Is the Taiwan Strait Still a Flash Point?
Is the Taiwan Strait
Still a Flash Point?
Scott L. Kastner
Rethinking the Prospects for Armed
Conºict between China and Taiwan
B
efore
2008,
the
Taiwan Strait was widely viewed as a dangerous ºash point for conºict.
The issue of Taiwan’s sovereign status was a persistent source of tensions in
U.S.-China relations, leading one of the most prominent U.S. experts on Asia
to refer to Taiwan as “the only issue in the world today that could realistically
lead to war between two major powers.”1 Since 2008, however, relations between the People’s Republic of China (PRC) and Taiwan have improved dramatically. Ofªcials from the two sides have engaged in frequent dialogues,
resulting in numerous cooperative agreements (including, most notably, the
2010 Economic Cooperation Framework Agreement). Given this unprecedented period of détente, does it still make sense to view the Taiwan Strait as
a ºash point for conºict? To what degree is a China-Taiwan military conºict a
continued possibility? Are the risks of armed conºict likely to increase or
decrease in the years ahead? What are the implications for U.S. policy in
the region?
Analysts differ on the degree to which the Taiwan Strait remains at risk of
armed conºict. Some scholars point to factors such as deepening cross-strait
economic exchange, frequent high-level contacts, greater institutionalization of
the cross-strait relationship, and pragmatism in both Beijing and Taipei as reasons to be optimistic about the prospects for continued stability in the Taiwan
Strait.2 Other analysts, however, are less sanguine about the future of ChinaScott L. Kastner is Associate Professor in the Department of Government and Politics at the University of
Maryland, College Park.
The author thanks Ping-Kuei Chen for helpful research assistance for this article. For valuable
comments on earlier drafts, the author thanks Monique Chu, Avery Goldstein, Todd Hall, Margaret Pearson, Ni Ting, Stein Tonnesson, the anonymous reviewers, and participants in workshops at
Stanford University, the University of Oxford, the University of Tokyo, and the Fourth Annual
East Asian Peace Conference.
1. Alan D. Romberg, “Rein in at the Brink of the Precipice: American Policy toward Taiwan and
U.S.-PRC Relations” (Washington, D.C.: Stimson Center, 2003), p. 14.
2. For relatively optimistic analyses of cross-strait relations, see, for instance, Bruce Gilley, “Not So
Dire Straits,” Foreign Affairs, Vol. 89, No. 1 (January/February 2010), pp. 44–60; Li Peng, “The Nature of the Economic Cooperation Framework Agreement and Its Implications for Peaceful Development of Cross-Strait Relations,” in Weixing Hu, ed., New Dynamics in Cross–Taiwan Strait
Relations: How Far Can the Rapprochement Go? (Hoboken, N.J.: Taylor and Francis, 2013), pp. 47–59;
and Phillip C. Saunders and Scott L. Kastner, “Bridge over Troubled Water? Envisioning a ChinaTaiwan Peace Agreement,” International Security, Vol. 33, No. 4 (Spring 2009), pp. 87–114. Emphasizing factors such as a “fading” independence movement in Taiwan and robust deterrence in the
Taiwan Strait, Robert S. Ross was relatively optimistic about the prospects for cross-strait stability
even during the turbulent early 2000s. See Ross, “Taiwan’s Fading Independence Movement,” ForInternational Security, Vol. 40, No. 3 (Winter 2015/16), pp. 54–92, doi:10.1162/ISEC_a_00227
© 2016 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.
54
Is the Taiwan Strait Still a Flash Point? 55
Taiwan relations. Those who are more pessimistic about the relationship tend
to highlight trends such as the PRC’s rapidly growing military capabilities
(which increasingly call into question Taiwan’s capacity for self-defense),
social and political trends in Taiwan that suggest little appetite for further
cross-strait economic liberalization or political integration, and the continued
presence of a fundamental underlying sovereignty dispute dividing Beijing
and Taipei.3 Part of the reason for these disagreements is that the cross-strait
relationship is characterized by complex, and at times competing, trends. Economic integration and increased institutionalization across the Taiwan Strait
should contribute to stability. In some ways, public opinion in Taiwan seems to
push toward stability (most Taiwanese, for example, are relatively pragmatic
on sovereignty issues and oppose formal independence if it were to trigger
conºict with China). In other ways, however, public opinion in Taiwan highlights the continued intractability of the underlying dispute (support on the
island for uniªcation continues to drop even as the PRC continues to view uniªcation as an important national goal). Growing PRC military power increases
the feasibility of Chinese coercion against Taiwan, but it could also make
Chinese leaders more conªdent about long-term trends and less inclined to act
rashly. What is needed, in short, is systematic analysis that considers how
these different trends ªt together in shaping the prospects for conºict and
peace in the Taiwan Strait.
I aim to provide this type of analysis. To do so, I ªrst consider three plausible conºict scenarios that frequently worried analysts of cross-strait relations
in the turbulent years prior to 2008. I then consider how fundamental trends in
eign Affairs, Vol. 85, No. 2 (March/April 2006), pp. 141–148; and Ross, “Navigating the Taiwan
Strait: Deterrence, Escalation Dominance, and U.S.-China Relations,” International Security, Vol. 27,
No. 2 (Fall 2002), pp. 47–85.
3. For recent pessimistic views on cross-strait relations, see, for instance, Denny Roy, “Collision
Course: The Looming U.S.-China Showdown over Taiwan,” National Interest, February 21, 2015,
http://www.nationalinterest.org/feature/collision-course-the-looming-us-china-showdown-overtaiwan-12293; Bonnie S. Glaser and Jacqueline Vitello, “Xi Jinping’s Great Game: Are China and
Taiwan Heading towards Trouble?” National Interest, July 16, 2015, http://nationalinterest.org/
feature/xi-jinpings-great-game-are-china-taiwan-headed-towards-13346; Sheryn Lee and Benjamin Schreer, “The Taiwan Strait: Still Dangerous,” Survival, Vol. 55, No. 3 (June/July 2013), pp. 55–
62; and David A. Shlapak et al., A Question of Balance: Political Context and Military Aspects of the
China-Taiwan Dispute (Santa Monica, Calif.: RAND Corporation, 2009). Weixing Hu makes an interesting middle-ground argument, suggesting that the cross-strait relationship is heading toward
an “involutional process” in which relations will probably not improve further, but where the
status quo is likely to remain relatively stable (and hence unlikely to degenerate into armed conºict). See Hu, “Prospects of Cross–Taiwan Strait Relations: Toward an Involutional Process?” in
Hu, New Dynamics in Cross–Taiwan Strait Relations, pp. 218–234. For a comprehensive overview of
recent trends in cross-strait relations and detailed discussion of continued challenges and possible
future scenarios, see Richard C. Bush, Uncharted Strait: The Future of China-Taiwan Relations (Washington, D.C.: Brookings Institution Press, 2013).
International Security 40:3 56
cross-strait relations, such as rapidly growing Chinese military power and
deepening cross-strait economic exchange, are affecting the likelihood that any
of these scenarios will emerge as future concerns. To preview, my analysis suggests that these trends are, on balance, stabilizing. Although the relationship
will continue to be characterized by periodic tensions, the risk of armed
conºict has been declining and is likely to continue to decline in the years
ahead. At the same time, however, the cross-strait relationship has not been
fundamentally transformed. Military conºict remains a real possibility, and I
emphasize in particular that a rapidly shifting balance of military power has
the potential to be highly destabilizing if it comes to dominate the effects of
other trends such as economic integration. I conclude the article with policy
implications for the United States.
Conºict Scenarios in the Taiwan Strait
The underlying dispute between the PRC and Taiwan dates to the end of the
Chinese Civil War, when the losing Nationalist Party (the Kuomingtang, or
KMT) retreated to Taiwan along with its government (the Republic of China,
or the ROC), while the Chinese Communist Party established the PRC on the
Chinese mainland. The KMT continued to view the ROC as the legitimate government of all of China, even as the PRC viewed Taiwan as a part of China that
must ultimately be reunited with the rest of the country. The cross-strait relationship was characterized by tensions and periodic crises during the 1950s
and 1960s, and the ROC in 1954 signed an alliance treaty with the United
States. Washington did not establish diplomatic ties with the PRC until 1979, at
which point it severed diplomatic ties and abrogated its alliance with the ROC.
China-Taiwan relations stabilized to some degree during the 1980s and
early 1990s, as the PRC adopted a less confrontational approach to Taiwan
(premised on peaceful reuniªcation) and the ROC government relaxed restrictions on travel and trade across the Taiwan Strait (which had been banned
since 1949). Yet fundamental political changes in Taiwan during the 1980s and
1990s helped to restructure the underlying sovereignty dispute in a way that
gave rise to renewed tensions by the mid-1990s. For several decades after 1949,
the KMT had ruled Taiwan in an authoritarian fashion; any advocacy of
Taiwan independence was strictly forbidden. During the late 1980s and early
1990s, however, Taiwan underwent a democratic transition. By the early 1990s,
a growing number of Taiwanese were openly questioning whether Taiwan
was a part of China, and Taiwan’s major opposition party (the Democratic
Progressive Party, or DPP) formally supported Taiwan independence. Meanwhile, the ROC president at the time, Lee Teng-hui, was calling for a greater
Is the Taiwan Strait Still a Flash Point? 57
international presence for Taiwan to help reverse decades of growing international isolation (by the 1990s, most countries of the world had severed relations with Taiwan and recognized the PRC instead). The PRC increasingly
feared that Taiwan under Lee was moving toward formal independence. As
Taiwan prepared for its ªrst direct presidential election in 1996, a prolonged
crisis erupted in the Taiwan Strait, culminating with PRC missile tests in
waters near Taiwan and the U.S. decision to deploy two aircraft carrier battle groups to the area. After some initial signs of thawing, the relationship
remained icy throughout the presidency of Chen Shui-bian (2000–08) of
the DPP.4
Relations across the Taiwan Strait have stabilized tremendously since 2008,
when Ma Ying-jeou, of the KMT, was elected Taiwan’s president. Ma had campaigned on an explicitly pro-status quo platform that rejects Taiwan independence. Shortly after Ma entered ofªce, the PRC and Taiwan were able to restart
their cross-strait dialogue (which had been moribund for nearly a decade) on
the basis of a construct called the “1992 consensus.”5 Since then, the two sides
have reached numerous bilateral agreements and engaged in regular dialogue,
culminating in a November 2015 meeting in Singapore between President Ma
and Chinese President Xi Jinping.
Despite the recent détente, however, future prospects are uncertain at the
time of this writing. The DPP, which rejects the 1992 consensus, will soon return to power after sweeping Taiwan’s 2016 presidential and legislative elections, and many analysts have become pessimistic about the future of the
cross-strait relationship. To what degree should scholars and policymakers
4. For a good history of Taiwan’s political transition and cross-strait relations, see Denny Roy, Taiwan: A Political History (Ithaca, N.Y.: Cornell University Press, 2003).
5. The 1992 consensus refers to an understanding reached by quasi-ofªcial representatives of the
PRC and ROC governments in 1992. At the time, the two sides were laying the groundwork for a
dialogue that would take place in Singapore the following year, but they were grappling with how
to deal with the issue of whether mainland China and Taiwan belong to “one China.” The two
sides agreed to express their views separately. The PRC’s view was that both sides upheld a oneChina principle, but that the meaning of one China would not be addressed in routine consultations. The ROC view was that each side upheld a one-China principle, but each had its own interpretation of what one China meant (for Taipei, one China meant the ROC). The term “1992
consensus” was invented several years later by a KMT ofªcial, Su Chi, to describe the accommodative approach both sides took on the one-China issue at the time. Ma Ying-jeou, in embracing
the 1992 consensus as a basis for resumed dialogue, has continued to interpret it as “one China,
separate interpretations.” Although the PRC does not agree with the “separate interpretations”
concept, it has also come to embrace the consensus, because the consensus implies that both
sides endorse a one-China principle. The DPP rejects the 1992 consensus (and indeed argues that
no real consensus was achieved in 1992). For discussions of this topic, see Bush, Uncharted Strait,
especially chap. 2; and Chi-hung Wei, “Producing and Reproducing the 1992 Consensus: The
Sociolinguistic Construction of the Political Economy of China-Taiwan Relations,” Asian Security,
Vol. 11, No. 1 (2015), pp. 72–88.
International Security 40:3 58
again start to worry about the possibility of armed conºict in the Taiwan
Strait? To answer this question, it is critical to ªrst have a solid understanding
of how conºict could occur in the Taiwan Strait. In the remainder of this section, I outline three plausible scenarios for a China-Taiwan military conºict.
These scenarios represent my effort to synthesize and, where appropriate, to
tease out the logic of conºict scenarios highlighted in the literature on crossstrait relations. Each of the scenarios drew considerable attention from analysts of the cross-strait relationship in the turbulent 1990s and early 2000s.6
the redline conºict scenario
During the Lee and Chen presidencies, the scenario for cross-strait conºict that
drew the most attention from analysts of the China-Taiwan relationship involved a “revisionist”7 Taiwan trying to formalize its independent status (or
taking steps in that direction8), and triggering a PRC military response by
crossing PRC “redlines.”9 A simple model helps to illustrate this scenario (see
ªgure 1).10 Assume that Taiwan and the PRC are bargaining, implicitly or explicitly, over Taiwan’s sovereign status, which can be represented on a single
dimension ranging from formal uniªcation with China (U) at one extreme to a
formally independent Taiwan (I) on the other. The status quo (SQ) lies somewhere in between.11 Assume further that a revisionist Taiwan’s leadership pre-
6. The discussion in this section draws from and builds on Scott L. Kastner, “Rethinking the Prospects for Conºict in the Taiwan Strait,” in Monique Ming-chin Chu and Kastner, eds., Globalization
and Security Relations across the Taiwan Strait: In the Shadow of China (London: Routledge, 2015),
pp. 42–60.
