  that the United States should make clear that its objective is not an independent Taiwan but peace and stability in the Taiwan Strait. U.S. leaders should restate key aspects of the U.S. “one China” policy consistently and clearly—particularly that the United States will accept any outcome that is peacefully agreed to by China and Taiwan, does not support Taiwan independence, and opposes unilateral changes to the status quo.  The U.S. should make clear that it is not pursuing sovereign status for Taiwan when it presses for Taiwan’s inclusion in international organizations that do not require statehood for membership, nor its meaningful participation in intergovernmental organizations such as the World Health Organization, or Taiwan’s accession to international trade and investment agreements.  Some of these recommendations are controversial or carry risks of emboldening Beijing. U.S. officials should vet these recommendations further and consult with Taiwan, Japan, and other allies and partners before implementing them.  

Political Foundations for Deterrence

A U.S.-Japan-Taiwan collaborative deterrence strategy relies on political foundations.  The United States, Japan, and Taiwan must be able to persuade key states, including NATO members, to oppose aggression by China and support sanctions and other punitive measures if it occurs.  The United States, Japan, and Taiwan must have sufficient confidence in each other’s commitments in various Taiwan contingencies.  The United States, Japan, and Taiwan must take steps to address issues that raise questions about resolve: Taiwan must maintain stable domestic politics and a society-wide commitment to resisting China.  The United States must maintain stable and robust commitments to U.S. allies and partners that can endure  polarization and possible turmoil in U.S. politics.  Japan must assure that its public has the collective will to support the United States and Taiwan during a potential conflict, even if doing so will require Japan to bear great risks and potential costs—possibly including a limited nuclear strike. Participants expressed concerns that publics in the United States and Japan may not adequately appreciate the importance of Taiwan’s security to their own countries’ security and that Beijing likely underestimates U.S., Japanese, and Taiwanese resolve.

Participants assessed that the prospect of a strong international response could help deter China from initiating a conflict or degrade its efforts to sustain a conflict. However, China’s economic leverage over many relevant states and China’s long-running and effective campaign to marginalize Taiwan diplomatically may impede efforts to mobilize the international community.

Participants’ comments pointed to the challenges for collaborative deterrence of limited mutual confidence among the U.S., Japan, and Taiwan governments. Participants from Taiwan and Japan expressed concern about U.S. commitments to respond quickly to Chinese aggression. U.S. participants expressed concern about Japan’s commitment to providing U.S. forces with base access during a conflict. U.S. and Taiwanese participants expressed concern about Japan’s governmental policy restricting official contacts with Taiwan. Participants from Japan and the United States expressed concern about Taiwan’s readiness to defend itself and commitment to making the defense investments required for a credible self-defense capability.

Participants agreed that the United States, Japan, and Taiwan should make concerted efforts to explain to their publics the consequences of failing to defend Taiwan for national security and national interests, as well as the costs that would have to be borne in a conflict over Taiwan. War over Taiwan would entail physical devastation and loss of life in Taiwan, attacks on military bases in Japan, significant damage to U.S. forces and assets in the region, and collateral costs. China’s occupation of Taiwan would allow the PLA to position missiles, naval units, and other assets on Taiwan, making it much more difficult to defend Japan, especially its southern islands and the Senkakus.  Occupation of Taiwan would significantly benefit China’s submarine warfare and ocean surveillance capabilities, including providing better placement for acoustic sensors used to detect and target U.S. surface combatants and by providing a more advantageous basing location for China’s sea-based nuclear deterrent.[6] Overall, it would greatly enhance China’s ability to operate with relative freedom beyond the first island chain and weaken U.S. ability to operate in the Western Pacific.  More broadly, China’s control of Taiwan or other nearby areas would accelerate its ability to undermine and rewrite international norms and rules.

Spillover Risks

Participants identified several paths to escalation and spillover risks in contingencies involving Taiwan and other flashpoints in the region. Participants did not reach consensus about the probability that escalation or spillover would occur in various scenarios.

China might attack and seize one or more offshore islands controlled by Taiwan.  Most of these islands are of little or no military value and in Chinese hands would become vulnerable targets for the United States and its allies.  Nonetheless, some participants argued China could choose this course of action, perhaps to demoralize the Taiwanese public or weaken the Taiwanese government.  Despite the lack of military significance, such a seizure by China likely would be politically impossible for a government in Taiwan to ignore.  Such a seizure could also act as a test of U.S. resolve, creating pressure for an escalatory response by the United States. Participants generally agreed that Taiwan, and perhaps the United States, would feel compelled to respond with military force, leading to escalation. 

China might attack and invade Taiwan, without targeting Japan or U.S. forces, in the hope of achieving a quick conquest or capitulation before the United States and Japan entered the war.  This approach would entail large risks for China (facing threats from as-yet un-degraded Japan and U.S. assets).  It would not reliably eliminate or greatly reduce the risk for China that the conflict would escalate in scale and expand geographically.  A successful initial invasion would not preclude the United States and Japan from intervening to challenge China’s advances.  A Chinese invasion that did not rapidly succeed would more likely face intervention by U.S. forces based in Japan and elsewhere.

In what many participants viewed as a more likely pattern, China would attack Taiwan and U.S. bases in Japan and U.S. military assets elsewhere in the region.  This would immediately escalate a China-attacks-Taiwan scenario into a region-wide—and possibly broader—great power and multi-party conflict.

Some participants (from all sides) expressed concern that the United States, Japan, and Taiwan have war aims that are unclear or different (or at least perceive one another’s aims as different) in a Taiwan conflict scenario.  This could lead to challenges for a coordinated response in an escalating or spreading conflict.

