Political Polarization and Gridlock as the Result of an Institutional Imbroglio in Taiwan

Institutionalist theory argues that fairly permanent economic structures and policy-making arrangements create “institutions” which shape future policy-making and economic performance to a considerable extent (March and Olsen 1989; North 1990; Riker 1982). This perspective suggests that it might be valuable to look for the political institutions that evolved during Taiwan’s development which could have inhibited the country’s adaptability after the mid-1990s. From this perspective, Taiwan’s current institutional imbroglio results from a combination of a complex and somewhat indeterminate constitutional system, a long-time election system that contained some perverse incentives, and its long era of authoritarian rule. This paper, hence, seeks to provide an “institutionalist” explanation for the polarization and gridlock in Taiwan’s politics in the early 21st century.


Introduction
Though somewhat delayed, Taiwan's very successful democratic transition in the late 1980s and early 1990s created a seeming "political miracle" to go with the nation's earlier "economic miracle" (Chao and Myers 1998;Clark 2006;Copper 2009;Rigger 1999;Tien 1996;Wu 1995).The lifting of martial law in 1987, the forced retirement of the "senior legislators" who gave the ruling Nationalist or Kuomintang (KMT) Party a guaranteed majority in 1991, and the first direct election of President in 1996 brought full democracy to the country.Then, the election of Chen Shui-bian of the opposition Democratic Progressive Party (DPP) as President in 2000 showed that all was possible politically in the previously authoritarian Taiwan.Yet, democracy, as desirable as it was, did not solve all of Taiwan's political problems.For example, the development of a fairly even balance of political power between the KMT and the DPP led to growing gridlock and partisan polarization and growing emphasis upon the highly divisive national identity issue (Clark 2006;Makeham and Hsiau 2005).In addition, democratization may have even enhanced corruption by creating the need for huge campaign funds (Clark 2006;Rigger 1999) and by politicizing such areas of decision-making as financial policy and regulation (Tan 2001 and2008).
Institutionalist theory argues that fairly permanent economic structures and policy-making arrangements create "institutions" which shape future policymaking and economic performance to a considerable extent (March and Olsen 1989;North 1990;Riker 1982).This perspective suggests that it might be valuable to look for the political institutions that evolved during Taiwan's development which could have inhibited the country's adaptability after the mid-1990s.Two such institutions stand out.First, Taiwan's government had long been dominated by strong authoritarian leaders, such as Chiang Kai-shek and his son Chiang Ching-kuo.While the movement to electoral democracy in the late 1980s and early 1990s went smoothly (Rigger 1999;Tien 1996), the give-and-take of normal democratic policy-making had not been established; and, indeed, the institutional legacies from the past seemed to push the nation toward a fairly dysfunctional politics.Second, Taiwan's electoral system of a single vote in multi-member legislative districts also stimulated some of the less desirable characteristics of Taiwan's democratic politics.
More specifically, John Fuh-sheng Hsieh (2006 and2009) has developed an interesting model of Taiwan's institutional legacy based on the difference between presidential or parliamentary governmental systems and between elections systems with single-member districts (SMDs) or proportional representation (PR).From this perspective, Taiwan's current institutional imbroglio results from a combination of a complex and somewhat indeterminate constitutional system, a long-time election system that contained some perverse incentives, and its long era of authoritarian rule.Indeed, Hsieh (2006, 99) concludes that "actual constitutional practice in Taiwan [is] ... contrary to the constitutional arrangement on paper."This certainly points toward an "institutional imbroglio" in Taiwan politics.This paper, hence, seeks to provide an "institutionalist" explanation for the polarization and gridlock in Taiwan's politics in the early 21st century.The first section describes the growing polarization over national identity.The second part describes the institutional incoherence in the nation's governmental system; and the third argues that constitutional change over the last two decades has been fairly ineffective in overcoming these problems.Finally, we present a conceptual model of how Taiwan's "institutional imbroglio" has promoted polarization and gridlock in policy-making.