7. I use the term “revisionist” here simply to denote dissatisfaction with the status quo in crossstrait relations. On the sources of revisionism in Taiwan’s foreign policy, see Robert S. Ross, “Explaining Taiwan’s Revisionist Diplomacy,” Journal of Contemporary China, Vol. 15, No. 48 (August
2006), pp. 443–458.
8. Such steps might include the creation of a new constitution to replace the ROC constitution, a
step advocated by Chen Shui-bian during his presidency.
9. For examples of studies that outline this sort of scenario, see Richard C. Bush and Michael E.
O’Hanlon, A War Like No Other: The Truth about China’s Challenge to America (Hoboken, N.J.: John
Wiley and Sons, 2007); Ted Galen Carpenter, America’s Coming War with China: A Collision Course
over Taiwan (New York: Palgrave Macmillan, 2005); and Gabe T. Wang, China and the Taiwan Issue:
Impending War at the Taiwan Strait (Lanham, Md.: University Press of America, 2006). Many studies
that have considered more speciªcally the feasibility or likely efªcacy of PRC military force
against Taiwan also point toward Taiwanese revisionism as a likely trigger of Chinese use of force.
See, for instance, Michael A. Glosny, “Strangulation from the Sea? A PRC Submarine Blockade of
Taiwan,” International Security, Vol. 28, No. 4 (Spring 2004), pp. 125–160.
10. The logic of the model draws heavily from James D. Fearon, “Rationalist Explanations for
War,” International Organization, Vol. 49, No. 3 (Summer 1995), pp. 379–414.
11. The assumption that Taiwan’s status can be thought of as a continuum is reasonable. The
status quo can be conceptualized broadly as de facto independence that is neither declared nor internationally recognized. Outcomes further toward the independence end of the continuum
would include an established legal groundwork for formal independence without actual declaration (such as a wholly new constitution, as former President Chen Shui-bian has proposed),
Is the Taiwan Strait Still a Flash Point? 59
Figure 1. A Model of Conºict in the Taiwan Strait
U: unification; I: independence; T: Taiwan’s war utility; W: the war outcome; SQ: status quo;
R: PRC’s war utility (PRC reversion point); Cc: PRC’s costs of war; and CT: Taiwan’s costs
of war
fers an outcome closer to I,12 and that China prefers an outcome closer to U.
Assume that if the two sides were to ªght a war, Taiwan’s status at the end of
the war would lie at point W.13 Finally, assume that the utility each side would
receive from the war outcome (W) would be reduced by the costs of actually
ªghting the war, represented by cc for China and ct for Taiwan. China, in this
declared independence (not internationally recognized), and widely recognized declared independence. Outcomes closer to the uniªcation end of the continuum would include a loose
confederal arrangement (as former KMT Chair and presidential candidate Lien Chan has proposed), a one-country, two-systems arrangement where Taiwan is afforded more autonomy than
Hong Kong (as the PRC has proposed), and a Hong Kong–style one-country, two-systems arrangement. For studies that conceptualize Taiwan’s status as a continuum of potential outcomes, see, for
instance, David M. Lampton, “Preparing for a Better Time in Cross-Strait Relations: Short-Term
Stalemate, Possible Medium-Term Opportunities,” in Donald S. Zagoria, ed., Breaking the ChinaTaiwan Impasse (Westport, Conn.: Praeger, 2003), pp. 105–114; and Richard C. Bush, Untying the
Knot: Making Peace in the Taiwan Strait (Washington, D.C.: Brookings Institution Press, 2005).
12. The degree to which this assumption reºects a reasonable approximation of the Taiwanese
government’s preferences obviously varies. The current Ma administration, for instance, almost
certainly does not view something approximating formal Taiwan independence as its ideal point.
To the degree that the Taiwanese government is satisªed with the current status quo in cross-strait
relations, the redline conºict scenario—as described here—is less relevant.
13. The logic of this model does not hinge on the exact location of W. W can be thought of either as
the expected amount of sovereignty Taiwan would hold at the conclusion of a war or as the probability that Taiwan (presumably with U.S. backing) would win the war and thus be able to claim
its preferred level of sovereignty.
International Security 40:3 60
model, should prefer all bargains to the left of R over ªghting a war. As such,
so long as Taiwan does not try to attain a level of sovereign status to the right
of R, it should avoid triggering a war with the PRC. R, then, represents China’s
reversion point, or “redline.”
Why would Taiwan cross this redline? Obviously, Taiwan would much prefer to avoid war by choosing a level of status to the left of China’s redline, because Taiwan’s war utility (point T in the ªgure) is far worse. If Taiwan could
know with certainty where China’s redline lies, it should choose a level of sovereign status just to the left of that redline and leave it at that. But Taiwan cannot know with certainty where the redline sits, because it cannot know with
certainty how China assesses the likely outcome of a war (W) and the PRC’s
likely costs of war (cc), both of which hinge on an array of factors, such as
PRC beliefs about the efªcacy of its military and its expectations of likely U.S.
behavior in the event of a war.14 Indeed, the PRC has clear incentives to overstate its resolve (meaning, in essence, to understate its expected costs of war),
so as to convince Taiwan’s leadership that its redline lies farther left (i.e., farther from independence and closer to uniªcation) than might truly be the
case. Taiwan, in turn, knows that Beijing has incentives to bluff.15 Indeed,
the redlines that Beijing communicates in practice can themselves be ambiguous: although the PRC has suggested that a formal declaration of independence would trigger a military response,16 it has been less clear concerning
what steps short of formal independence would be sufªcient to trigger such
a response.17
The above scenario can be conceptualized as a bargaining failure that arises
as a consequence of imperfect information (and incentives to misrepresent
that information). Of course, the bargaining problem itself exists only to the
extent the two sides have intense, zero-sum preferences relating to the issue in
dispute, Taiwan’s sovereign status. Taiwan’s leaders must strongly desire to alter Taiwan’s status, and PRC leaders must strongly oppose such changes.
Zero-sum preferences are not sufªcient to trigger conºict, however, because
war would be almost unimaginably costly for Taiwan (and possibly China
as well). As Steve Tsang emphasizes, no Taiwanese president would “in14. The PRC, for instance, would have a much more pessimistic view of a war’s likely outcome if
it believed that the United States would be likely to intervene militarily. Even U.S. actions short of
military intervention, such as economic sanctions, could have a large effect on expected PRC costs.
15. On private information, with incentives to misrepresent that information, as a cause of war,
see especially Fearon, “Rationalist Explanations for War.”
16. The Anti-Secession Law of 2005, for instance, emphasizes that the PRC shall use “nonpeaceful” means in the event of Taiwan’s “secession” from China.
17. On ambiguity in PRC redlines, see Steve Tsang, ed., If China Attacks Taiwan: Military Strategy,
Politics, and Economics (London: Routledge, 2006).
Is the Taiwan Strait Still a Flash Point? 61
tentionally announce a policy that will immediately cause China to attack
Taiwan.”18 Rather, war arises in this scenario because there is some uncertainty
regarding the precise location of Beijing’s redlines, and it can be hard for China
to credibly communicate the true location of those redlines. As such, a revisionist Taiwan could trigger a military conºict by unintentionally crossing a
Chinese redline.
the pessimistic trends analysis scenario
In a second scenario, conºict arises between China and Taiwan as a consequence of pessimism in Beijing about long-term trends in the Taiwan Strait.
Thomas Christensen, in particular, articulated this scenario in a series of articles published during the Chen Shui-bian administration.19 Christensen
warned that a defensively minded, pro-status quo Beijing could initiate the use
of military force even in the absence of a speciªc Taiwan action that crosses
PRC redlines. Triggering such a decision would be deep pessimism in Beijing
about long-term trends in Taiwan’s politics and U.S.-Taiwan relations, combined with a belief that a military attack could alter the trajectory of
those trends. More speciªcally, trends that make PRC elites believe that they
are faced with a “closing window of opportunity” to prevent Taiwan independence could lead to the decision to use military force as a way to arrest
those trends.20
As Christensen points out, PRC pessimism was especially pronounced in
the months following Chen Shui-bian’s narrow re-election as Taiwan’s president in 2004, when there also appeared to be a realistic chance that the DPP
could capture the Legislative Yuan in year-end elections.21 Chen emphasized,
18. Ibid., p. 1.
19. See Thomas J. Christensen, “Posing Problems without Catching Up: China’s Rise and Challenges for U.S. Security Policy,” International Security, Vol. 25, No. 4 (Spring 2001), pp. 5–40;
Thomas J. Christensen, “Beijing’s Views of Taiwan and the United States in Early 2002: The Renaissance of Pessimism,” China Leadership Monitor, Summer 2002, http://media.hoover.org/sites/
default/ªles/documents/clm3_TC.pdf; Thomas J. Christensen, “The Contemporary Security Dilemma: Deterring a Taiwan Conºict,” Washington Quarterly, Vol. 25, No. 4 (Autumn 2002), pp. 7–
21; Thomas J. Christensen, “Taiwan’s Legislative Yuan Elections and Cross-Strait Security
Relations: Reduced Tensions and Remaining Challenges,” China Leadership Monitor, Winter 2005,
http://www.hoover.org/publications/clm/issues/2903801.html; and Thomas J. Christensen,
“Windows and War: Trend Analysis and Beijing’s Use of Force,” in Alastair Iain Johnston and Robert S. Ross, eds., New Directions in the Study of China’s Foreign Policy (Stanford, Calif.: Stanford University Press, 2006), pp. 50–85. On PRC willingness to use force to alter unfavorable trend lines, see
also Allen S. Whiting, “China’s Use of Force, 1950–96, and Taiwan,” International Security, Vol. 26,
No. 2 (Fall 2001), pp. 103–131; and M. Taylor Fravel, Strong Borders, Secure Nation: Cooperation and
Conºict in China’s Territorial Disputes (Princeton, N.J.: Princeton University Press, 2008).
20. Christensen, “Posing Problems without Catching Up,” p. 36; and Christensen, “Windows and
War.”
21. Christensen, “Taiwan’s Legislative Yuan Elections and Cross-Strait Security Relations.” I also
International Security 40:3 62
for instance, that a DPP majority in this body would enable him to push
through a new constitution that would be more “suitable” for Taiwan. From
Beijing’s vantage point, a DPP ªrmly in control of Taiwan’s political system
could institute political changes (such as constitutional reform meant to
redeªne Taiwan’s sovereign status) that would be difªcult to undo later on.
This type of scenario can be thought of as a credible commitment problem
that resembles the type of commitment problem that can emerge as a consequence of long-term changes in the balance of power between countries. Imagine a powerful country, A, and a weaker country, B, that is expected to become
stronger in the future. If A fears that B, when it becomes powerful, will adopt
policies at odds with A’s interests, A might be tempted to act today to halt B’s
rise while it still has the power to do so. Recognizing this danger, B might
promise to act with restraint when it becomes powerful. B’s commitment lacks
credibility, however: once B becomes powerful, A will be unable to hold B to
its word. If A launches a preventive war meant to arrest B’s rise, the root cause
of the conºict ultimately lies in B’s inability to commit credibly to restraint as it
becomes stronger.22 Robert Powell writes that some of the largest wars in history have been rooted in this sort of dynamic.23
In the Taiwan Strait case, the logic of prevention centers on political trends
in Taiwan and the nature of the U.S. commitment to Taiwan, not on trends in
the balance of power per se. As such, the comparison to power transition theory is not exact, but the basic logic of the underlying credible commitment
problem is similar. For instance, if Chinese leaders were to become deeply
pessimistic about long-term political and societal trends in Taiwan, it is not
clear that today’s Taiwan leaders could sufªciently assuage those concerns.
A Taiwan president could promise that Taiwan would not preclude the possibility of future uniªcation under the right circumstances, and a future Taiwan
government could even pass laws to this effect. These sorts of promises would
lack credibility, however, if trend lines in Taiwan society were to suggest that
any sort of uniªcation would be fundamentally anathema to most future
Taiwan residents. That is, if Chinese leaders were to conclude that long-term
trends suggest an ever deepening sense of Taiwan-centric identity for most
Taiwanese, and a view among most Taiwanese of a China that is a distinct
“other,” then these leaders may come to view future uniªcation as impossible
found considerable pessimism in my own interviews with analysts in Shanghai during the summer of 2004.
22. Fearon, “Rationalist Explanations for War.”
23. Robert Powell, “War as a Commitment Problem,” International Organization, Vol. 60, No. 1 (January 2006), pp. 169–203. For a detailed discussion of the preventive logic, see Stephen Van Evera,
Causes of War: Power and the Roots of Conºict (Ithaca, N.Y.: Cornell University Press, 1999); and
Jack S. Levy, “Declining Power and the Preventive Motivation for War,” World Politics, Vol. 40,
No. 1 (October 1987), pp. 82–107.