Participants noted other possible spillover scenarios.  Disputes between China and Japan over the Senkaku Islands/East China Sea claims could trigger Chinese military action.  Participants saw Chinese actions around the Senkakus or in the Miyako Strait as low priority or low value in a scenario where China attacks Taiwan, but noted the possibility in the context of China-Japan frictions.  U.S. officials have reaffirmed that the U.S. interpretation of the mutual security treaty with Japan includes a commitment to defend the Japanese-administered Senkaku islands, meaning a China-Japan conflict likely would immediately prompt U.S. involvement.[7]  Participants differed over whether a potential China-Japan conflict in the East Chins Sea would inevitably involve Taiwan, but participants agreed that if China were to find itself in a conflict with the United States and Japan that did not initially involve Taiwan, China’s leadership might conclude that it is already suffering the primary costs of a Taiwan conflict and therefore expanding the conflict to Taiwan would not incur significantly more costs. Inadvertent spillover may also occur if fighting is in close proximity to Taiwan.

A similar scenario was possible in the South China Sea, but seemed less likely to most participants.  Still, the risk of spillover and escalation exists, given Taiwan’s control of the largest island in the area, the multiple claimant states whose interests are at stake, and the extensive presence of U.S. and allied navies in the region.

Participants also noted the possibility of spillover to or from the Korean peninsula.  North Korea is a recurrently destabilizing factor and could act dangerously or provocatively to exploit perceived opportunities in a U.S.-Japan-Taiwan-China conflict.  Alternatively, as a Taiwan conflict escalated into a U.S.-China great power conflict, China might strike U.S. assets in South Korea.

Finally, participants noted the possibility of vertical escalation from gray zone coercion.  This could take the form of inadvertent escalation from an incident at sea or in the air (although it could take other forms as well).

Nuclear Weapon Challenges

China has been modernizing, diversifying, and expanding its nuclear forces.[8]  Some of these developments introduce new challenges for nuclear stability.

China has developed and deployed high precision, dual-use intermediate-range ballistic missiles: the DF-21 and the hot-swappable DF-26.[9]  These missiles are potentially destabilizing or escalatory in several respects.  First, because they can be equipped with either nuclear or conventional warheads, they raise risks of entanglement and “conventional counterforce” risks: U.S. strikes targeting China’s conventional forces inadvertently degrade or endanger components of China’s nuclear arsenal during a conflict, triggering limited Chinese nuclear escalation.[10]  

Second, when nuclear-armed, these missiles have ambiguous purposes. Because they are intermediate-range weapons that cannot reach the U.S. homeland, they cannot contribute to either a countervalue or a counterforce nuclear deterrence mission. The former requires holding U.S. cities at risk. The latter requires the capacity to destroy nuclear forces outside the region. The DF-21 and DF-26 therefore must serve other nuclear missions. They might strike U.S. bases in Guam, as well as Japan and Taiwan, or on the battlefield against U.S. and allied forces. They might be used for nuclear blackmail (particularly against Japan, Taiwan, and others who lack their own nuclear forces and may doubt U.S. commitments to extended deterrence when the U.S. homeland is not at risk from these weapons).  Their purpose might be to enhance deterrence of regional nuclear powers (India, primarily). Alternatively, they might serve to provide a response option if the United States were to deploy comparable weapons to the region. Some, but not all, of these possible purposes create pressure for the United States to forward-deploy equivalent capability, which would raise the risk of one side choosing to use the weapons in-theater and on the battlefield, particularly if that side faces a use-or-lose scenario.  The imperative for the U.S. to deploy—or for states in the region to develop—comparable capabilities depends on which possible missions they foresee for China’s missiles.

China has been working to create a larger and more survivable strategic nuclear capability.  This effort includes deploying intercontinental ballistic missiles equipped with multiple independently-targetable reentry vehicles on mobile launchers and, soon, from hardened silos (e.g., the DF-5, DF-31, DF-41).  China is testing a FOBS-like weapon system that can place a hypersonic glide vehicle in low Earth orbit.  It is building up its submarine-based platform and increasing anti-submarine warfare capacity (to address the threat to China’s nuclear arsenal from U.S. platforms).[11]

China has invested in early warning capabilities and nuclear platforms that can maintain a higher state of peacetime readiness, which give China greater ability to adopt a “launch on warning” posture, with launch authority pre-delegated further down the chain of command, increasing the risk of accidental escalation or unauthorized nuclear use. Nuclear experts remain divided on whether China’s leadership would choose to adopt this posture.[12] The CCP’s organizational culture has always insisted on tight political control over strategic military capabilities and a degree of civilian distrust of the military, which suggests the CCP would be reluctant to pre-delegate launch authority to points down the chain of command. However, the DoD’s 2021 Annual Report to Congress on China’s military capabilities suggests that China is in the process of moving toward a “launch on warning” posture referred to as “early warning counterstrike.”[13]

U.S., Japanese, and Taiwanese participants were skeptical of China’s commitment to no first use and cautioned against viewing it as a reliable guarantee in a Taiwan conflict.  While some expressed the opinion that the doctrine was never strongly constraining, others observed signs that the doctrine is becoming more flexibly interpreted or marginalized in Chinese discussions of use-of-force scenarios.  While China’s nuclear arsenal would not be useful in an initial attack on Taiwan, and the use of nuclear weapons would have major international reputational costs, participants agreed there were some plausible first strike scenarios. Some participants expressed concern that the danger that China would use nuclear arms is higher because its conditions for using them are not clear (and thus might be more expansive than we understand) and that China may be overly confident it its ability to control the ladder of in-conflict escalation.