The Rise, Fall, and Rise of the National Identity Issue
The nature of authoritarian rule on Taiwan at the end of the Chinese Civil War created a bitter legacy of ethnic hostility and tensions that reverberates in the nation's politics even today.Since the evacuation of the Chiang Kai-shek regime to Taiwan in 1949 at the end of the Chinese Civil War, the island has suffered from a clear ethnic cleavage between the Mainlanders who came with Chiang (a little under 15% of the population) and the long-time residents of the Taiwan or Islanders who also were almost all ethnically Han Chinese.The Mainlanders dominated the government and imposed a harsh and repressive rule termed the "White Terror" until the country's democratic transition in the late 1980s and early 1990s.More broadly, especially after the initiation of the Chinese Cultural Renaissance Movement in 1966, the China-centric regime denigrated and discriminated against local culture and dialects by, for example, treating the Mandarin dialect as the official language of government and education, leading to ongoing resentments among the Islanders (Cheng 1994;Lee 2005;Lynch 2004;Makeham and Hsiau 2005;Mendel 1970;Phillips 2003;Wachman 1994).
Thus, the questions of ethnic justice and national identity, even if repressed by martial law, were clearly important to many Taiwanese.Taiwan's democratization, therefore, was widely expected to unleash Taiwanese nationalism on two interlinked but distinct issues: 1) rejection of the Mainlander-dominated political regime and 2) growing hostility toward and the absolute rejection of China's claim to sovereignty over Taiwan which was ironically at least tacitly supported by the Kuomintang's policy of "Mainland Recovery" (Gold 1986;Makeham and Hsiau 2005;Rigger 1999 and2001;Tu 1998;Wachman 1994).
As the 1990s opened, both the opposition DPP and the ruling Kuomintang KMT seemingly placed opposing bets on how the citizens of Taiwan would respond to this issue.The former bet that the end of authoritarian controls would permit Islander resentments to be expressed, winning over voters to the DPP as the champion of Taiwanese nationalism.For example, the DPP added support for Taiwan Independence to its Charter in 1991.Conversely, the KMT bet that the satisfaction of the general population with the prosperity created by Taiwan's "economic miracle" would make them supportive of the political status quo both domestically and in cross-Strait relations with China (Clark 2002).
Political forces soon began to push both parties away from these stark alternatives.In the KMT, Islander Lee Teng-hui, who as Vice-President succeeded Chiang Kai-shek's son Chiang Ching-kuo as President after his death in 1988, responded to this opportunity with what appeared to be inspired statesmanship on the national identity question.As Lee consolidated his power, he not so subtly pushed the KMT's position on cross-Strait relations in a new direction.Lee, in fact, managed to straddle the national identity issue quite astutely, implicitly portraying himself as a moderate between the pro-Independence DPP and the pro-Unification members of the KMT and (after 1993) the New Party who tended to be Mainlanders.While retaining a commitment to Unification with China in the indefinite future, he aggressively began to pursue the "pragmatic diplomacy" of trying to upgrade Taiwan's international status.For example, in 1993 he co-opted a popular issue from the DPP by launching a campaign to join the United Nations, which the KMT had strongly opposed up to then (Cabestan 1998;Chao 2002;Lasater 2000;Sutter and Johnson 1994).Furthermore, the victory of Lee's "Mainstream" faction clearly promoted the "Taiwanization" of the party--which made it hard to blame it for the repression of the "old" KMT.Consistent with the view that the old Kuomintang had been rejected, he formally apologized in 1995 for the massacre of thousands of Taiwanese by Nationalists troops in the Spring of 1947, the "February 28th or 2-2-8 Incident" (Chao and Myers 1998;Hood 1997).
For its part, the DPP began to moderate its position on Taiwan Independence in the early 1990s after the inclusion of a pro-Independence plank in the party charter cost it significantly at the polls in 1991.In particular, the Chinese military threats during the 1996 presidential elections and the woeful showing of the pro-Independence DPP candidate evidently convinced most of its leaders that Taiwan Independence was simply unfeasible.Consequently, the DPP began to downplay Independence without ever formally renouncing it.For example, some (but far from all) DPP leaders began to argue that Taiwan already was an independent country, so there was no need for a formal declaration of Independence (Rigger 2001;Wang 2000).Taiwan's political dynamics in the late 1990s, therefore, suggested that partisan differences over national identity were narrowing and losing their intensity (Clark 2006;Fell 2005).
However, a re-escalation soon erupted following the dramatic victory of the DPP's Chen Shui-bian in the 2000 presidential election, which he won with just under 40% of the vote in a three-way race (plus a couple of inconsequential minor candidates) with the KMT's Lien Chan and with James Soong who ran as an Independent after failing to get on the KMT ticket.Two distinct types of issues were involved in this polarization.The first was an ongoing struggle over the "localization" or Bentuhua of the country's politics and especially culture which was consistently pushed by the Chen administration.The second involved cross-Strait relations with the People's Republic of China and was more episodic; and here Chen Shui-bian's policies were far from consistent over time.Domestically, Chen displayed a strong commitment pursuing Bentuhua to create a "Taiwan-centric paradigm" for the nation (Hsiau 2005;Jacobs 2005).This, in turn, stimulated substantial opposition and pushback from the old guard KMT (Wang 2005).The administration used its executive power to promote what it called a "Taiwanese subjectivity" that certainly was aimed at its base constituency.