Is the Taiwan Strait Still a Flash Point? 63
in practical terms—regardless of the policies or promises of current Taiwan
leaders. Such thinking is dangerous because it suggests that Chinese leaders
could conclude that war is necessary in the short term to prevent Taiwan’s permanent separation from China: delay only makes success less likely.
inadvertent war and the logic of preemption
In a third conºict scenario, an accident or incident is misinterpreted by one
side as the initiation of hostilities, leading to a rapid escalation.24 Analysts
have suggested numerous possibilities along these lines in the PRC-Taiwan relationship. Kenneth Allen, for instance, notes that frequent ROC and PRC military ºights over the Taiwan Strait raise the risk of an inadvertent collision; if
such a collision is misinterpreted as an intentional provocation, it could lead
one or both sides to scramble more aircraft to the area, potentially triggering
an escalatory spiral.25 Allen notes, for instance, that Beijing might view such a
situation as the “right time to implement its operations plan, especially if the
United States is not in a position to come to Taiwan’s assistance in a hurry.” At
the same time, if the PRC is rapidly building up forces, “Taiwan might want to
retaliate with a strike on mainland targets while its forces are still capable.”26
Bernard Cole likewise describes a naval scenario where some of the escalation
is unintentional,27 and James Mulvenon highlights the possibility that Chinese
missile tests conducted near Taiwan, and intended as a signal of resolve, could
be misinterpreted by Taiwanese leaders as the beginning of a PRC attack; presumably this would especially be the case if a missile were to unintentionally
strike some Taiwanese asset.28
These sorts of scenarios are ultimately grounded—at least in part—in a logic
of preemptive war.29 Escalation occurs because one side believes that an attack
is imminent and that there are advantages to acting ªrst. Beijing might feel the
24. On the possibility of accidental war in the Taiwan Strait, see, for instance, Craig S. Smith,
“War Talk Is Cheap. War Isn’t. Keep Talking,” New York Times, March 19, 2000, http://www
.nytimes.com/2000/03/19/weekinreview/the-world-war-talk-is-cheap-war-isn-t-keep-talking
.html?pagewanted⫽all&src⫽pm.
25. Kenneth W. Allen, “Air Force Deterrence and Escalation Calculations for a Taiwan Strait
Conºict: China, Taiwan, and the United States,” in Michael D. Swaine, Andrew N.D. Yang, and
Evan S. Medeiros, eds., Assessing the Threat: The Chinese Military and Taiwan’s Security (Washington,
D.C.: Carnegie Endowment for International Peace, 2007), pp. 153–183.
26. Ibid., p. 171.
27. For instance, frigate commanders’ doubts about rules of engagement in a tense environment,
or overly aggressive behavior by individual pilots, could trigger incidents unintended by national
leaders. See Bernard D. Cole, “The Military Instrument of Statecraft at Sea: Naval Options in an
Escalatory Scenario Involving Taiwan: 2007–2016,” in Swaine, Yang, and Medeiros, Assessing the
Threat, pp. 185–209.
28. James Mulvenon, “Missile Defenses and the Taiwan Scenario” (Washington, D.C.: Stimson
Center, 2002), http://www.stimson.org/images/uploads/research-pdfs/44.pdf.
29. On preemption, see especially Thomas C. Schelling, Arms and Inºuence (New Haven, Conn.:
Yale University Press. 1966), chap. 6.
International Security 40:3 64
need to act quickly, for instance, before the United States can mobilize forces to
intervene. Alternatively, Taiwan, fearing an impending PRC attack, may feel
the need to attack PRC forces while it still can. James Fearon argues that a commitment problem, arising as a consequence of strong ªrst-strike advantages,
lies at the root of preemptive wars. Under conditions in which attacking is
much more advantageous than defending (meaning the state initiating conºict
will gain the lion’s share of the good being bargained over), peaceful bargains
can become unenforceable because both states know the other has incentives
to attack, and both know they are better off acting ªrst.30 Fearon and others
note, however, that wars arising purely as a consequence of a preemptive logic
appear to be extremely rare. Rather, ªrst-strike advantages are more likely to
exacerbate other potential causes of conºict than they are to precipitate conºict
on their own.31
In the cross-strait case, the risk of inadvertent war would seem highest if one
or both sides were deliberately undertaking risky military exercises that could
potentially lead to mishaps. They might take such risks as a way of demonstrating resolve, which might otherwise be in doubt: risky exercises are a relatively credible signal precisely because they are costly (and they are costly
because they leave something to chance). For example, in Allen’s scenarios of
possible air force escalation in the Taiwan Strait, the PRC initially engages in
deliberately provocative (and potentially risky) behavior—such as ºying sorties near or around Taiwan—as a way to “send a ‘message’ to Taiwan.” The
fundamental underlying factor motivating a desire to signal resolve in this
way would most likely be uncertainty over PRC redlines.32 In this regard, an
inadvertent war in the Taiwan Strait—if one were to occur—would likely arise
as a consequence of both a commitment problem (stemming from possible
ªrst-strike advantages and the logic of preemption) and an information problem (stemming from the difªculties the PRC faces in credibly signaling the
true location of its redlines).
30. Fearon, “Rationalist Explanations for War.” For a discussion of the concept of the offensedefense balance, see Robert Jervis, “Cooperation under the Security Dilemma,” World Politics,
Vol. 30, No. 2 (January 1978), pp. 167–214.
31. Fearon, “Rationalist Explanations for War.” Dan Reiter suggests that preemptive wars “almost
never happen.” See Reiter, “Exploding the Powder Keg Myth: Preemptive Wars Almost Never
Happen,” International Security, Vol. 20, No. 2 (Fall 1995), pp. 5–34. On this point, see also Van
Evera, Causes of War. For a theoretical discussion of preemption in the cross-strait case, see Hengyu
Lee, “The Weak Preempting the Strong: The Case of the Taiwan Strait,” Issues and Studies, Vol. 47,
No. 2 (June 2011), pp. 205–230.
32. Allen, “Air Force Deterrence and Escalation Calculations for a Taiwan Strait Conºict.” On
costly signals and risk-taking, see, for instance, Schelling, Arms and Inºuence; and James D. Morrow, “The Strategic Setting of Choices: Signaling, Commitment, and Negotiation in International
Politics,” in David A. Lake and Robert Powell, eds., Strategic Choice and International Relations
(Princeton, N.J.: Princeton University Press, 1999), pp. 77–114.
Is the Taiwan Strait Still a Flash Point? 65
Key Trends in Cross-Strait Relations
In this section, I consider trends in three broad variables that have seen
tremendous change in recent years, and that are likely to affect future prospects for conºict and peace in cross-strait relations: China-Taiwan economic
integration and cooperation; the cross-Strait balance of military power; and
patterns in Taiwan public opinion relating to Taiwan’s identity and future
status. Changes in these variables have important implications for the three
conºict scenarios outlined earlier because they have potentially large effects
on the likely costs of a cross-strait conºict, the likely outcome of a conºict, and
the willingness of Taiwan’s leaders to challenge PRC redlines or acquiesce to
PRC demands.
deepening cross-strait economic integration and cooperation
Economic integration across the Taiwan Strait has proceeded rapidly since the
early 1990s. Bilateral trade in 2014 approached $200 billion (see ªgure 2),33 and
the stock value of Taiwan investment in the PRC is estimated to exceed
$150 billion.34 Economic ties between the PRC and Taiwan grew steadily despite the relatively high tensions that characterized cross-strait relations during the 1990s and early 2000s. Indeed, by 2003 the PRC had replaced the
United States as Taiwan’s largest trading partner, and the bulk of Taiwanese
outbound foreign direct investment was ºowing to mainland China.35 The interesting dichotomy of deepening economic integration against a backdrop of
continued intense political tension, in turn, motivated a considerable amount
of scholarship.36
Still, even as economic ties began to burgeon, the economic relationship was
33. This ªgure is according to Chinese customs statistics. According to Taiwan customs statistics,
cross-strait trade in 2014 was $130 billion. Regardless, trade has grown rapidly since the 1990s. See
Mainland Affairs Council, Cross-Strait Economic Statistics Monthly (Taipei: Mainland Affairs Council, various editions), http://www.mac.gov.tw/lp.asp?ctNode⫽5934&CtUnit⫽4152&BaseDSD⫽7
&mp⫽3.
34. See, for instance, U.S. Department of State, “2012 Investment Climate Statement—Taiwan”
(Washington, D.C.: U.S. Department of State, June 2012), http://www.state.gov/e/eb/rls/othr/
ics/2012/191245.htm. The report notes that some estimates suggest that total Taiwan investments
in China may be as high as $300 billion.
35. See, for instance, Robert G. Sutter, U.S.-Chinese Relations: Perilous Past, Pragmatic Present
(Lanham, Md.: Rowman and Littleªeld, 2010), p. 232.
36. See Yun-han Chu, “The Political Economy of Taiwan’s Mainland Policy,” Journal of Contemporary China, Vol. 6, No. 15 (1997), pp. 229–257; Scott L. Kastner, Political Conºict and Economic Interdependence across the Taiwan Strait and Beyond (Stanford, Calif.: Stanford University Press, 2009);
Tse-kang Leng, The Taiwan-China Connection: Democracy and Development across the Taiwan Straits
(Boulder, Colo.: Westview, 1996); and John Q. Tian, Government, Business, and the Politics of Interdependence and Conºict across the Taiwan Strait (New York: Palgrave Macmillan, 2006).
International Security 40:3 66
Figure 2. Cross-Strait Trade, 1994–2014
SOURCE: Mainland Affairs Council, Cross-Strait Economic Statistics Monthly, various edi-
tions (Taipei), http://www.mac.gov.tw/lp.asp?ctNode⫽5934&CtUnit⫽4152&BaseDSD⫽7
&mp⫽3.
NOTE: Figures are from PRC customs statistics.
far from normal. For instance, most direct travel and trade across the Taiwan
Strait was prohibited; rather, travel and trade ºowed through third locations such as Hong Kong. Relatively few mainlanders were permitted to
visit Taiwan, and most Chinese investment in Taiwan was likewise prohibited. The situation changed dramatically after the election of Ma Ying-jeou as
Taiwan’s president in 2008. Since then, the two sides have reached numerous
agreements that have, among other things, opened up direct travel, trade, and
communications across the Taiwan Strait; opened Taiwan to PRC tourists;
legalized some Chinese investment in Taiwan; and reduced tariffs on a wide
array of items. The centerpiece agreement, the 2010 Economic Cooperation
Framework Agreement (ECFA), both liberalizes trade and serves as a roadmap
for an eventual cross-strait free trade agreement.37
37. For an analysis of the economic implications of ECFA, see Daniel H. Rosen and Zhi Wang, The
Implications of China-Taiwan Economic Liberalization (Washington, D.C.: Peterson Institute for International Economics, 2011). On ECFA, see also Li, “The Nature of the Economic Cooperation
Framework Agreement and Its Implications for Peaceful Development of Cross-Strait Relations.”
Ma believes that economic integration not only beneªts Taiwan’s economy, but also contributes to
better political relations between the two sides by facilitating increased contact and understanding. See, for instance, “An Interview with Ma Ying-jeou,” New York Times, June 19, 2008, http://
www.nytimes.com/2008/06/19/world/asia/19taiwan-interview.html?pagewanted⫽all&_r⫽0.
Is the Taiwan Strait Still a Flash Point? 67
Growing economic integration and cooperation across the Taiwan Strait
have beneªted both sides greatly. Taiwan’s economy has become highly dependent on mainland China, as Taiwan companies have since the late 1980s increasingly located their manufacturing activities inside the PRC. Taiwan’s
exports to China, which mostly consist of materials used in the manufacturing
process by Taiwan companies with operations in China, account for roughly
40 percent of the island’s total exports.38 Meanwhile, some estimates suggest
that as much as 80 percent of Taiwan’s outbound foreign direct investment
ºows to the PRC.39 The recent opening of Taiwan to mainland tourists has
brought a surge of new visitors to Taiwan (see ªgure 3), which will have ripple
effects on Taiwan’s economy. Also, Taiwan has much to gain from continuing
the process of liberalization in cross-strait economic ties. Daniel Rosen and Zhi
Wang estimate, for instance, that a fully implemented cross-strait free trade
agreement would result in a Taiwan economy 4.4 percent larger than it would
otherwise be.40
There are some clear asymmetries in the cross-strait economic relationship.
Although China is Taiwan’s largest trading partner, for instance, Taiwan is
only the ªfth-largest partner of the PRC. Nevertheless, China too has beneªted
enormously from cross-strait economic ties. Excluding Hong Kong, Taiwan is
by far the largest foreign investor in China.41 And, as Douglas Fuller demonstrates, Taiwanese investment in the PRC “has made an overwhelmingly
positive contribution to China’s development, especially in the area of technology.”42 For instance, Taiwanese high-technology ªrms in China are transfer-
38. From 2010 to 2012, Taiwan exports to mainland China and Hong Kong combined to represent
40.4 percent of Taiwan’s total exports. See Taiwan Bureau of Foreign Trade, “Trade Statistics” (Taipei: Taiwan Bureau of Foreign Trade, September 2015), http://cus93.trade.gov.tw/ENGLISH/
FSCE/.
39. See, for instance, Jonathan Lin, “Is Taiwan Too Economically Dependent on China?” American
Chamber of Commerce in Taipei Taiwan Business Topics, Vol. 41, No. 11 (November 2011), http://
www.amcham.com.tw/topics-archive/topics-archive-2011/vol-41-no-11/3459-is-taiwan-tooeconomically-dependent-on-china. According to Taiwan’s ofªcial statistics, approximately
63 percent of outbound foreign direct investment ºows to the PRC. See Mainland Affairs Council,
Cross-Straits Economic Statistics Monthly, No. 245 (Taipei: Mainland Affairs Council), http://www
.mac.gov.tw/lp.asp?ctNode⫽5934&CtUnit⫽4152&BaseDSD⫽7&mp⫽3. It is widely recognized
that the ofªcial numbers signiªcantly understate actual Taiwan investment in the PRC, as many
investments have not been reported because they have violated Taiwan’s laws governing crossstrait exchange.