Participants identified several circumstances in which China might escalate to nuclear weapons.  First, China might attempt to engage in nuclear blackmail to induce Taiwan to capitulate, to deter Japan (which China might consider especially sensitive/vulnerable) from intervening or supporting the United States, or to deter the United States from intervening.  A Japanese participant highlighted the risk of nuclear blackmail against Japan. When asked, the Japanese participants could not provide a definitive answer about whether Japan’s political leaders and public would be more likely to resist or capitulate if subjected to nuclear blackmail by China. Such attempts at blackmail can lead to brinksmanship and actual use.  Second, China might turn to nuclear weapons if it is losing, incurring great losses, or becoming hopelessly bogged down in an effort to invade and conquer Taiwan (or in some other conflict scenario).  A Taiwanese participant believed that it was unlikely that China would directly target Taiwan with a nuclear strike, but did believe China might target U.S. or Japanese forces (specifically U.S. or Japanese aircraft carriers).  Third, a U.S. participant believed China might conclude that (in a conflict with the United States) it has no adequate, non-nuclear means to attack hardened or hard-to-reach U.S. military targets (e.g., hardened bunkers, aircraft carriers, mainland U.S. targets).  Fourth, a Japanese participant (and U.S. participant) believed China might conclude that it is facing a “use it or lose it” situation with its nuclear weapons. This could be the result of actual or impending attacks on its command and control or ISR systems or because China construed an attack on its dual-use intermediate-range missiles as an attack on nuclear, rather than conventional, assets.  Finally, two U.S. participants and one Japanese participant believed the CCP regime might conclude that it was facing an existential threat of attack from or defeat by an adversary (presumably the United States).

Participants also assessed the mode that a Chinese (first) use of nuclear weapons would take.  China might undertake some form of demonstration, rather than actual targeting use, likely reducing the risk of further escalation.  Examples included a nuclear test, a high-altitude detonation, an EMP blast, and striking a small, uninhabited landform controlled or claimed by an adversary (for example, in the South China Sea or East China Sea).  The principal “non-demonstration” use that caused participants concern was a nuclear strike on military assets in Japan or U.S. assets elsewhere, whether to preempt intervention or degrade capabilities in an in-progress intervention.

China could strike Taiwan with a nuclear weapon, but one Taiwanese and one U.S. participants considered this unlikely (with the exception of a high-altitude demonstration strike over Taiwan, which a Taiwanese participant suggested was conceivable).  Participants who were skeptical that China would target Taiwan military assets or infrastructure reasoned that it was unlikely because there is little China could not achieve with conventional weapons.  Using a nuclear weapon would cost China significantly in terms of attitudes toward China in Taiwan (as well as abroad) and (depending on the scale of use) in significant physical destruction to a place China hoped to govern (according to a U.S. participant, China’s current nuclear warheads are suspected to have yields of approximately 300kT or greater).  Participants considered China very unlikely to use a very low-yield nuclear weapon, primarily because China is not known to possess such weapons (although it could easily develop them). In the most likely scenarios for a Taiwan or other regional conflict, use of such a weapon would bring no tactical advantage but would have reputational and political costs and could risk further nuclear escalation. However, the DoD’s 2021 defense white paper on China’s military capabilities indicates that PLA strategists have raised arguments in favor of unspecified lower-yield nuclear weapons and a 2017 defense industry publication reportedly indicated that a lower-yield weapon had been developed, so observers should continue to monitor China’s arsenal and doctrine for signs that China is deploying tactical nuclear weapons.[14]

Participants also noted the possibility that a conflict elsewhere (perhaps on the Korean peninsula or with Russia or Iran) that escalated to nuclear weapons use could increase the likelihood that a U.S./Japan/Taiwan and China conflict could go nuclear.

Participants were also concerned that China might draw dangerous nuclear lessons from the Ukraine conflict.[15]  Beijing may conclude that Putin’s nuclear blackmail successfully deterred the United States from direct conventional (as well as nuclear) intervention to protect a state that lay outside the United States’ nuclear (as well as security treaty) umbrella.  This is especially concerning given that Taiwan also lies outside the U.S. nuclear umbrella (and defense treaties).

Lessons from Ukraine aside, the U.S. vulnerability to nuclear attack by China creates challenges for U.S. extended deterrence and, in turn, for U.S.-Japan-Taiwan cooperation to address threats from China.  Japan and Taiwan inevitably will worry that the United States will not risk a devastating attack on its homeland in order to protect Taiwan or Japan.[16]  These concerns create demands for reassurance from the United States and warrant discussions about possible forward/theater deployment of U.S. nuclear weapons or NATO-like nuclear sharing arrangements.  If Japan requested to host U.S. nuclear weapons, it would require Japan to take the controversial and contentious step of revising the third of its Three Non-Nuclear Principles (Japan will not possess, manufacture, or introduce nuclear weapons).[17]  Deployment in Taiwan is not plausible.

Participants expressed concern that U.S. policies concerning the use of nuclear weapons are unclear or inconsistent.  Some discerned significant differences in the views of the President, defense officials, and Congress, or significant shifts from one administration to the next.  Some noted ambiguities over how much to rely on nuclear rather than conventional forces for deterrence, and whether nuclear deterrence was framed primarily in terms of denial or cost-imposition.

Participants also expressed concern about inadequate information about relevant Chinese behavior—especially in real-time—and inadequate bases for communication and cooperation among the United States, Japan, and Taiwan in situations where the United States or China might use nuclear weapons.

Despite these many challenges, several features that limit the risks of nuclear use and escalation persist.  In addition to challenges mentioned earlier in this section, there is no indication that China is seeking nuclear parity or superiority with the United States in the near or medium term.  C4ISR vulnerabilities and limits to production capacity would make such a goal unachievable for the time being (although these factors can change and the United States should monitor them closely).[18]  China also does not yet appear to be manufacturing low-yield nuclear warheads for so-called tactical nuclear weapons.