Wei-chin Lee (2005), for example, argues that Chen promoted a Cultural Reconstruction Movement that included such initiatives as changing the name of many agencies and organizations to stress "Taiwan," promoting Islander dialects in language policy, revising the official policy toward the mass media to reverse the previous KMT domination of outlets (including the encouragement of underground radio stations), and changing the focus from Chinese to Taiwanese history in education policy.Daniel Lynch (2004), for example, concluded that Chen and his "Green" bloc (named for the primary color of the DPP flag) were trying to create a new nation rooted in Taiwanese history and culture.
Relations with China were much more volatile, despite Chen Shui-bian's image as a zealot in promoting the declaration of de jure Taiwan Independence, which very probably would have resulted in military action by the PRC.Chen's pushing the envelope on the Independence issue commenced in the summer of 2002 when he proclaimed the theory that "one country on each side of the Taiwan Strait" existed, provoking significant unhappiness in both Beijing and Washington.After that, he periodically set off contretemps with Beijing and Washington until he left office in 2008, as he challenged China's "red lines" on Taiwan Independence by, for example, proposing or holding referenda on issues that might affect Taiwan's international status and by advocating fundamental change to the country's Constitution.
Yet, there were also signs of pragmatism in Chen's policies toward cross-Strait relations.He was fairly conciliatory toward an unresponsive PRC for his first two years in office and negotiated a "Ten Point Consensus" with the widely perceived pro-China James Soong in early 2005.More broadly, he followed a pattern of being aggressive toward China during electoral campaigns to appeal to the "deep Green" Taiwanese nationalists and then sounding much more conciliatory after the election was over.Indeed, he only became stridently pro-Independence consistently in 2006 when burgeoning scandals deprived him of support from almost everybody except the deep Greens (Clark 2006).
For their part, the KMT and its "Blue" coalition (named for one of the colors in the KMT flag) returned to a much more "China-centric" stance after Lee Tenghui left the party following its defeat in the 2000 presidential election.According to the model developed by Yu-shan Wu (2011), this represented a direct response to their electoral situation.During elections, Wu argues, the KMT acts like a catch-all party and appeals to the median voter with centrist policies.Between elections when the party is out of power (as it was from 2000 to 2008), in contrast, it focuses its appeals on keeping the support of the pro-China "deep Blues," while acting in a more pragmatic or "realist" manner when it controls the government.
By the middle of the first decade in the 21st century, therefore, a harsh and viciously divisive debate over cross-Strait relations and national identity had come to dominate Taiwan's politics.The Greens argued that they must "stand up for Taiwan" and accused the Blues of selling Taiwan out to China.In stark contrast, the Blues contended that the Greens were needlessly provocative and that a more accommodating policy could defuse the threat from China.Taken to the extreme (which they often were), these positions implied that one side was the savior and the other the destroyer of Taiwan and its statehood.Unfortunately, both critiques seem to have had some merit.President Chen's periodic appeals to his pro-Independence "base constituency" for primarily domestic purposes both infuriated China and at times strained relations with the United States, thereby threatening to undermine Taiwan's position in the Taipei-Beijing-Washington "triangle."Conversely, the Blue attempts to "do business" with Beijing undermined Chen's ability to deal with China; and there were even fairly credible rumors that Blue leaders had urged both the PRC and US to "get tough" with the Chen administration which in itself might have created a security threat to Taiwan (Clark 2006;Hickey 2006;Rigger 2005).

Institutional Incoherence in Taiwan
Originally, the government for the Republic of China on Taiwan was (and still essentially is) structured around the 1947 Constitution, which was based upon earlier constitutional and institutional developments in the interwar Republican period of China.This Constitution created the institutions for a liberal democracy based on the five branches of government proposed by Sun Yat-sen, the founder of the Kuomintang, and on constitutional guarantees of civil rights and liberties.Thus, at the national level, there were five basic governmental organizations: the Executive Yuan, the Legislative Yuan, the Judicial Yuan, the Control Yuan, and the Examination Yuan.An indirectly elected President stood above these five branches and served as the top political official in the nation.Freedom of speech and other political rights were guaranteed; and universal suffrage and the secret ballot were mandated (Ch'ien 1950;Copper 1979;Winckler 1984).