40. Rosen and Wang, The Implications of China-Taiwan Economic Liberalization, p. 69.
41. Ibid., p. 32. Ofªcial statistics often place Taiwan just behind Japan and the United States in this
regard, but again, these statistics dramatically understate actual Taiwan foreign direct investment
ºows into China.
42. Douglas B. Fuller, “The Cross-Strait Economic Relationship’s Impact on Development in Taiwan and China: Adversaries and Partners,” Asian Survey, Vol. 48, No. 2 (March/April 2008),
pp. 239–264.
International Security 40:3 68
Figure 3. PRC Tourist Visits to Taiwan
SOURCE: Mainland Affairs Council, Cross-Strait Economic Statistics Monthly, various edi-
tions (Taipei), http://www.mac.gov.tw/lp.asp?ctNode⫽5934&CtUnit⫽4152&BaseDSD⫽7
&mp⫽3.
ring signiªcant amounts of know-how to the PRC,43 and venture capitalists
from Taiwan are investing in numerous high-technology start-ups in China.44
Considerable uncertainty will inevitably surround any effort to forecast future trends in cross-strait economic ties: the relationship has been dynamic,
and it is likely to continue to change in dramatic ways. That said, there is some
reason to question whether the gains associated with cross-strait integration
and cooperation will grow as fast in the future as they have in the past. Most
obviously, the burst of economic cooperative agreements in recent years is unlikely to be sustainable. China and Taiwan have pursued cooperation under a
tacit agreement to address easy issues ªrst and save difªcult ones for later.45
Many of the easy issues, such as opening up direct links, have already been resolved. The Cross-Strait Trade in Services Agreement, a follow-up agreement
to the ECFA that would signiªcantly reduce barriers to the cross-strait service
43. For instance, Taiwan’s semiconductor industry is training large numbers of Chinese engineers
in design skills. See ibid., pp. 257–260, at p. 257.
44. Ibid., pp. 260–261.
45. See, for instance, Alan Romberg, “Cross-Strait Relations: First the Easy Steps, Then the
Difªcult Ones,” China Leadership Monitor, September 2008, http://www.hoover.org/research/
cross-strait-relations-ªrst-easy-steps-then-difªcult-ones.
Is the Taiwan Strait Still a Flash Point? 69
trade, was signed in 2013 but remains stalled in Taiwan’s legislature. Indeed,
KMT efforts to expedite passage of the agreement helped to trigger large-scale
protests in Taiwan (the Sunºower Movement), highlighting the tenuousness of
support for further liberalization of cross-strait economic relations in Taiwan.46
the shifting balance of military power across the taiwan strait
A second critical trend in the cross-strait environment is the shifting balance of
military power as a consequence of the PRC’s rapid growth and military modernization. Preparation for a conºict in the Taiwan Strait—in particular, deterring Taiwan independence and possible U.S. intervention in a Taiwan Strait
conºict—has been the primary driver of PRC military modernization efforts
dating to the 1990s.47 PRC military spending has grown rapidly over the past
two decades even as military spending has been relatively stagnant in Taiwan.
Given Taiwan’s prominence as a driver in PRC military modernization efforts,
rapid growth in PRC military spending has led to a sharp increase in PRC capabilities relevant to a possible conºict in the Taiwan Strait. Some analysts believe that the PRC’s modernization efforts have already “fundamentally
altered Taiwan’s security options,” and they emphasize that Taiwan’s defense
will increasingly need to rely on asymmetrical strategies.48
At the turn of the century, U.S. analysts tended to be fairly conªdent
in Taiwan’s capacity to rebuff a PRC attack. For instance, the 2000 U.S.
Department of Defense Annual Report on the Military Power of the People’s
Republic of China highlighted some of the challenges the PRC would face in a
Taiwan Strait conºict, noting that in the short term the People’s Liberation
Army (PLA) would “have only a limited capability to conduct integrated operations against Taiwan.”49 The RAND Corporation issued a study the same
46. A 2014 poll commissioned by Taiwan’s Mainland Affairs Council found that the public was
split concerning the Trade in Services Agreement, with 42.5 percent supporting the agreement and
40.1 percent opposing. See Taiwan Mainland Affairs Council, “Summarized Results of the Public
Opinion Survey on the ‘Public’s Views on Supervision of Cross-Strait Agreements’” (Taipei: Taiwan Mainland Affairs Council, 2014), http://www.mac.gov.tw/public/Attachment/47316281810
.pdf. Some other polls, however, found more substantial public opposition. See, for instance,
Chrystal Hsu, “Majority Opposes Trade Agreement: Poll,” Taipei Times, March 27, 2014, http://
www.taipeitimes.com/News/taiwan/archives/2014/03/27/2003586647.
47. See Michael A. Glosny, “Getting beyond Taiwan? Chinese Foreign Policy and PLA Modernization” (Washington, D.C.: Institute for National Strategic Studies, National Defense University, January 2011), http://ndupress.ndu.edu/Portals/68/Documents/stratforum/SF-261.pdf.
48. William S. Murray, “Asymmetric Options for Taiwan’s Deterrence and Defense,” in Chu and
Kastner, Globalization and Security Relations across the Taiwan Strait, pp. 61–79. See also William S.
Murray, “Revisiting Taiwan’s Defense Strategy,” Naval War College Review, Vol. 61, No. 3 (Summer
2008), pp. 12–38; and Shlapak et al., A Question of Balance.
49. Ofªce of the Secretary of Defense, Annual Report on the Military Power of the People’s Republic of
China (Washington, D.C.: U.S. Department of Defense, June 2000), http://archive.defense.gov/
news/Jun2000/china06222000.htm.
International Security 40:3 70
year that also highlighted the difªculties the PRC would face in mounting
an attack against Taiwan, concluding that “any near-term PRC attempt to invade Taiwan would likely be a very bloody affair with a signiªcant probability
of failure.”50
The past ªfteen years, however, have seen dramatic improvement in the
PRC’s relative military capabilities in the Taiwan Strait. For instance, the PRC
has steadily increased its number of short-range ballistic missiles (SRBMs) deployed within range of Taiwan, from approximately 350 in 2002 to more than
1,100 in 2012.51 Although Taiwan has invested in Patriot missiles to defend
against China’s SRBMs, the growing number of SRBMs, their increasing accuracy, and the high cost of ballistic missile defense lead some to be pessimistic
about the defensibility of Taiwan’s ªxed military targets. William Murray
writes, for instance, that the “competition between Chinese SRBMs and
Taiwan’s Patriot interceptors is . . . one Taiwan cannot win, and probably cannot afford to continue.”52 A recent study by the RAND Corporation likewise
concludes that China’s improving missile capabilities mean that it “soon may
be impossible for the United States and Taiwan to protect the island’s military
and civilian infrastructure from serious damage.”53
Recent years have also seen a sharp improvement in PRC relative air and naval capabilities in the Taiwan Strait. By 2014 China possessed approximately
600 modern combat aircraft (up from approximately 65 in 2000) and 90 modern submarines and surface ships (up from about 16 in 2000). Meanwhile, in
2014 Taiwan possessed approximately 330 modern combat aircraft and about
25 modern surface ships (and had not acquired any new modern combat aircraft or naval combatants since the mid-2000s).54 A qualitative edge for
Taiwan—including more advanced pilot training—has helped to blunt these
50. David A. Shlapak, David T. Orletsky, and Barry A. Wilson, Dire Strait? Military Aspects of the
China-Taiwan Confrontation and Options for U.S. Policy (Santa Monica, Calif.: RAND Corporation,
2000), p. xvi.
51. Murray, “Asymmetric Options for Taiwan’s Deterrence and Defense,” p. 64. For similar estimates, see Piin-Fen Kok and David J. Firestein, “Threading the Needle: Proposals for U.S. and Chinese Actions on Arms Sales to Taiwan” (New York: EastWest Institute, 2013), http://www.ewi
.info/idea/threading-needle-proposals-us-and-chinese-actions-arms-sales-taiwan.
52. Murray, “Asymmetric Options for Taiwan’s Deterrence and Defense,” p. 66.
53. Shlapak et al., A Question of Balance, p. 126.
54. The 2014 numbers for China and Taiwan come from U.S.-China Economic and Security Review Commission, 2014 Report to Congress of the U.S.-China Economic and Security Review Commission (Washington, D.C.: Government Printing Ofªce, 2014), p. 487. The 2000 data for China come
from pages 300–309 of the same report, where the term “modern” is also deªned as: for submarines, “a second-generation submarine that is capable of employing antiship cruise missiles” or
intercontinental ballistic missiles, and for surface combatants, a ship that “possesses a multimission capability, is armed with more than a short-range air defense capability, and has the ability to embark a helicopter.” For combat aircraft a more ºuid deªnition is used, based on existing
technologies.
Is the Taiwan Strait Still a Flash Point? 71
quantitative trends,55 though this edge may also be eroding with improving
PLA training.56 The 2009 RAND Corporation study noted earlier concluded
that improvements in China’s air and missile capabilities mean that Taiwan
can no longer be conªdent of maintaining air supremacy in the event of a
cross-strait conºict, even if the United States intervenes.57 Bernard Cole, writing in 2007, argued that the naval balance was also shifting sharply in China’s
favor: although China and Taiwan were “roughly equal in surface combatant
capability” at the time, the PLA Navy already possessed “much superior”
submarine and aviation capabilities.58 Although Cole suggested that U.S. intervention would shift the naval balance in Taiwan’s favor,59 China’s improving antiaccess capabilities increasingly complicate Washington’s calculus in
this regard.60
These trends imply that the PRC would be increasingly able to impose tremendous costs on Taiwan in the event of a cross-strait war. As Mark Stokes has
remarked, China’s growing missile arsenal implies that “every citizen of
Taiwan lives within seven minutes of destruction, and they know that.”61
Nevertheless, although many analysts believe that the cross-strait military bal-
55. Ian M. Easton, “Prepared Statement,” Hearing on China’s Relations with Taiwan and North Korea
before the U.S.-China Economic and Security Review Commission (Washington, D.C.: U.S.-China Economic and Security Review Commission, June 5, 2014), p. 4, http://www.uscc.gov/sites/default/
ªles/Easton_USCC_Taiwan_Hearing_Statement_2014.pdf.
56. On PLA Air Force training, see Kevin Lanzit, “Education and Training in the PLAAF,” in Richard P. Hollion, Roger Cliff, and Phillip C. Saunders, eds., The Chinese Air Force: Evolving Concepts,
Roles, and Capabilities (Washington, D.C.: National Defense University Press, 2012), pp. 235–254.
See also Allen, “Air Force Deterrence and Escalation Calculations for a Taiwan Strait Conºict,”
p. 175.
57. Shlapak et al., A Question of Balance, p. 131.
58. Cole, “The Military Instrument of Statecraft at Sea,” p. 195.
59. Ibid.
60. For instance, the 2013 Quadrennial Defense Review of the Republic of China argues that the
PRC’s antiaccess and area-denial capabilities are growing rapidly and in the future will increase
the risks of U.S. intervention and call into question the United States’ security commitments in the
region. See Republic of China Ministry of National Defense, Zhonghua Minguo 102 Nian Sinianqi
Guofang Zongjiantao [2013 Quadrennial Defense Review of the Republic of China] (Taipei: Republic
of China Ministry of National Defense, 2013), p. 12, http://qdr.mnd.gov.tw/content.html. On PRC
anti-access capabilities, see also Roger Cliff et al., Entering the Dragon’s Lair: Chinese Antiaccess
Strategies and Their Implications for the United States (Santa Monica, Calif.: RAND Corporation,
2007); Richard D. Fisher Jr., “Reºections on China’s Military Trajectory and the U.S. Pivot,” in Peter C.Y. Chow, ed., The U.S. Strategic Pivot to Asia and Cross-Strait Relations: Economic and Security
Dimensions (New York: Palgrave Macmillan, 2014), pp. 207–225; and M. Taylor Fravel, “China’s
Search for Military Power,” Washington Quarterly, Vol. 31, No. 3 (Summer 2008), pp. 125–141.
61. Mark Stokes, “Statement of Mark Stokes,” Hearing before the U.S.-China Economic and Security
Review Commission (Washington, D.C.: U.S.-China Economic and Security Review Commission, March 18, 2010), p. 71, http://origin.www.uscc.gov/sites/default/ªles/transcripts/3.18.10
HearingTranscript.pdf. Ross, “Navigating the Taiwan Strait,” argued in 2002 that the PRC’s growing capabilities already acted as an effective check on Taiwan independence, given the high costs
that they could impose.