A U.S. participant argued that China’s nuclear modernization has the principal aim of creating and maintaining a state of mutual nuclear vulnerability with the United States to create a shield for China to use conventional forces in Taiwan and other scenarios, free from the threat of U.S. nuclear intervention.  If true, China’s nuclear modernization is creating the conditions for the “stability-instability” paradox. As the United States and China approach mutual vulnerability and nuclear stability rises, it may embolden rather than inhibit Chinese willingness to use conventional force and conventional stability may break down. With the United States and China mutually deterred from using nuclear weapons, China will be more willing to initiate a conventional conflict over Taiwan, particularly if China believes it has an advantage in conventional forces in the area and China’s leadership continues to attach an extremely high priority to its goal of unification. Although this is cold comfort, and it does not eliminate the prospect of nuclear escalation, it is a feature that can limit that risk.

De-Escalation Challenges

Participants agreed that de-escalation would be difficult to achieve in a U.S.-China conflict, including one centered on Taiwan and involving Japan, but agreed that it is imperative to develop means that could do so.  The most likely initiation of such a conflict would be an attack by China on Taiwan.  If China were to make such an attempt, it would have done so in the expectation that the conflict would escalate to include direct military conflict with the United States, and that China would suffer serious direct (in-conflict) and collateral (economic and other) costs.  Chinese nationalism—whether as driver for the initial Chinese action or stoked in an in-progress conflict—would make de-escalation more difficult.  Considerations other than traditional and tangible security interests would constrain or impede rational de-escalation.  In some scenarios, the Chinese regime would believe it faced an existential threat if it were not to prevail and achieve its goals.  For the United States and China, a conflict that began over “unification” of Taiwan would quickly become a contest for preeminence.  With the stakes so high and far-reaching, the risks of losing become greater and de-escalation can become still more difficult, compared to a limited “Taiwan-only” scenario.

At the same time, it is very unlikely that either side would escalate the conflict intentionally in an effort to decisively defeat the other.  The costs of such an approach, including the risk of nuclear war, would deter such moves.  Therefore, a protracted and costly conventional conflict is more likely.  Japan and Taiwan would have limited ability, and perhaps limited incentive, to achieve de-escalation.  For Taiwan, some de-escalatory outcomes would be existentially threatening.  For Japan, they could have dire long-term security implications.  In a protracted conflict for preeminence between great powers, the influence of such lesser powers is limited.

Participants noted the possibility, and risks, of “escalating to de-escalate,” potentially by either side.  One concern is that this could take the form of the threat or use of a nuclear weapon, which creates risks of nuclear escalation that are hard to control or predict (especially given differences and ambiguities in U.S. and PRC policies).  A related and likely possibility is that the United States would seek to do something akin to “escalating to de-escalate” by seeking to impose (with allies and partners) massive collateral economic and other costs on China.

De-escalation is typically difficult.  As participants noted, different parties have different understandings of the conflict, incompatible win (or non-loss) sets, different red lines, different perception-shaping strategic cultures, and so on.  Still worse, the United States and China lack robust crisis management and in-crisis communication mechanisms, and China has shown little interest in developing them (preferring to focus narrowly and unhelpfully on crisis-avoidance mechanisms, which often amount to demands for concessions from potential adversaries).  China’s highly centralized and top-down decision-making limits opportunities for incremental de-escalation and innovative de-escalation efforts.  Its practices in past conflicts are discouraging, including its approach of “hitting hard” or escalating in a conflict’s opening phases to bring the adversary to the negotiating table, refusing to engage in negotiations mid-conflict, and rejecting third-party mediation.  A strategic culture of proactivity rather than reactivity may underlie or reinforce this pattern.

Finally, participants identified two conditions that might make de-escalation more likely: one side (in the discussion, China) might be convinced that the conflict was going very badly and that de-escalation, including conflict termination, therefore offered a better outcome; and/or the other side offered a mutually acceptable off ramp.  Participants agreed that both of these conditions would be difficult to achieve and suggested that possible solutions would be highly context-dependent.

Key Findings and Recommendations

Finding: The risk of a Taiwan conflict is rising and poses challenges for collaborative deterrence. China has increased its capacity for invading and otherwise using force against Taiwan. China’s will to use force appears to be increasing.  Potential triggers for action include scenarios where China perceives that Taiwan is approaching or has crossed red lines on Taiwan independence.  The arrival of its deadline for achieving unification (possibly by 2049) or a decision on Xi’s part that unifying Taiwan is a vital “legacy” issue could also trigger conflict.  Similarly, if Xi—or the party—faces a threat to his hold on power, he might act against Taiwan to stir nationalism or show progress toward “national rejuvenation.”  China may determine that its window of opportunity for recovering Taiwan peacefully is closing as Taiwan turns more against unification and as the United States increases military capabilities and commitments to Taiwan. China could perceive a moment of opportunity to take Taiwan (due to apparent U.S. distraction or a weak government in Taiwan) or unintended escalation could occur because of a Taiwan crisis or another regional conflict. Successful deterrence requires a concerted and cooperative response from the United States, Japan, and Taiwan.

Recommendations:

  • Invest in U.S. capabilities: The United States should continue investing in military and non-military capabilities to address threats from China in a Taiwan scenario. Assessments should account for prospects that the United States may need to address multiple conflicts or crises simultaneously.
  • Increase survivability of bases in Japan: The United States and Japan should take steps (possibly including missile defenses and medium/intermediate-range missile deployments in Japan) to increase the survivability of U.S. forces and bases and thereby deter a Chinese attack.
  • Tailor responses to shape the decision-making process of specific Chinese leaders: U.S., Japanese, and Taiwanese planning must tailor response plans to the complex set of possible scenarios and take into account—and seek to shape or exploit—the hard-to-discern perceptions and choices of Chinese leaders concerning when and how to act against Taiwan.
  • Define goals and responsibilities: The United States, Japan, and Taiwan should define common strategic and operational goals to address threats from China, clarify defense/military roles in a potential Taiwan-centered conflict, and increase defense and intelligence cooperation. 
  • Improve U.S.-Japan-Taiwan communication: U.S. allies and partners are a force multiplier, but effective defense cooperation requires robust communication. The United States and Japan should improve channels for military-to-military communication with Taiwan at senior and middle levels.
  • Cooperate with like-minded states: Cultivating support from other states that can impose costs on China enhances deterrence, whether through defense cooperation with the United States or through cooperating to impose severe economic costs on China.
     