The keystone of the ROC's government was the President, who initially was indirectly elected for six-year terms by the National Assembly.The National Assembly itself was conceived as a major representative body that, in addition to selecting the President and Vice-President, was charged with adopting and amending the Constitution.National Assembly Members were elected by constituencies of 500,000 in China for six-year terms; and some representation was also given to occupational groups, racial minorities, and overseas Chinese (Barnett 1963;Ch'ien 1950).
The President possessed important constitutional powers, but there were also significant limitations on them as well.He appointed the Premier who headed the Executive Yuan and also had appointment powers for the Judicial and Examination Yuans.Moreover, the President became the focal point for several important decision-making bodies, such as the National Security Council that was founded by Chiang Kai-shek in 1967.The NSC has been generally composed of some of the top officials in the regime and seemingly has served as a "super cabinet" at many times.Constitutionally, however, the President did not really appear to be the chief executive.It was the Premier who selected and presided over the cabinet; and, at least on paper, the Premier and the cabinet were responsible to the Legislative Yuan (Ch'ien 1950;Gurtov 1968;Hsieh 2006).
It is somewhat ambiguous, therefore, whether the ROC Constitution created a presidential or cabinet system because the exact division of labor between the President and Premier has been somewhat unclear and has depended upon their personal power positions.In reality, except for the brief period after Chiang Kaishek's death when his Vice President finished out his term, the President has always been preeminent.Still, Presidents and Premiers have not always been fully compatible.For example, President Lee Teng-hui and his Mainstream faction of the KMT clearly had strong differences with Premier Hau Pei-tsun, a leader of the Anti-Mainstream KMT in the early 1990s.This led to a joke about why Taiwan was the most democratic nation in the world: "The United States has a President and people drive on the right; the United Kingdom has a Prime Minister and people drive on the left; while Taiwan has both a President and a Premier and people drive on both sides of the street." The Legislative Yuan or Taiwan's parliament was a directly elected body that was constituted much like the National Assembly.Even during the long authoritarian era, it passed budgets and legislation and exercised oversight over the executive (e.g., the Executive and Legislative Yuans had vetoing and overriding powers fairly similar to those exercised by the President and Congress in the United States).In reality, the Legislative Yuan was fairly weak; and it is probably fair to describe it as a "rubber stamp" on major policies before the 1990s.Still, the Legislative Yuan held the very important formal power of having to approve presidential appointments of Premiers (Ch'ien 1950;Hsieh 2006).More informally, legislators did exercise considerable initiative in such important areas as amending legislation, constituent service, local development projects, and overseeing the executive in public interpellation sessions (Chou, Clark and Clark 1990).
The other three major branches of government or Yuan have been much less important but still significant in Taiwanese politics.The Judicial Yuan, whose members were appointed by the President, interpreted the Constitution and served as the Supreme Court for the ROC.The Control Yuan, whose members were initially elected indirectly by sub-national governments (provinces in China), exercised oversight over other parts of the government.It held general auditing powers, had to consent to the appointments to the Judicial and Examination Yuans, and could censure and (with the approval of the National Assembly) impeach government officials.Finally, the Examination Yuan, which was also appointed by the President, oversaw the system of civil service examinations and served as a personnel agency for the government (Ch'ien 1950).
As originally established in China, the governmental structure was a federal one with three levels--national, provincial, and county.The retreat from the Mainland left essentially one Province (Taiwan) that encompassed almost all the territory governed by the ROC, with some "offshore" islands near China under military administration.The provincial administration was directed by a Governor who was appointed by the President.There also was a directly elected Provincial Assembly, whose relationship to the provincial executive paralleled that between the Executive and Legislative Yuans.Legislative politics in the Provincial Assembly were much livelier than in the Legislative Yuan both because the members were directly elected and because the issues that the Assembly could deal with were not all that important (Lerman 1978).In addition, the two largest cities in Taiwan, Taipei in 1967 and Kaohsiung in 1979, were made "special municipalities" under the Executive Yuan with structures fairly similar to the provincial government (including their mayors becoming appointive rather than elective).Figure 1 outlines this basic governmental structure and the relationships among its various components.Solid arrows indicate the formal power to select and dashed arrows show check-and-balance relations.This diagram suggests three central characteristics of Taiwan's politics during the authoritarian era of the 1950s through the 1980s.First, the electorate possessed substantial power in theory, but it was quite circumscribed in practice by two important factors: 1) the indirect election or selection of a large number of key leaders and bodies and 2) the fact that large majorities of the Legislative Yuan and National Assembly continued to be "senior legislators" elected in China in the late 1940s.Second, the large number of dotted lines has ambiguous implications.On the one hand, it could have promoted checks and balances among the various governmental institutions.On the other, there was certainly the potential for institutional incoherence, especially in the key set of relationships among the President, Premier and cabinet, and Legislative Yuan.Third, the foundation certainly existed for a strong presidency, despite the ambiguity over whether the President could even be considered the chief executive.