International Security 40:3 72
ance is rapidly shifting in the PRC’s favor,62 less clear is whether the balance
already favors the PRC. Indeed, in recent testimony before the U.S.-China
Economic and Security Review Commission, Ian Easton emphasizes that PLA
analysts are themselves pessimistic about the PLA’s ability to achieve air and
sea superiority in the event of a cross-strait war.63 Furthermore, many—
perhaps most—analysts of cross-strait military trends are skeptical of the
PRC’s current ability to successfully invade and occupy Taiwan. These analysts point to complicating factors such as limitations in China’s amphibious
assault capabilities, unfavorable geography, and continued military modernization in Taiwan, as well as the high costs of such an operation and the risks
of escalation into a major war with the United States.64
Future trends in the cross-strait balance of power—in particular, the PRC’s
ability to defeat Taiwan in a military confrontation and impose its preferred
outcome—will hinge on several factors. First, Taiwan’s resolve is a critical variable. If Taiwan’s population and government are strongly committed to maintaining de facto independence and resisting political integration with the PRC,
then China’s growing capacity to impose high costs on Taiwan will not be
enough to compel surrender in the event of a cross-strait conºict. High
Taiwanese resolve also implies a willingness to continue to invest in the acquisition and development of weapons systems needed to deter a PRC attack. The
PRC’s amphibious assault capabilities represent a second critical variable: absent such capabilities, the PRC will lack the capacity to invade and occupy
Taiwan. China’s ability to conduct an amphibious assault on Taiwan, as noted
earlier, remains limited. If the PRC were to begin investing heavily in such capabilities in the future, however, it would potentially signal a more fundamental shift in the cross-strait balance of power, perhaps implying increased PRC
conªdence about China’s ability to achieve air and sea superiority in a crossstrait conºict.65
62. Shlapak et al., A Question of Balance; and U.S.-China Economic and Security Review Commission, 2014 Report to Congress of the U.S.-China Economic and Security Review Commission. Most contributors to Swaine, Yang, and Medeiros, Assessing the Threat, viewed the cross-strait balance as
shifting in the PRC’s favor.
63. Ian M. Easton, “Prepared Statement.”
64. Even Shlapak et al., A Question of Balance, which views the balance of power in the Taiwan
Strait as shifting rapidly, highlights the challenges that China would face in an actual invasion of
Taiwan. See chap. 5. On China’s limited amphibious capabilities and the risks associated with an
amphibious assault on Taiwan, see also Ofªce of the Secretary of Defense, Annual Report to Congress: Military and Security Developments Involving the People’s Republic of China (Washington, D.C.:
U.S. Department of Defense, 2014), http://www.defense.gov/Portals/1/Documents/pubs/2014
_DoD_China_Report .pdf. On the high costs of a PRC coercive attack on Taiwan, see Paul H.B.
Godwin and Alice L. Miller, China’s Forbearance Has Limits: Chinese Threat and Retaliation Signaling
and Its Implications for a Sino-American Military Confrontation (Washington, D.C.: National Defense
University Press, April 2013), pp. 19–20.
65. In testimony to the U.S.-China Economic and Security Review Commission, Ian Easton sug-
Is the Taiwan Strait Still a Flash Point? 73
Perhaps most importantly, future U.S. capabilities in the region and the evolution of the U.S. commitment to Taiwan’s security will have a large effect on
the cross-strait balance of power. At a minimum, the possibility of U.S. intervention means that PRC initiation of the use of force would be a tremendously
risky decision; the United States is also the only country willing to sell advanced weapons to Taiwan. As PRC capabilities continue to grow, Taiwan’s
ability to deter and ultimately repel an attack will hinge to a great degree on
future U.S. policy and capabilities. To what degree will the United States continue to be willing to sell advanced weapons to Taiwan, sales that typically
generate turbulence in the U.S.-China relationship? To what degree will the
United States continue to signal interest in Taiwan’s security—and hence
the possibility of intervention in the event of a cross-strait military conºict? To
what degree will the United States possess the capacity to intervene quickly
and effectively in the event of a cross-strait confrontation?
There has been some recent debate within the United States concerning the
desirability and sustainability of a continued U.S. commitment to Taiwan’s security, with some prominent scholars and analysts suggesting that the United
States scale back its commitment.66 So far there is little evidence of such a shift
in U.S. policy. For instance, the United States continues to sell advanced weapons to Taiwan. Announcements of major arms sales occur on an irregular basis, and such announcements have occurred somewhat less frequently over the
past decade. A ªve-year moving average, however, reveals that the average
annual sale (in constant dollars) since 2008 has been higher than at any other
point since the mid-1990s (see ªgure 4). Moreover, some analysts argue that
Taiwan has played a quiet but important role in the broader U.S. “pivot” (or
rebalance) toward Asia, exempliªed by growing bilateral military exchanges
and greater cooperation.67 Still, it is hard to predict future trends in this re-
gested that China’s unwillingness to expand the size of its amphibious ºeet in recent years may
reºect pessimism in Beijing about the PRC’s ability to attain air and sea superiority in a cross-strait
conºict. See Easton, “Prepared Statement.”
66. See especially Charles L. Glaser, “A U.S.-China Grand Bargain? The Hard Choice between Military Competition and Accommodation,” International Security, Vol. 39, No. 4 (Spring 2015), pp. 49–
90; Gilley, “Not So Dire Straits”; and Bill Owens, “America Must Start Treating China as a Friend,”
Financial Times, November 17, 2009, http://www.ft.com/cms/s/0/69241506-d3b2-11de-8caf00144feabdc0.html#axzz3d9SM9kwf. For a critical response to this line of argument, see Nancy
Berhkopf Tucker and Bonnie Glaser, “Should the United States Abandon Taiwan?” Washington
Quarterly, Vol. 34, No. 4 (Fall 2011), pp. 23–37. Some scholars suggest that the U.S. commitment to
Taiwan may be unsustainable as the PRC continues to become more powerful. See especially
John J. Mearsheimer, “Say Goodbye to Taiwan,” National Interest, March/April 2014, http://
nationalinterest.org/article/say-goodbye-taiwan-9931; and Robert Sutter, “Taiwan’s Future: Narrowing Straits” (Seattle, Wash.: National Bureau of Asian Research, May 2011).
67. See, for instance, Easton, “Prepared Statement”; and Dennis Van Vranken Hickey, “The U.S.
Pivot to the Asia-Paciªc: Implications for Taiwan,” in Shihoko Goto, ed., Taiwan and the U.S. Pivot
to Asia: New Realities in the Region? (Washington, D.C.: Wilson Center, 2013), pp. 13–24.
International Security 40:3 74
Figure 4. Major U.S. Arms Sales to Taiwan, 1992–2014
SOURCE: Shirley A. Kan, “Taiwan: Major U.S. Arms Sales since 1990,” Congressional Re-
search Service Report RL30957, August 29, 2014.
NOTE: Solid line represents a five-year moving average.
gard, and if the United States were in the future to signal a reduced interest in
Taiwan’s security, the ramiªcations for the cross-strait balance of power would
be substantial.
taiwan’s changing conceptions of identity and views on sovereignty
A third trend central to the prospects for stability in the Taiwan Strait centers
on how Taiwan’s citizens view themselves and Taiwan’s relationship with
China. On a broad level, opinion polls suggest a Taiwan electorate that on
the one hand is becoming more opposed to uniªcation with the PRC
while holding a greater sense of Taiwan-centric identity. But on the other hand,
polls also suggest a Taiwan electorate that remains deeply pragmatic on sovereignty issues.
Public opinion polling suggests that support for uniªcation with China has
been declining in Taiwan. To be clear, support for near-term political uniªcation with the PRC has long been miniscule among Taiwan voters. Polls conducted by National Chengchi University’s (NCCU) Election Study Center, for
instance, found that only 4 percent of Taiwan respondents favored uniªcation
“as soon as possible” in 1994.68 Nevertheless, in the mid-1990s, about 20 per68. The NCCU polls give respondents a range of options and ask which they prefer: uniªcation as
Is the Taiwan Strait Still a Flash Point? 75
Figure 5. Views of Taiwanese on Uniªcation/Independence, 1994–2014
SOURCE: Election Study Center, National Chengchi University, “Taiwan Independence vs.
Unification with the Mainland Trend Distribution in Taiwan,” http://esc.nccu.edu.tw/
course/news.php?Sn⫽167.
NOTES: Status quo includes those favoring status quo indefinitely and those favoring status
quo for now, with a decision later. Toward independence includes those supporting immediate independence, or the status quo for now with independence later. Toward unification
includes those supporting immediate unification or status quo for now with unification
later.
cent of Taiwan respondents in the NCCU polls favored eventual uniªcation—
including those saying they wanted uniªcation as soon as possible along with
those saying they wanted to preserve the status quo for now but who favored
uniªcation later. By 2014 this ªgure had dwindled to less than 10 percent (see
ªgure 5), with only 1 percent favoring uniªcation as soon as possible. Even
when asked about support for uniªcation under hypothetical favorable conditions (where the political, social, and economic conditions on mainland China
and Taiwan were similar), a declining percentage of Taiwanese voters express
support for the idea. For instance, the 2012 Survey of the Presidential and
Legislative Elections conducted by Taiwan’s Elections and Democratization
Study (TEDS) found that only 35 percent of those surveyed supported
uniªcation even under these favorable conditions, whereas a majority—
soon as possible; status quo for now, uniªcation later; status quo for now, decide later; status quo
indeªnitely; status quo for now, independence later; independence as soon as possible; and don’t
know. See Election Study Center, NCCU, “Taiwan Independence vs. Uniªcation with the Mainland Trend Distribution in Taiwan (June 1992–June 2015)” (Taipei: Election Study Center, NCCU,
July 9, 2014), http://esc.nccu.edu.tw/course/news.php?class⫽203.
International Security 40:3 76
51 percent—were opposed. In the 2004 Presidential Election Survey, 41 percent supported uniªcation under these favorable conditions, while only
42 percent were opposed.69
Meanwhile, a growing percentage of Taiwan’s citizens self-identify as
Taiwanese, rather than as Chinese or both Chinese and Taiwanese. According
to the NCCU surveys, in the early 1990s more respondents self-identiªed as
solely Chinese (about 25 percent) than as solely Taiwanese (about 20 percent), with about half of respondents self-identifying as both Taiwanese and
Chinese. By 2014 fewer than 4 percent of respondents self-identiªed as
solely Chinese, with a clear majority (more than 60 percent) self-identifying
as solely Taiwanese (with about 33 percent self-identifying as both Chinese
and Taiwanese). These trends have intensiªed since Ma Ying-jeou became
Taiwan’s president in 2008.
Yet even as Taiwanese support for cross-strait political uniªcation declines,
and as a tendency for Taiwan’s citizens to self-identify as solely Taiwanese
continues to grow, public opinion polls suggest that Taiwan’s population
remains pragmatic on sovereignty issues. For instance, although the NCCU
surveys point toward growing support for independence, a majority of respondents favor either maintaining the status quo and deciding on sovereignty issues later, or maintaining the status quo indeªnitely (see ªgure 5).
And of those supporting independence, only a small percentage (6 percent of
all respondents in 2014) favor independence as soon as possible; the rest favor
maintaining the status quo for now, with independence later. Other polls suggest that these preferences are pragmatic and reºect a fear that formal independence would trigger a cross-strait military conºict. For example, the 2012
TEDS survey found that 56 percent of respondents supported a declaration of
independence if peaceful relations with the PRC could be maintained; the
ªgure drops to 31 percent support, however, if a declaration of independence
were to trigger a PRC attack. This ªnding suggests that although a majority of
Taiwanese support a formally independent Taiwan in principle, a large majority does not support formalization of independence in practice because of how
the PRC would likely respond.70
69. These data are drawn from Taiwan’s Election and Democratization Study, The Survey of Presidential Election in 2004, http://www.tedsnet.org/main.php; and Taiwan’s Election and Democratization Study, The Survey of the Presidential and Legislative Elections, 2012, http://www.tedsnet.org/
main.php.
70. On the conditionality of Taiwanese preferences relating to uniªcation and independence, see
especially Emerson M.S. Niou, “Understanding Taiwan Independence and Its Policy Implications,” Asian Survey, Vol. 44, No. 4 (July/August 2004), pp. 555–567. Niou also ªnds in his survey
that a large majority of Taiwan voters supports independence if it could be achieved peacefully,
but that a large majority opposes independence if it were to cause the PRC to attack.
Is the Taiwan Strait Still a Flash Point? 77
Rethinking Cross-Strait Conºict Scenarios in Light of Recent Trends
How do the complex trends outlined above—deepening economic integration
and cooperation; a shifting balance of military power; and declining support in
Taiwan for cross-strait political integration, tempered by continued pragmatism on sovereignty issues—affect the prospects for conºict and peace
in the Taiwan Strait? In this section, I reconsider each of the three conºict scenarios discussed earlier in light of these trends.
the obsolescence of redlines?
The redline conºict scenario outlined above became much less likely after the
2008 election of Ma Ying-jeou as Taiwan’s president. Ma has shown little interest in pushing the envelope on sovereignty-related issues, and it is hard to
imagine him adopting policies that would cross a PRC redline, thereby triggering a military response. Less clear, however, is whether analysts should continue to worry about a redline conºict scenario in a post-Ma Taiwan. Below,
I argue that the major trends identiªed in the previous section combine in a
way that is reducing the likelihood of the redline conºict scenario. I suggest,
moreover, that this argument will apply even after the Democratic Progressive
Party returns to power in Taiwan.
Consider ªrst the impact of shifting Taiwanese views on sovereignty and
identity issues. Although support for uniªcation continues to decline in
Taiwan, this does not imply an increased willingness among the Taiwanese
public to challenge PRC redlines on sovereignty issues. Likewise, that a growing majority of Taiwan’s citizens now self-identify as solely Taiwanese does
not necessarily imply a growing dissatisfaction with the status quo among
Taiwan’s population. To the contrary, support for the status quo has trended
upward in recent years (ªgure 5), and some polls point to a considerable
amount of pragmatism among Taiwan’s public regarding sovereignty issues.
Thus, there is no clear reason to believe that public opinion in Taiwan is
evolving in a way that would reward politicians who aggressively challenge
PRC redlines.