Finding: Trilateral defense cooperation faces challenges that the parties can address to some extent. Although all three governments recognize that Chinese threats to Taiwan imperil all parties’ security and a Taiwan conflict scenario likely would draw them into a war with China, effective defense cooperation faces several obstacles.  These include a lack of clear and commonly defined goals and roles, as well as policy and other impediments to inter-operability and intelligence sharing (including a U.S. lack of trust about the security of information, operations, and methods shared with Taiwan).  The challenges vary across issues.  All three parties could improve cooperation, addressing the easily resolved issues and building on existing or feasible areas of cooperation and inter-operability.

Recommendations:

  • Clarify roles for the United States, Japan, and Taiwan: The United States, Japan, and Taiwan should pursue clearer consensus on their respective defensive (and counteroffensive) roles in potential Taiwan conflict scenarios in order to enhance deterrence of China.
  • Encourage Taiwan to focus on asymmetric defense and resilience: Taiwan should continue to adopt a “porcupine”/asymmetrical defense strategy to impede and deter a Chinese invasion, including through prepositioning necessary weapons.  The United States, Japan, and Taiwan should work to increase Taiwan’s governmental and physical infrastructure resilience.
  • Reduce restrictions on defense interactions with Taiwan: The United States and especially Japan should consider removing the political/policy obstacles to more open and extensive contacts with higher-level defense and military personnel from Taiwan.
  • Improve intelligence sharing: While recognizing that intelligence sharing will remain limited, the United States, Japan, and Taiwan should work to remove impediments to intelligence sharing by resolving remediable issues including: resistance to sharing intelligence information across agencies within the same government; updating domestic laws to be compatible with other states’ requirements for intelligence sharing; and modifying bilateral intelligence sharing agreements to allow for trilateral or multilateral sharing.  National security concerns, particularly the security of sources and methods, may prevent the sharing of detailed intelligence, at least until all parties can guarantee that they have adequate safeguards against adversary infiltration and espionage efforts, and the mishandling or leaking of intelligence.
  • Build on cooperation in non-military and informal/unofficial channels: The United States, Japan, and Taiwan should explore and encourage possible building blocks for closer cooperation and potentially greater interoperability.  Military personnel could meet with and develop informal communication channels with their counterparts by attending or participating in multilateral non-military activities, like multinational Coast Guard exercises, humanitarian assistance and disaster relief (HA/DR) exercises, or noncombatant evacuation operation (NEO) exercises. Military personnel can also develop informal communication with their counterparts by serving as visiting fellows at the same research institutions, observers at the same Track 2 or Track 1.5 dialogues, or attendees at the same conferences or ceremonies.  The Global Cooperation Training Framework (GCTF)—a rare, existing forum that fully includes Taiwan—can address security-related topics and can allow military personnel to engage with one another on the sidelines of GCTF events. Routinizing these contacts and ensuring key action officers and operational personnel know their counterparts and how to contact them quickly in a crisis  would be of significant benefit to trilateral security cooperation.
     

Finding: Deterrence signaling is difficult and needs improvement. The United States, Japan, and Taiwan have increasingly signaled resolve to coercive threats or the use of force by China against Taiwan.  Even well-designed signals may not be effective, given: the uncertain efficacy of signaling (especially to China, where leadership decision-making is opaque); the difficulty of deterring China from using extreme means to pursue the high-priority goal of unifying Taiwan; China’s possible underestimation of the U.S., Japan, Taiwan, and possibly others; and the lessons China may be drawing from Ukraine. Inconsistent U.S. messaging and a lack of attention to the “assurance” component of deterrence (signaling China that its restraint will lead to the United States refraining from measures threatening China’s vital interest) compound the difficulties.

Recommendations:

  • Focus on preserving the status quo: The United States, Japan, and Taiwan should strive for consistency and a clear emphasis on deterrence when signaling to China (rather than statements likely to be construed as trying to change the status quo).
  • Ensure policy statements are clear and consistent: The United States should ensure high-level statements of U.S. policy achieve and maintain greater clarity, coherence, and consistency over time, particularly when addressing U.S. commitments to Taiwan and the U.S. “one China” policy.
  • When making gestures of support for Taiwan, focus on substantive rather than symbolic measures: The United States should focus on substantive measures to signal support for Taiwan (e.g., arms sales or meaningful participation in substantive international organizations), rather than primarily symbolic gestures (e.g., congressional visits or legislation with few substantive requirements).
  • Dissuade China from learning the wrong lessons from Ukraine: U.S. deterrence signaling to China should make clear that the absence of direct U.S. military intervention in Ukraine does not imply that the United States would refrain from direct intervention in a Taiwan conflict, or that “nuclear blackmail” by China would work against the United States and allies in a Taiwan conflict scenario.
  • Consider offering assurances to China: The United States, Japan, and Taiwan should address the “reassurance” dimension of deterrence and consider providing assurances that they will not target China’s vital interests if it refrains from uses of force and coercion to change the status quo.  Measures of reassurance should be assessed within and among the United States, Japan, and Taiwan, particularly to address risks that they may embolden China or weaken U.S.-Japan-Taiwan mutual confidence.
     

Finding: Deterrence requires strong political foundations. Effective collaborative deterrence rests on political foundations. Trust among the United States, Japan, and Taiwan in one another’s commitments is vital.  International support, from NATO and regional states, would strengthen the political foundations of deterrence.  However, while many states are now warier of China, China’s economic leverage and Taiwan’s diplomatic isolation undermines international support. The vital domestic political foundations of deterrence are questionable in the United States, Japan, and Taiwan.  Effective deterrence requires public appreciation of the threats to their nation’s security posed by China’s threats or actions against Taiwan, and support for the commitments and costs that effective deterrence requires.

Recommendations:

  • Build public buy-in: The United States, Japan, and Taiwan should explain to their publics that a Chinese attack on Taiwan or Japan is a national security threat to their respective countries. They must also be candid with their people that a Chinese attack on Taiwan will include attacks on Japan and on U.S. forces, and that any conflict may devolve into a long and costly struggle.   Public appreciation of these points is important for each party to sustain necessary commitments in a conflict scenario, to have confidence that the others will do so too, and to reduce risks that China will underestimate U.S., Japanese, or Taiwanese resolve.
  • Cultivate international support and cooperation to prepare non-military cost-imposition tools: The United States, Japan, and Taiwan, in conjunction with other like-minded states, should frame the threats from China as threats to the status quo and international norms. This can strengthen cooperation, especially to lay foundations for collectively imposing major collateral economic costs on China, and to address concerns about China’s underestimation of the United States’ and others’ resolve.
  • Support Taiwan’s meaningful international participation: Pushing back against China’s international isolation of Taiwan and supporting meaningful participation for Taiwan in international contexts that have substantive, not merely symbolic, value and that do not appear to seek to change the status quo or conflict with the United States’ “one China” policy can enhance international political foundations of deterrence.
     

Finding: Risks of horizontal escalation and spillover are significant: Contingencies involving Taiwan or other possible flashpoints in the region present major escalation and spillover risks. These include: a Chinese attack on Taiwan that immediately or (less likely) at a second stage involves an attack on U.S. bases and forces in Japan and elsewhere; escalation of an incident or conflict in the East or South China Seas or on the Korean Peninsula or elsewhere; or escalation (possibly inadvertent) from China’s gray zone activities targeting Taiwan.  These scenarios will likely result in great power conflict between the United States and China and involve other states.

Recommendations:

  • Articulate and assess military goals and objectives: Defense planners from the United States, Japan, and Taiwan should establish trilateral consultations (likely through informal channels) to discuss and attempt to align war aims for a potential conflict with China over Taiwan.  Planners should discuss the conflict’s interaction with, or expansion to, potential conflicts elsewhere in the region, including the Korean Peninsula, East China Sea, and South China Sea. Planners should also discuss the possible roles for other U.S. treaty allies, such as South Korea and the Philippines, in various scenarios.
  • Increase Japan’s resilience: The United States and Japan should invest in capabilities to increase the resilience of bases in Japan (e.g., runway repair, air defense, counterstrike capabilities).
     

Finding: China’s nuclear modernization program creates risks of instability and nuclear escalation. China has been modernizing its nuclear forces and increasing its nuclear capabilities.  It aims to maintain a state of mutual nuclear vulnerability with the United States to create a shield for China to use conventional forces in Taiwan (and elsewhere).  This creates a risk of the “stability-instability” paradox, wherein China’s increasing confidence in nuclear deterrence emboldens its leaders to use conventional force.  U.S. extended deterrence efforts face challenges: theater forces and the U.S. homeland are vulnerable to Chinese attack; China may infer from the war in Ukraine that nuclear blackmail works; Japan and Taiwan doubt U.S. commitments, including extended nuclear deterrence, in a conflict with China.

Conflict with China would significantly risk nuclear escalation. Dual-use intermediate range missiles raise risks that strikes on Chinese conventional forces inadvertently endanger its nuclear arsenal, triggering escalation.  China’s nuclear-armed intermediate missiles have ambiguous in-theater purposes, which complicate U.S., Japan, Taiwan, and allied assessments and reactions, including possible U.S. nuclear force deployments.  China’s more potent, varied, and strategic arsenal also poses challenges for U.S. deterrence efforts.  China may adopt “launch on warning” and may not be committed to “no first use.”  Nuclear escalation scenarios may include Chinese efforts to: induce Taiwan’s capitulation or to deter Japan (likely through nuclear blackmail); change the course of a conventional conflict that was going poorly; hit otherwise unreachable U.S. military targets; respond to a perceived use-or-lose situation; or react to an existential threat to the regime.

Recommendations:

  • Invest in conventional deterrence: The United States, Japan, and Taiwan should invest in conventional capabilities to lessen reliance on a nuclear deterrent and vulnerability to nuclear blackmail by China—a tactic China may attempt in the Taiwan context.
  • Modernize U.S. nuclear forces: The United States should continue to invest in nuclear force modernization to uphold the credibility of the U.S. nuclear deterrent and to dissuade China from attempting brinksmanship or nuclear blackmail against the United States in a Taiwan crisis or conflict scenario.
  • Nuclear-sharing agreements: If requested by Japan or others in the region, the United States should consider allowing nuclear-sharing agreements that would permit U.S. allies to host U.S. nuclear weapons under conditions comparable to the NATO nuclear-sharing agreements.
  • Consider missile defense investments: The United States and Japan should consider increasing capacity to intercept or preempt a Chinese nuclear delivery system (although this creates some risk of an escalating spiral of measures from both sides).
  • Consider clarifying U.S. nuclear use policy vis-à-vis China: The United States should strive for greater clarity, coherence, and stability in its nuclear deterrence and use policy specific to China scenarios, including those that may emerge during a potential Taiwan conflict.  The United States also should address problems, if any, that might arise from the differentiated roles of USINDOPACOM as theater command and USSTRATCOM.
  • Increase intelligence sharing: The United States, Japan, and Taiwan should increase intelligence sharing about China’s nuclear capabilities, intentions, and real-time actions to the extent possible.
  • Dissuade China from considering nuclear blackmail: As noted above, U.S. deterrence signaling to China should make clear that the absence of direct U.S. military intervention in Ukraine does not imply that the United States would refrain from direct intervention in a Taiwan conflict, or that “nuclear blackmail” by China would intimidate the United States and U.S. allies in a Taiwan conflict scenario (or any other context).
  • Reassure Japan and Taiwan: The United States, Japan, and Taiwan should proactively discuss collaborative measures they can take to resolve doubts about the credibility of U.S. extended deterrence and U.S. commitment to respond to potential Chinese aggression.
     