The previous section showed that although cross-strait economic integration
is asymmetric—constituting a much larger percentage of Taiwan’s economy
than the PRC’s—both sides have nevertheless beneªted considerably from that
integration. Meanwhile, the PRC has become deeply embedded in global markets over the past three decades, and today stands as the world’s largest exporting country. Deepening economic integration and cooperation across the
Taiwan Strait, and China’s deepening integration into global markets more
generally, should increase the opportunity costs of cross-strait military conºict
International Security 40:3 78
for both sides (because many of these beneªts would disappear if conºict were
to erupt, especially if it were to escalate into a U.S.-China war).71 With this in
mind, consider again ªgure 1: all else equal, increased economic integration
should increase the size of both cc (China’s costs of war) and ct (Taiwan’s costs
of war). This, in turn, implies that the ceteris paribus impact of increased economic integration is to push China’s redline, R, to the right, while Taiwan’s expected war payoff, T, shifts left.
Now consider the effect of a shifting cross-strait military balance of power.
Improving Chinese military capabilities relative to Taiwan’s should affect both
the likely outcome of a cross-strait military conºict and the costs each side
would pay in such a conºict. More speciªcally, improving Chinese relative capabilities mean that any war that does occur will tend to end on terms more favorable to the PRC; in ªgure 1, this suggests that a shifting military balance is
having the ceteris paribus effect of pushing the war outcome (W) to the left.
Improving Chinese capabilities should also tend to reduce the PRC’s expected
costs of ªghting while increasing Taiwan’s costs of ªghting. In ªgure 1, this
means that China’s war cost (cc) decreases while Taiwan’s war cost (ct) increases. A shifting military balance, then, should have the ceteris paribus effect
of pushing China’s redline, R, to the left—because both a shrinking cc and a shift
left in W will independently push the redline left. A military balance shifting in
China’s favor should also ceteris paribus reduce the attractiveness—from
Taipei’s perspective—of Taiwan’s war payoff: both a shift left in the war outcome (W) and an increase in Taiwan’s war costs (ct) independently push
Taiwan’s war payoff (T) left.
In short, while deepening economic integration across the Taiwan Strait uniformly reduces the attractiveness of the war outcome for both sides, a shifting
military balance of power has differential effects on the two sides, reducing the
attractiveness of the war outcome for Taiwan but increasing the attractiveness
of the war outcome for Beijing. These effects are summarized in ªgure 6: here,
dark arrows show the independent effects of increasing economic ties across
the strait, while the lighter arrows show the independent effects of a shifting
71. The idea that economic integration increases the opportunity costs of conºict is central in
much of the literature linking economic integration to a reduced likelihood of military conºict.
See, for instance, Solomon William Polachek, “Conºict and Trade,” Journal of Conºict Resolution,
Vol. 24, No. 1 (March 1980), pp. 55–78; and Bruce Russett and John Oneal, Triangulating Peace: Democracy, Interdependence, and International Organizations (New York: W.W. Norton, 2001). For critical
analyses of the opportunity cost argument, see, for instance, Jack S. Levy, “Economic Interdependence, Opportunity Costs, and Peace,” in Edward D. Mansªeld and Brian M. Pollins, eds., Economic Interdependence and International Conºict: New Perspectives on an Enduring Debate (Ann Arbor:
University of Michigan Press, 2003), pp. 127–147; and Norrin M. Ripsman and Jean-Marc F.
Blanchard, “Commercial Liberalism under Fire: Evidence from 1914 and 1936,” Security Studies,
Vol. 6, No. 2 (Winter 1996/97), pp. 4–50.
Is the Taiwan Strait Still a Flash Point? 79
Figure 6. The Effects of Economic Integration and a Shifting Balance of Power
NOTE: Dark arrows represent the independent impact of increasing cross-strait economic ties;
light arrows represent the independent impact of China’s growing relative power.
military balance of power. Taken together, the trends suggest that Taiwan’s
war utility has been getting unambiguously worse: both a shifting military
balance and deepening economic ties have the effect of pushing Taiwan’s war
payoff, T, to the left (toward the uniªcation end of the continuum). But the
trends have contradictory, and hence ambiguous, effects on China’s war utility
(which deªnes China’s redline, R). While economic integration has the independent effect of pushing the redline R to the right (away from China’s ideal
point of uniªcation), the shifting military balance pushes R to the left (closer to
uniªcation). Thus, the net effect on China’s redline hinges on which is the
dominant trend, and drawing conclusions in this regard is no easy task.
Taken together, recent trends suggest that the attractiveness of revisionist policies—even for governments committed in principle to formal
independence—is declining for Taiwan, and it will continue to decline if crossstrait economic integration continues to deepen and the cross-strait military
balance of power continues to shift in China’s favor. First, it is unclear whether
Taiwan has more leeway than it had in the past to redeªne the status quo in
cross-strait relations, given that recent trends have had an ambiguous effect on
the PRC’s redline. Future Taiwan revisionism, were it to occur, would be no
less likely than it was in the past to trigger a PRC military response. Second,
the costs associated with crossing China’s redlines have been unambiguously
increasing for Taipei: deepening cross-strait economic links and a shifting balance of military power in the strait have combined to make a war with China
almost unimaginably costly for Taiwan. Given continued uncertainty over the
precise location of China’s redline, the structural environment in the Taiwan
Strait increasingly encourages moderation in Taiwan’s policies; the potential
International Security 40:3 80
beneªts of revisionism—even for a future DPP government—are increasingly
outweighed by the risks associated with it.72 Therefore, the scenario of a revisionist Taiwan government triggering military conºict by crossing PRC
redlines should be less likely in the future than it was in the past.
reasons for optimism and pessimism in beijing
A second conºict scenario that worried analysts prior to 2008 involved a pessimistic assessment in Beijing of trends in the Taiwan Strait. The danger highlighted in this scenario, in essence, is that Chinese leaders could embrace a
preventive logic. As with the redline conºict scenario, there are reasons to
think that the pessimistic trends scenario is less relevant today than perhaps
was the case under the Chen administration. Most obviously, the election and
re-election of Ma Ying-jeou—combined with Chen’s deep unpopularity during
his ªnal years in ofªce—should put to rest concerns that Taiwan’s political system is moving ineluctably toward more revisionist (pro-formal independence)
policies on sovereignty issues. Although future leaders in Taiwan might again
move in such a direction, there is no reason to think that revisionism is an
inevitable trend on the island.
Furthermore, some of the trends highlighted in the previous section—
especially deepening cross-strait economic ties and the shifting balance of
military power in the Taiwan Strait—are giving the PRC some reason to be optimistic about the future of cross-strait relations. Beijing clearly hopes that
deepening economic exchanges across the Taiwan Strait can help to facilitate
long-term changes in the attitudes that individuals in Taiwan have concerning
mainland China; such linkages, Beijing hopes, can help to “win hearts and
minds” in Taiwan.73 Some PRC analysts believe that the ECFA is helping to
further stabilize the security environment in the Taiwan Strait,74 and that average Taiwanese increasingly recognize their economic well-being is linked to a
stable cross-strait relationship.75 Other analysts believe that growing crossstrait economic ties—especially PRC efforts to increase economic engagement
72. Note that I am not suggesting that pro-independence sentiments, or hostility to uniªcation, are
likely to fade in Taiwan. Rather, I am suggesting that it is increasingly difªcult for these sorts of
sentiments to manifest themselves in concrete policy steps.
73. On recent PRC efforts to facilitate such effects through cross-strait economic exchanges, see,
for instance, “China Flaunts Buying Power,” Taipei Times, May 15, 2010, http://www.taipeitimes
.com/News/taiwan/archives/2010/05/15/2003473036; and Annie Huang, “China Uses Trade to
Inºuence Taiwan Election,” Associated Press, January 9, 2012.
74. See, for example, Li, “The Nature of the Economic Cooperation Framework Agreement and Its
Implications for Peaceful Development of Cross-Strait Relations.”
75. See, for instance, Chen Xing, “Lun Heping Fazhan Zhanlue Dui Taiwan Zhengdang Zhengzhi
de Yingxiang” [On the inºuence of the peaceful development strategy on Taiwan’s political parties
and politics], Beijing Lianhe Daxue Xuebao, Vol. 10, No. 3 (July 2012), pp. 87–92.
Is the Taiwan Strait Still a Flash Point? 81
with southern Taiwan—were a factor contributing to the KMT’s victory in the
2012 Taiwan presidential election.76 More generally, the “expanding gap” between China and Taiwan’s relative economic power “has made more people
on the mainland believe that time is on their side.”77
Trends in the cross-strait security relationship also appear to be moving in
Beijing’s favor as the PRC’s relative military capabilities continue to improve.
While—for reasons outlined in the previous section—the U.S. security commitment to Taiwan appears to remain quite strong, there is some reason for
optimism in Beijing, too. For instance, one key factor that may have helped to
generate pessimism in Beijing about long-term trends in the Taiwan Strait
in the past was a sense that the United States appeared to be moving toward
an unconditional commitment to Taiwan’s security—especially in the ªrst
years of the George W. Bush administration.78 Subsequent public U.S. criticisms of the Chen administration beginning in 2003, however, helped to allay
this concern.79
Nevertheless, Taiwan’s future politics—and how future PRC leaders are
likely to assess those politics—are hard to predict with any conªdence. After a
landslide victory, Tsai Ing-wen of the DPP will become Taiwan’s president
in May 2016, and for the ªrst time the DPP will also control a majority in
Taiwan’s legislature. How this will affect cross-strait relations remains unclear.
On the one hand, there are reasons to think that a future DPP government
76. See, for instance, Zhang Hua, “2012 Taiwan ‘Daxuan’ Xuanmin Toupiao Xingwei Xin Tedian ji
qi dui Liang’an Guanxi de Yingxiang Yanjiu” [A study of the new characteristics of the behavior of
Taiwan voters during the 2012 Taiwan “election” and their impact on cross-strait relations], Taiwan
Yanjiu Jikan, No. 5 (2012), pp. 24–32; Zhang Hua, “Cong Taiwan Diqu Liangxiang Xuanju Kan Taiwan Minyi” [Examining Taiwan public opinion from the standpoint of the Taiwan area’s two elections], Liang’an Guanxi, February 2012, pp. 25–26; and Xu Chuan, “Lun Nanbu Taiwan Zai
Liang’an Jiaowangzhong de Diwei yu Zuoyong” [On the standing and effects of southern Taiwan
with regard to cross-strait ties], Guizhou Shehuizhuyi Xueyuan Xuebao, No. 3 (2012), pp. 33–38.
77. Lin Gang, “Beijing’s Evolving Policy and Strategic Thinking on Taiwan,” in Hu, New Dynamics
in Cross–Taiwan Strait Relations, p. 74.
78. On the dangers of an unconditional U.S. commitment to Taiwan, see Christensen, “The Contemporary Security Dilemma.”
79. In a December 2003 meeting with Chinese Premier Wen Jiabao at the White House, U.S. President George W. Bush pointedly criticized Chen Shui-bian’s plans to hold a national referendum at
the same time as the 2004 Taiwan presidential election. See David Stout, “Bush Tells China Leader
He Opposes Taiwan’s Referendum,” New York Times, December 9, 2003, http://www.nytimes
.com/2003/12/09/international/asia/09CND-BUSH.html. The United States was openly critical
of several subsequent Chen initiatives, particularly Chen’s decision to hold a referendum seeking
support for Taiwan being admitted to the United Nations using the name “Taiwan.” Tsai Ing-wen,
the DPP candidate for president in 2012 and 2016, has sought to reassure the United States that,
if she were to be elected president, she would not pursue policies that would generate renewed
tensions in cross-strait relations. Her efforts in this regard suggest that the DPP recognizes at least
some conditionality in the U.S. security commitment to Taiwan. See, for example, Matthew Pennington, “Taiwan President Candidate Seeks Stable Relations with China,” Associated Press,
June 3, 2015.
International Security 40:3 82
would take a pragmatic approach to cross-strait relations, for reasons discussed in the previous subsection. Indeed, during the 2012 and 2016 presidential election campaigns, Tsai (who ran as the DPP candidate in both)
downplayed cross-strait relations as an issue, focusing instead on domestic issues (though opponents criticized her for a lack of clarity concerning how she
would manage to engage China).80 On the other hand, however, many PRC
analysts remain deeply skeptical that the DPP will implement fundamental
changes to its China policy. Such analysts note that the DPP’s voter “base” continues to include pro-independence fundamentalists, and that signiªcant moderation of the party’s China policy risks losing these voters in future elections.
Moreover, continued factional divides within the party will make it hard to
construct a new approach regarding China policy.81 As such, a return to DPP
governance in Taiwan—especially if it appears likely to be long-lasting—could
lead to a renewed sense of pessimism in Beijing.
Furthermore, though I noted above that some PRC analysts believe that PRC
efforts to foster increased economic ties with Taiwan have generally been successful, economic integration does not appear to be having a transformative effect on Taiwan society. While more microlevel data on this question are needed
(how do the views of individuals directly affected by increased exchanges
change as a consequence of those exchanges?), the aggregate trends presented
in the previous section are at best ambiguous. Although Taiwan public opinion
appears pragmatic on sovereignty issues, support for uniªcation has continued to drop, and the percentage of Taiwan’s citizens self-identifying solely as
Taiwanese has continued to increase, even as economic ties have deepened. If
these sorts of trends persist, Beijing could become less optimistic about the effectiveness of increased cross-strait exchange in winning Taiwan hearts and
minds. Indeed, a good deal of skepticism already pervades some PRC analysis
in this regard.82
80. On Tsai’s approach in 2012 and DPP postelection soul-searching, see Gunter Shubert, “No
Winds of Change: Taiwan’s 2012 National Elections and the Post-Election Fallout,” Journal of Current Chinese Affairs, Vol. 41, No. 3 (2012), pp. 143–161.