Finding: If conflict occurs, de-escalation and termination will be difficult: A conflict that China begins over Taiwan would become a contest for preeminence between the United States and China.  Terminating the conflict would be very difficult.  The stakes for all sides would be very high.  If China begins a conflict, it will have done so with the expectation that it will bear huge costs or face unacceptable alternatives (for example, the existential threat of “losing” Taiwan).  The parties likely have incompatible understandings of the conflict, “red lines,” and strategic cultures.  U.S.-China crisis management and communications mechanisms are weak.  China’s centralized decision-making and behavior in past conflicts and crises (such as hitting hard to gain an early advantage, and foregoing mid-crisis negotiations and third-party mediation) present challenges.  It will be difficult to find mutually acceptable off ramps.

Recommendations:

  • Establish U.S.-China crisis management mechanisms: The United States should consider seeking to establish hotlines and other in-crisis communications mechanisms with China even though China has a history of ignoring hotlines during crises and would be unlikely to use them in a conflict situation.  A U.S. participant suggested hotlines between the USINDOPACOM commander and the PLA Eastern Theater Commander and/or PLA Southern Theater Commander as a specific example.
  • Address possible off ramps: The United States, Japan, and Taiwan should try to design possible off ramps acceptable to all sides for the most likely conflict scenarios.
  • Prioritize avoiding conflict: The challenges of de-escalation after a conflict starts reinforces the imperative for the United States, Japan, and Taiwan to limit the risk of conflict in the first place. The U.S., Japan, and Taiwan should follow the recommendations in this report for effective collaborative deterrence of China and reassurance among the U.S., Japan, and Taiwan.  They should also refrain from actions that could inadvertently trigger a conflict.
     

 

For any questions or comments about the report, please contact the Strategic Trends Division at [email protected]

Strategic Trends Research Initiative: STRI is a sponsored research program that encompasses studies, strategic dialogues, and tabletop exercises. STRI focuses on delivering operationally relevant, credible, timely, and actionable research for DTRA and the warfighter community and is intended to inform future operations, activities, and investments within DTRA.

 
 

[1]     For example, during a speech in 2019, Xi Jinping stated, “两岸长期存在的政治分歧问题是影响两岸关系行稳致远的总根子,总不能一代一代传下去。” (“The long-standing political differences between the two sides of the strait are the root cause that affects the stability and long-term development of cross-strait relations, and cannot be passed down from generation to generation.”) This language was repeated in the 2022 PRC white paper on Taiwan, which stated, “The long-standing political differences between the two sides are the fundamental obstacles to the steady improvement of cross-Straits relations, but we should not allow this problem to be passed down from one generation to the next.”

Xi Jinping, “为实现民族伟大复兴推进祖国和平统一而共同奋斗——在《告台湾同胞书》发表40周年纪念会上的讲话” (“To Achieve the great Rejuvenation of the Nation Promote the Peaceful Reunification of the Motherland and Work Together—Speech at the 40th Anniversary Commemoration of the ‘Message to Compatriots in Taiwan’”), Xinhua, 2 January 2019, http://www.xinhuanet.com/politics/leaders/2019-01/02/c_1123937757.htm

PRC State Council Taiwan Affairs Office and PRC State Council Information Office, “The Taiwan Question and China’s Reunification in the New Era,” Xinhua, 10 August 2022, https://english.news.cn/20220810/df9d3b8702154b34bbf1d451b99bf64a/c.html

[2]      PRC State Council Taiwan Affairs Office and PRC State Council Information Office, “The Taiwan Question and China’s Reunification in the New Era,” Xinhua, 10 August 2022, https://english.news.cn/20220810/df9d3b8702154b34bbf1d451b99bf64a/c.html[3]      “Japan-U.S. Joint Leaders’ Statement: Strengthening the Free and Open International Order,” The White House, 23 May 2022, https://www.whitehouse.gov/briefing-room/statements-releases/2022/05/23/japan-u-s-joint-leaders-statement-strengthening-the-free-and-open-international-order/
[4]      Josh Rogin, “Has Biden Altered the One-China Policy? Yes and No.” Washington Post, 13 October 2022, https://www.washingtonpost.com/opinions/2022/10/13/biden-china-taiwan-policy-contradiction/

       Zack Cooper, “The Fourth Taiwan Strait Slip-Up,” AEI, 19 September 2022, https://www.aei.org/foreign-and-defense-policy/the-fourth-taiwan-strait-slip-up/
[5]      Jude Blanchette, Thomas Chsristensen, et al. “Avoiding War Over Taiwan,” Asia Society, 12 October 2022, https://china.ucsd.edu/_files/2022-avoiding-war-over-taiwan.pdf

       Thomas Christensen, M. Taylor Fravel, et al. “How to Avoid War Over Taiwan: Threats, Assurances, and Effective Deterrence,” Foreign Affairs, 13 October 2022, https://www.foreignaffairs.com/china/how-avoid-war-over-taiwan.
[6]      Brendan Rittenhouse Green and Caitlin Talmadge, “Then What? Assessing the Military Implications of Chinese Control of Taiwan,” International Security, Vol. 47, No. 1, Summer 2022, https://direct.mit.edu/isec/article-abstract/47/1/7/112577/Then-What-Assessing-the-Military-Implications-of?redirectedFrom=fulltext

       Brendan Rittenhouse Green and Caitlin Talmadge, “The Consequences of Conquest: Why Indo-Pacific Power Hinges on Taiwan,” Foreign Affairs, July/August 2022, https://www.foreignaffairs.com/articles/china/2022-06-16/consequences-conquest-taiwan-indo-pacific
[7]      “Joint Statement of the U.S.-Japan Security Consultative Committee (“2+2”),” U.S. Department of Defense, 6 January 2022, https://www.defense.gov/News/Releases/Release/Article/2891314/joint-statement-of-the-us-japan-security-consultative-committee-22/
[8]      Fiscal Year 2023 Strategic Forces Posture Hearing, Before the House Armed Services Committee on Strategic Forces, 117th Congress (2022) (statement of Charles A. Richard, Commander, United States Strategic Command) https://www.stratcom.mil/Portals/8/Documents/2022%20USSTRATCOM%20Posture%20Statement.pdf?ver=CUIoOCLyos9xe9C9I0XjMQ%3D%3D

Office of the Secretary of Defense, Annual Report to Congress: Military and Security Developments Involving the People’s Republic of China, 2021, https://media.defense.gov/2021/Nov/03/2002885874/-1/-1/0/2021-CMPR-FINAL.PDF
[9]      Missile Defense Project, “DF-21 (CSS-5),” Missile Threat, Center for Strategic and International Studies, last modified 28 March 2022, https://missilethreat.csis.org/missile/df-21/

       Missile Defense Project, “DF-26,” Missile Threat, Center for Strategic and International Studies, last modified 6 August 2021, https://missilethreat.csis.org/missile/dong-feng-26-df-26/
[10]    Caitlin Talmadge, “The U.S.-China Nuclear Relationship: Growing Escalation Risks and Implications for the Future,” Hearing on China’s Nuclear Forces [Testimony before the U.S.-China Economic and Security Review Commission], 7 June 2021, https://www.uscc.gov/sites/default/files/2021-06/Caitlin_Talmadge_Testimony.pdf
[11]    Matthew Funaiole, Joseph Bermudez, and Brian Hart, “A Glimpse of Chinese Ballistic Missile Submarines,” Center for Strategic and International Studies, 4 August 2021, https://www.csis.org/analysis/glimpse-chinese-ballistic-missile-submarines
[12]    Philip Saunders, “Testimony before the U.S.-China Economic and Security Review Commission Hearing on China’s Nuclear Forces,” 10 June 2021, https://www.uscc.gov/sites/default/files/2021-06/Phillip_Saunders_Testimony.pdf
[13]    Office of the Secretary of Defense, Annual Report to Congress: Military and Security Developments Involving the People’s Republic of China, 2021, https://media.defense.gov/2021/Nov/03/2002885874/-1/-1/0/2021-CMPR-FINAL.PDF
[14]    Ibid.
[15]    Jacques deLisle, “China’s Russia/Ukraine Problem, and Why It’s Bad for Almost Everyone Else Too,” Orbis, Vol. 66, No. 3, Summer 2022, pp.402-423, https://doi.org/10.1016/j.orbis.2022.05.009

Joel Wuthnow, “Rightsizing Chinese Military Lessons from Ukraine,” Strategic Forum, September 2022, https://inss.ndu.edu/Portals/68/Documents/stratforum/SF-311.pdf

       Bonny Lin and John Culver, “China’s Taiwan Invasion Plans May Get Faster and Deadlier,” Foreign Policy, 19 April 2022, https://foreignpolicy.com/2022/04/19/china-invasion-ukraine-taiwan/

       David Sacks, “What Is China Learning From Russia’s War in Ukraine?” Foreign Affairs, 16 May 2022, https://www.foreignaffairs.com/articles/china/2022-05-16/what-china-learning-russias-war-ukraine

       David Finkelstein, “Beijing’s Ukrainian Battle Lab,” War on the Rocks, 2 May 2022, https://warontherocks.com/2022/05/beijings-ukrainian-battle-lab/
[16]    Masashi Murano, “US-China Mutual Vulnerability: A Japanese Perspective,” US-China Mutual Vulnerability Perspectives on the Debate, Pacific Forum Issues and Insights, Vol. 22, Series 2, May 2022, https://pacforum.org/wp-content/uploads/2022/05/Issues-Insights-Vol.-22-SR-2.pdf
[17]    Prime Minister Sato Eisaku: “本土としては、私どもは核の三原則、核を製造せず、核を持たない、持ち込みを許さない、これははっきり言っている。” (“As a mainland, we clearly state the three principles of nuclear weapons: no nuclear production, no nuclear possession, no nuclear import.”)

House of Representatives Budget Committee No. 2 Remark #103 [Meeting Minutes], 57th National Diet, 11 December 1967, https://kokkai.ndl.go.jp/txt/105705261X00219671211/103
[18]    The DOD’s annual report to Congress on China’s military power estimates that China’s nuclear expansion may enable China to accumulate up to 700 deliverable nuclear warheads by 2027 and approximately 1,000 warheads by 2030.

Office of the Secretary of Defense, Annual Report to Congress: Military and Security Developments Involving the People’s Republic of China, 2021, https://media.defense.gov/2021/Nov/03/2002885874/-1/-1/0/2021-CMPR-FINAL.PDF