81. For analysis that is more pessimistic—or at least skeptical—concerning the prospects for fundamental change in the DPP’s approach to cross-strait relations, see, for instance, Wang Jianmin,
“Minjindang Nengfou Tiaochu Liang’an Lunshu de Kunjing” [Can the Democratic Progressive
Party extricate itself from the predicament of its cross-strait discussion?], Shijie Zhishi, No. 13
(2013), pp. 54–56; Peng Danyu, “Minjindang Liang’an Zhengce Tiaozheng de Quxiang yu Kunrao”
[The Democratic Progressive Party’s cross-strait policy adjustment prospects and pitfalls], Liang’an
Guanxi, No. 8 (2013), pp. 11–12; and Liu Qiang, “Minjindang Dalu Zhengce Tiaozheng Lunxi” [An
analysis of the Democratic Progressive Party’s mainland policy adjustment], Zhongyang
Shehuizhuyi Xueyuan Xuebao, August 2013, pp. 56–60.
82. Chinese Academy of Social Sciences researcher Wang Jianmin has noted, for instance, that today no major political forces in Taiwan advocate for uniªcation; to the contrary, a growing consensus has emerged that recognizes the ROC and Taiwan to be one and the same. These political
Is the Taiwan Strait Still a Flash Point? 83
the inadvertent war scenario
There are reasons to think that the risk of an inadvertent war in the Taiwan
Strait was limited even when tensions were especially high during the Lee and
Chen administrations, and that the risk of a conºict arising purely by accident
remains relatively small. As Stephen Van Evera notes, accidents—such as an
unintentional air collision—are most dangerous when ªrst-strike advantages
are large. If an accident is viewed in state A as the beginning of a larger attack
by state B, then ªrst-strike advantages mean state A will be under pressure
to “shoot ªrst and ask questions later.” State A, that is, will have incentives to
act preemptively.83
In the Taiwan Strait, it is unlikely that the PRC would interpret an accident
as the initiation of a Taiwan attack on the mainland: such an attack would be
suicidal and has not been a realistic concern for many decades. Thus, the inadvertent war scenario is more fundamentally rooted in the possibility that
Taiwan decisionmakers could misinterpret an accident as the beginning of a
mainland attack. Yet even if an accident were to be misinterpreted in this way
in Taiwan, the potential for rapid escalation should be contingent on the degree to which Taiwan has acquired signiªcant offensive capabilities as part of
an outward-oriented defense posture. Were Taiwan to adopt such a defense
strategy, and were Taiwan’s leaders to misinterpret some incident or accident as a mainland initiation of hostilities, Taiwan’s leadership could conceivably adopt a “use it or lose it” mentality with regard to their offensive
capabilities and thus act preemptively.84 On the other hand, if Taiwan’s reliance on offensive weaponry in its defense strategy is limited, it is less clear
why an accident would lead to rapid escalation (and potentially war).
trends in Taiwan lead Wang to be skeptical of the ability of cross-strait economic exchange to fundamentally alter political and social trends in Taiwan. See Cheng Hongli and Wang Jianmin, “2013
Liang’an Guanxi: Cheng Cuowei Fazhanzhishi” [Cross-Strait relations in 2013: A mismatched
state of development], Shijie Zhishi, No. 8 (2013), pp. 56–58. See also “‘New Taiwan Consensus’ a
Challenge to Beijing: Researcher,” Want China Times, August 23, 2013, http://www.wantchinatimes
.com/news-subclass-cnt.aspx?id⫽20130823000072&cid⫽1101. In other writings Wang does note,
however, that increasing economic exchange—and changes in the cross-strait economic relationship that allow southerners to reap beneªts from that exchange—are leading to increased support
for cross-strait economic cooperation in southern Taiwan. See Wang Jianmin, “Nan Taiwan Jingji
Shehui Jiegou dui Daonei Zhengzhi yu Liang’an Guanxi Yingxiang Chuxi” [A preliminary analysis of the impact of South Taiwan’s economic and societal structure on the island’s politics and on
cross-strait relations], Taiwan Yanjiu, No. 2 (2012), pp. 1–6.
83. Van Evera, Causes of War, p. 43.
84. If Taiwan were to possess signiªcant offensive capabilities, PRC decisionmakers would likewise be attuned to this possibility in the aftermath of an accident. That is, China would be unlikely
to view an accident as the initiation of a Taiwan attack, but Chinese ofªcials might worry about the
possibility that Taiwanese leaders would misinterpret an accident as a PRC attack and react using
a “use it or lose it” mentality. PRC leaders, then, could conceivably be faced with preemptive pressures of their own.
International Security 40:3 84
The shifting balance of military power in the Taiwan Strait has led to some
increased interest in Taiwan in developing a stronger capability to strike targets in China both to deter more effectively a Chinese attack and to enhance
the effectiveness of Taiwan’s defense capabilities in the event of deterrence
failure (i.e., to be able to strike PRC offensive capabilities, such as surface-tosurface missiles or airªelds).85 Taiwan, moreover, has recently deployed landbased cruise missiles that can strike targets in the PRC—and reportedly is
pursuing longer-range capabilities in this regard.86 Nevertheless, the scope of
these programs has, to date, been limited; in turn, it is unlikely that Taiwanese
leaders would view a preemptive strike as greatly improving Taiwan’s chances
in a cross-strait war—even if they were correct in their belief that war was inevitable. Preemption would also have massive risks. It would undoubtedly
leave many PRC assets intact and would virtually guarantee a signiªcant PRC
response. Preemption could also undercut the likelihood of U.S. intervention
in a conºict, and might thus actually worsen Taiwan’s war prospects.
Could the Taiwan Strait Again Become a Flash Point for Conºict?
The analysis in the previous section suggests that recent trends in the Taiwan
Strait are reducing the likelihood of the cross-strait conºict scenario that, arguably, most worried analysts prior to 2008—a revisionist Taiwan crossing PRC
redlines, thus triggering a military response. This does not imply, however,
that the underlying dynamics of the scenario are becoming irrelevant as regards the prospects for cross-strait conºict. To the contrary, Beijing and Taipei
85. On Taiwanese considerations about moving toward a more outward-oriented defense posture
during the early Chen administration, see Mulvenon, “Missile Defenses and the Taiwan Scenario.”
See also Dennis V. Hickey, “China’s Military Modernization and Taiwan’s Defense Reforms: Programs, Problems, and Prospects,” in Martin Edmonds and Michael M. Tsai, eds., Taiwan’s Defense
Reform (London: Routledge, 2006), especially pp. 52–53. The DPP under Chen’s leadership released a white paper in 1999 that focused on defense issues. One key point of emphasis in the paper was a need to develop the capacity to carry out attacks deep within the PRC. See York W. Chen
and Martin Edmonds, “An Overview of Taiwan’s Defense Reform,” in Edmonds and Tsai, Taiwan’s
Defense Reform, pp. 63–78. See also Lin Chieh-yu, “Chen Unveils DPP’s Platform on National Security,” Taipei Times, November 23, 1999, http://www.taipeitimes.com/News/local/archives/1999/
11/23/0000011885. For a discussion of whether the development of offensive missile capabilities is
likely to be an effective deterrent, see David A. Shlapak, “The Red Rockets’ Glare: Implications of
Improvements in PRC Air and Missile Strike Capabilities,” in Roger Cliff, Phillip C. Saunders, and
Scott Harold, eds., New Opportunities and Challenges for Taiwan’s Security (Santa Monica, Calif.:
RAND Corporation, 2011), p. 77. See also Bush, Uncharted Strait, pp. 178–179.
86. On recent deployments, see U.S.-China Economic and Security Review Commission, Report to
Congress (Washington, D.C.: Government Printing Ofªce, 2012), p. 259. On reports of a program to
develop longer-range missiles, see, for instance, Wendell Minnick, “Taiwan Working on New
‘Cloud-Peak’ Missile,” Defense News, January 18, 2013, http://www.defensenews.com/article/
20130118/DEFREG03/301180021/Taiwan-Working-New-8216-Cloud-Peak-8217-Missile?odyssey
⫽nav|head.
Is the Taiwan Strait Still a Flash Point? 85
continue to hold strongly divergent preferences over the issue of Taiwan’s
status, and it is likely that tacit bargaining on the issue will continue. Moreover, the optimistic assessment sketched out earlier hinges in part on the assumption that major trends in cross-strait relations—especially the shifting
balance of power and deepening economic ties—have ambiguous effects on
the location of China’s redline, R. It is unclear, however, whether the net effects
of economic integration and a shifting balance of power will remain ambiguous. Rather, as China’s military capabilities continue to improve (and as the
momentum in cross-strait economic cooperation appears to stall), it is possible
that the shifting balance of military power in the Taiwan Strait could become the dominant trend in cross-strait relations. Such a development could
generate renewed instability in the strait.
Consider again ªgure 6. Recall that a shifting balance of power has the ceteris paribus effect of pushing the likely war outcome, W, to the left and reducing the size of China’s war costs, cc. To the degree that these effects overwhelm
the effects of other trends on these parameters, China’s war utility should
improve—which in the model means that China’s redline, R, should begin to
shift left. The shift would be most pronounced if a changing cross-strait military balance (driven, for instance, by improving PRC amphibious capabilities
or a declining U.S. commitment to Taiwan) were accompanied by a decline in
PRC expected costs of conºict (which might occur, for example, if cross-strait
economic ties were to decay or if U.S.-China relations were to founder). Yet
just as conºict could occur if a revisionist Taiwan were to redeªne the status
quo to a point to the right of China’s redline, so too could conºict occur if a
changing cross-strait environment were to push China’s redline to the left of
the status quo. If the redline, R, were to shift to the left of the status quo, the
model suggests that the PRC would gain more from war than from a continuation of that status quo; in other words, Taiwan would now be risking military
conºict if it failed to accommodate this new reality by redeªning its sovereign
status more to the PRC’s liking.
If China’s redline were ever to drift to the left of the status quo, Taiwan
would have incentive to be accommodating, as a war would impose tremendous costs on the island. Indeed, several scholars have suggested in recent
years that China’s growing power makes Taiwan accommodation on sovereignty issues virtually inevitable. Robert Sutter writes, for instance, that
“China’s economic, military, and diplomatic leverage over Taiwan increasingly
constrains Taipei to follow a path leading to accommodation of and eventual
reuniªcation with China.”87 John Mearsheimer similarly warns that “if China
87. Sutter, “Taiwan’s Future,” p. 4.
International Security 40:3 86
continues its impressive rise, Taiwan appears destined to become part of
China.”88 And, recalling Finland’s strategic accommodation of Soviet interests
during the Cold War, Bruce Gilley likewise suggests that “Taiwan is moving in
the direction of eventual Finlandization.”89
Although a continued shift in the cross-strait military balance of power will
increase pressures on Taipei to move closer to the PRC, it is important to recognize the substantial barriers that stand in the way of Taiwanese accommodation on sovereignty-related issues. Generally speaking, cross-strait relations in
recent decades have been characterized by a deterrence dynamic, where the
PRC tries to deter Taiwan from redeªning its sovereign status in a way that
complicates PRC hopes for uniªcation. If China’s expected war payoff were to
improve dramatically, however, then the relationship could come to be characterized instead by a compellence dynamic.90 War in such a dynamic would occur not because the PRC fails to deter unilateral Taiwan changes to the status
quo that cross PRC redlines, but because the PRC fails to compel Taiwan to alter the status quo in a way that is more to Beijing’s liking (remember that once
China’s redline, R, shifts left of the status quo, China’s expected war payoff exceeds its utility for a continuation of the status quo). It is widely believed, however, that it is harder to compel than it is to deter,91 and future PRC efforts at
compellence in the Taiwan Strait would likely face signiªcant political, strategic,
and psychological obstacles. As such, the shifting cross-strait military balance of
power has the potential to be highly destabilizing in ways that may not be fully
appreciated in existing studies predicting future Taiwanese accommodation.
First, accommodation on sovereignty issues is likely to face considerable political resistance in Taiwan. Most people in Taiwan view their country as sovereign and independent, even if it is not internationally recognized as such,92
whereas support for uniªcation with mainland China—even in the long term,
under favorable conditions—is thin. President Ma has been able to pursue a
policy of détente with the PRC by embracing the ambiguous “1992 consensus,” which he interprets as an agreement that “one China” exists, but that
Beijing and Taipei have different interpretations as to what that “one China”
88. Mearsheimer, “Say Goodbye to Taiwan.”
89. Gilley, “Not So Dire Straits.”
90. Schelling conceptualizes the difference between deterrence and compellence as “the difference
between a threat intended to make an adversary do something [compellence] and a threat intended to keep him from starting something [deterrence].” See Schelling, Arms and Inºuence, p. 69.
91. See especially ibid.
92. See, for instance, Joy Su, “Survey Backs Sovereign and Independent Taiwan,” Taipei Times, November 9, 2004, http://www.taipeitimes.com/News/taiwan/archives/2004/11/09/2003210299.
Even KMT politicians who have generally opposed formal Taiwan independence view Taiwan’s
government, the ROC, as possessing sovereign statehood.
Is the Taiwan Strait Still a Flash Point? 87
means; for Ma, the Republic of China is the one China. Although this sort of
ambiguous and pragmatic formulation has been acceptable to the majority
of Taiwanese, further accommodation of China on the sovereignty issue would
be a risky political move in Taiwan. Consider, for instance, the decision by Ma
to broach the possibility of a cross-strait peace agreement in the midst of his reelection campaign in the fall of 2011.93 Even though Ma’s proposal was, as
Alan Romberg writes, “heavily caveated from the beginning,”94 it generated a
tremendous amount of domestic criticism, with DPP candidate Tsai Ing-wen
attacking the idea as putting Taiwan’s sovereignty at risk.95 Ma was forced to
add further caveats, such as promising that the ROC government would not
pursue an agreement until ªrst obtaining the consent of the Taiwan people in a
national referendum.96 The entire episode was generally viewed as a major political blunder by Ma, and it suggests that opposition would be ªerce if a future Taiwan government were to contemplate signiªcant accommodation on
sovereignty issues.
Insights from prospect theory reinforce this basic point. In the language of
prospect theory, being a functionally independent state forms Taiwan’s reference point. A shift from deterrence to compellence means that the PRC would
be in a position where it could demand—under credible military threats—that
Taiwan cede some of the status it currently has; that is, it would imply
that cross-strait bargaining—from Taiwan’s perspective—would shift from the
domain of gains to the domain of losses. Prospect theory predicts, however,
that individuals are more willing to take risks in the domain of losses: people
want to protect what they already have. As such, Taiwan’s people—and hence
its popularly elected politicians—might resist PRC demands to yield sovereign
status even if doing so is risky.97
Finally, commitment problems would likely complicate the search for a
peaceful accommodation even if a future Taiwan government comes to recognize that China’s redline has moved to the left of the status quo and is, in prin-
93. Wang Yu-chung, “Ma Talks Peace Deal with China,” Taipei Times, October 18, 2011, http://
www.taipeitimes.com/News/front/archives/2011/10/18/2003516029.
94. Alan D. Romberg, “Taiwan Elections Head to the Finish: Concerns, Cautions, and Challenges,” China Leadership Monitor, Winter 2012, http://www.hoover.org/research/taiwan-electionshead-ªnish-concerns-cautions-and-challenges.
95. Chris Wang, “Peace Proposal Puts Taiwan at Risk: Tsai,” Taipei Times, October 20, 2011, http://
www.taipeitimes.com/News/front/archives/2011/10/20/2003516192.
96. Romberg, “Taiwan Elections Head to the Finish,” p. 13.
97. On the use of prospect theory in international relations, see Jack S. Levy, “Prospect Theory, Rational Choice, and International Relations,” International Studies Quarterly, Vol. 41, No. 1 (March
1997), pp. 87–112. For an application to the cross-strait relationship, see Kai He and Huiyun Feng,
“Leadership, Regime Security, and China’s Policy toward Taiwan: Prospect Theory and Taiwan
Crises,” Paciªc Review, Vol. 22, No. 4 (September 2009), pp. 501–521.
International Security 40:3 88
ciple, willing to bargain away some of its sovereignty to preserve the peace.98
The problem here is that the issue being bargained over, Taiwan’s sovereign
status, affects Taiwan’s future bargaining power: to the degree that a Taiwan
government bargains away some of Taiwan’s sovereign status today,
Taiwan’s future bargaining power with Beijing will be further diminished.
Perhaps most importantly, even a loose uniªcation bargain would almost
certainly reduce even further Taiwan’s conªdence that the United States
would intervene in a future cross-strait conºict—because in the aftermath of
a uniªcation agreement U.S. involvement would represent intervention in a
civil, rather than an international, conºict. Thus, any bargain involving a reduction in Taiwan’s sovereignty should independently reduce Beijing’s expected costs of war, thus pushing China’s redline, R, even further to the left.
The credibility of Beijing’s commitment to such a bargain would therefore be
suspect, because once the commitment was implemented, Beijing would have
incentives to demand an even more favorable bargain (and Taiwan would not
be in a position where it could refuse). This situation thus represents a dynamic commitment problem, where the good being bargained over (in this
case, Taiwan’s sovereignty) has implications for future bargaining power of
the parties involved.99 Absent a mechanism that obliges Beijing to honor the
agreement, Taiwan could reject the bargain even if it recognizes that this could
lead to war.
In short, a shifting balance of power in the PRC’s favor has the potential to
create renewed instability in cross-strait relations, if it comes to overwhelm the
effects of other, more stabilizing, trends. How worried should analysts be
about this sort of a dynamic emerging in the Taiwan Strait?
On the one hand, there are several reasons to be optimistic that the Taiwan
Strait will not again become a ºash point for conºict even if the PRC’s military
capabilities continue to improve. First, as noted earlier, it is hard to predict the
long-term evolution of the balance of military power in the Taiwan Strait, and
as such it is not a given that a shift in the PRC’s favor will indeed become the
dominant trend in cross-strait relations. Some analysts have suggested, for instance, that Taiwan’s defenses are quite strong, particularly if the U.S. commitment to the island’s security remains robust.100 Second, even if a shifting
98. China would likely have difªculty credibly communicating to Taiwan that the PRC would
prefer war over a continuation of the status quo, given obvious incentives to bluff.
99. On dynamic commitment problems, see Fearon, “Rationalist Explanations for War,” pp. 408–
409. For an elaboration of commitment problems as an obstacle in cross-strait bargaining, see
Scott L. Kastner and Chad Rector, “National Uniªcation and Mistrust: Bargaining Power and the
Prospects for a PRC/Taiwan Agreement,” Security Studies, Vol. 17, No. 1 (January 2008), pp. 39–71.
100. See, for example, Easton, “Prepared Statement.”
Is the Taiwan Strait Still a Flash Point? 89
balance of power does become the dominant trend in cross-strait relations, it
will not necessarily become so dominant that Beijing concludes it could reap
greater beneªts from a cross-strait war than it receives from the status quo.
That is, even if the PRC is conªdent that it would “win” a cross-strait war (in
the context of ªgure 1, the war outcome might lie very close to the uniªcation
end of the continuum), it might still view the costs of war as prohibitively high
(enough to keep China’s redline to the right of the status quo). Also, there is
good reason to think that these costs are in fact substantial and will remain
so for the foreseeable future. As Paul Godwin and Alice Miller write, a PRCinitiated war in the Taiwan Strait would damage the PRC’s “effort to be perceived as a constructive, responsible member of the international community,”
which in turn would “have undesirable consequences for China’s global economic and commercial links.”101 Finally, Taylor Fravel has shown that the PRC
historically has been most likely to escalate territorial disputes when its bargaining power is declining, not when it is improving.102 This is a point emphasized in the pessimistic trends-analysis scenario highlighted earlier. Indeed, if
Beijing is conªdent that, over the long term, factors such as increasing crossstrait economic integration or increasing people-to-people contacts across the
Taiwan Strait will ultimately help to transform Taiwan’s identity and make
uniªcation more palatable to Taiwan’s citizens, then Chinese decisionmakers
could forgo the use of force even if they believed they could win a better outcome through war.
On the other hand, a combination of a shifting balance of military power
and renewed PRC pessimism about political and societal trends in Taiwan
would be especially worrisome. To see why, consider once more ªgure 1. Suppose again that the shifting balance of military power has pushed China’s redline, R, to the left of the status quo, implying that China expects a better war
payoff than it obtains from that status quo. From a pessimistic PRC’s vantage
point, this shift could open what is seen as a potentially short-lived window of
opportunity, where uniªcation could be achieved militarily at acceptable costs
today (China’s redline has shifted left of the status quo), but where the PRC expects future trends to raise those costs again (and push the redline back to the
right of the status quo). For instance, unfavorable (from Beijing’s perspective)
trends in Taiwan’s identity could dramatically increase the governance costs
associated with an attempted occupation of Taiwan: Beijing might expect that
101. Godwin and Miller, China’s Forbearance Has Limits, p. 20.
102. M. Taylor Fravel, “Power Shifts and Escalation: Explaining China’s Use of Force in Territorial
Disputes,” International Security, Vol. 32, No. 3 (Winter 2007/08), pp. 44–83; and Fravel, Strong Borders, Secure Nation. See also Christensen, “Windows and War.”
International Security 40:3 90
as more of Taiwan’s citizens self-identify as Taiwanese and not Chinese, the
PRC will face a more and more daunting task trying to govern the island.
Thus, a PRC that again becomes pessimistic about political and social trends in
Taiwan would potentially be less patient in a scenario where Chinese
decisionmakers come to believe that they have the capacity to achieve uniªcation at an acceptable cost.
Conclusion
Broadly speaking, this analysis of recent trends in cross-strait relations offers
some reason for optimism. To be clear, the relationship between China and
Taiwan remains on a fundamental level untransformed, and it continues to be
deªned by a sovereignty dispute that is unlikely to be resolved in the foreseeable future. Against this backdrop, future cross-strait relations will almost certainly be characterized by periodic tensions, particularly after the DPP returns
to power in Taiwan. Key trends in the relationship, however—including a
shifting balance of military power, deepening economic integration and cooperation, and changing views in Taiwan on identity and sovereignty issues—
combine to make less likely the cross-strait conºict scenarios that most worried
analysts in the past.
That the relationship has become more stable does not mean, however, that
military conºict in the Taiwan Strait has become unthinkable. The shifting balance of military power has the potential to create renewed instability if it begins to dominate other, more stabilizing, trends such as deepening cross-strait
economic integration. As Chinese military capabilities continue to improve, future PRC leaders may become less willing to continue to accept the status quo
in cross-strait relations. Analysts should not assume, however, that future
Taiwanese leaders will readily accommodate new power realities. To the contrary, a number of factors—relating to both Taiwan’s domestic politics and
structural commitment problems—will make it difªcult for Taiwan to yield
signiªcant ground on sovereignty issues in the years ahead. A shifting balance
of power, then, has the potential to give rise to renewed instability in the
Taiwan Strait if it dominates other trends.
The analysis presented in this article has signiªcant implications for U.S.
policy toward the Taiwan Strait. On the one hand, the potential dangers posed
by a shifting cross-strait balance of power suggest that proposals calling for a
reduced U.S. commitment to Taiwan’s security carry with them signiªcant
risks. For instance, a U.S. decision to terminate arms sales to the island would
almost certainly exacerbate changes in the cross-strait balance of power. Such
an effect could occur both directly, by undercutting Taiwan’s access to ad-
Is the Taiwan Strait Still a Flash Point? 91
vanced weaponry, and indirectly, by signaling a reduced U.S. interest in
Taiwan’s security (and hence reduced willingness to intervene in the event of
conºict in the Taiwan Strait). This is not to say that a concept such as Charles
Glaser’s grand bargain, where the United States would end its commitment to
Taiwan in exchange for the PRC peacefully settling other regional maritime
disputes and accepting a U.S. role in the region,103 would necessarily contribute to instability in the Taiwan Strait. Rather, I have argued that the destabilizing effect of a shifting military balance is conditional on the evolution
of other trends, particularly those that affect China’s expected costs of war and
those that affect the degree to which Chinese policymakers are optimistic or
pessimistic about where Taiwan is heading over the long term. Yet, because a
reduced U.S. commitment to Taiwan would affect the cross-strait military balance, such a policy shift is risky and should occur only in the context of a
broader understanding that signiªcantly increases China’s stake in a stable
status quo.
For similar reasons, U.S. policymakers should be skeptical of calls to dramatically increase the U.S. commitment to Taiwan’s security or to more visibly
incorporate Taiwan into rebalancing initiatives.104 Such moves would be
viewed as inºammatory in Beijing and could generate renewed pessimism
about long-term trends in the strait. Perhaps more fundamentally, a much
stronger U.S. commitment to Taiwan would stoke increased tensions in the
U.S.-China relationship. As alluded to earlier, a key component of China’s expected costs of a Taiwan Strait conºict—regardless of whether the United
States intervened—is the damage such a conºict would likely inºict on the
broader U.S.-China relationship. But if the U.S.-China relationship is in tatters
to begin with, then Beijing has less to lose in a cross-strait war. In essence, policies that increase U.S. support for Taiwan at a cost of a much worse U.S.-China
relationship risk being self-defeating, as the increased stability generated by
slowing the cross-strait power shift is canceled out by decreasing expected
costs of war for China.
Finally, although I have downplayed in my analysis the risk of inadvertent
war, it is something that cannot be discounted entirely. As such, the United
States should continue to advocate for U.S.-China and China-Taiwan
conªdence-building measures (e.g., hotlines, limits in risky military exercises
103. Glaser, “A U.S.-China Grand Bargain?”
104. For a nuanced analysis calling for a greater Taiwanese role in the pivot, see, for instance, Michael Mazza, “Taiwan’s Crucial Role in the U.S. Pivot to Asia” (Washington, D.C.: American Enterprise Institute, July 9, 2013), https://www.aei.org/publication/taiwans-crucial-role-in-the-uspivot-to-asia/.
International Security 40:3 92
undertaken in close proximity to the other party, military-to-military dialogues) that could help further mitigate this risk.105
The relationship across the Taiwan Strait has stabilized greatly in recent
years, but it remains untransformed on a more fundamental level. The underlying issue in dispute, Taiwan’s sovereign status, remains unresolved. China’s
military modernization—and the resulting shift in the cross-strait balance of
power—have the potential to introduce dangerous new dynamics that could
give rise to future conºict. Instability, though, is not inevitable. This article has
provided a number of reasons to be optimistic about the future of cross-strait
relations. Still, ofªcials in Washington, Beijing, and Taipei should not be lulled
by the recent détente; long-term stability will require continued careful management of the Taiwan issue in all three capitals.
105. For a detailed discussion and set of recommendations concerning conªdence-building measures in the Taiwan Strait, see Bonnie S. Glaser, “Building Trust across the Taiwan Strait: A Role for
Military Conªdence-Building Measures” (Washington, D.C.: Center for Strategic and International
Studies, January 2010), http://csis.org/ªles/publication/100107_Glaser_BuildingTrust_Web.pdf.