The formal constitutional structure for any country, of course, is subject to significant modification by informal political practices.This occurred to an extreme extent in authoritarian Taiwan.Not only were the Constitution's democratic intent and institutions subverted by the authoritarian controls exercised by the state, but the outline of the nation's political bodies in Figure 1 completely ignores the central role of the ruling Kuomintang Party.The party's institutions were greatly influenced by Soviet advisors to Sun Yat-sen in the 1920s; and once it came to power it created something of a Leninist party-state.Consequently, major decisions about policy and personnel were evidently approved, if not made, by the top party organizations rather than the official government (e.g., the KMT's Central Standing Committee had to approve the Premier's cabinet choices); and the Executive Yuan was viewed as much more of a policy implementer than initiator.The party also had fairly extensive ties with society through such organizations as the China Youth Corps and Farmers' Associations.The strong presidential leadership that has marked Taiwan's politics was based to a goodly extent on control of the ruling party as both Chiang Kai-shek and Chiang Chingkuo were strong leaders of the KMT through their presidencies (Bedeski 1981;Clough 1978;Copper 1979;Tai 1970).This continued in the democratic era, as Lee Teng-hui was Chairman of the Kuomintang throughout his presidency (1988)(1989)(1990)(1991)(1992)(1993)(1994)(1995)(1996)(1997)(1998)(1999)(2000) and Chen Shui-bian and Ma Ying-jeou have been Chairs of their respective parties for part of their terms.
Despite this structure of a party-state and a substantial amount of repression directed against those who challenged the regime, the KMT on Taiwan departed from the classic Leninist strategy of control in one vital respect.Rather than destroying all pre-existing political and social groups, the regime tried to co-opt and manipulate them whenever possible.This resulted in the Mainlander "national" elite playing "local" Islander factions off against each other and retaining power by acting as the arbitrator among them.This also made elections for local governments and Farmers' Associations "real" and often fiercely competitive, which had somewhat contradictory implications for Taiwan's future political development.On the one hand, a significant basis or starting point was created for democratic development and expansion; on the other, these local bodies were strongly focused on political patronage which was often tied to corruption (Bosco 1994;Clark 1989;Rigger 1999;Tien 1989).This turns attention to the somewhat rare type of election system which Taiwan imported from its former colonial master Japan.As noted in the Introduction, the two major types of election systems are single-member districts (SMDs) in which the person who gets the most votes wins and proportional representation (PR) in which a party wins the number of seats in a multi-member district that is proportionate to its share of the vote.Taiwan's system of what is called the single nontransferable vote or SNTV combined elements of both systems (the SNTV system was abolished at the time of the 2008 Legislative Yuan elections).The candidates for legislative seats ran in multi-member districts as in PR.However, each voter only could cast one ballot for a specific candidate (not a party) which could not be transferred to a second or third alternative if the candidate did not win.The candidates were then ranked according to the votes they received; and the number elected was determined by the size of the district.For example, if a district had eight seats, the eight candidates with the highest number of votes were the winners.Consequently, in large districts fairly small minorities could elect a representative (Hsieh 2009).
This system appears to have ambiguous implications for the party system.SMDs are usually considered to promote competitive two-party systems because minor party candidates are difficult to elect.This promotes the ability of the citizenry to hold a government accountable but makes the representation of some specific constituencies hard because the catch-all major parties must retain broad appeal.Conversely, a PR system promotes a multi-party system that is good for representation but can undermine accountability.The SNTV system promoted representation by individual legislators but undermined representation by a party because candidates of the same party had to compete against each other as well as against the representatives of other parties, which undercut the cohesion and responsibility of the parties.As John Hsieh (2009, 12) explains nicely: Since the vote shares of these two parties [the DPP and KMT] are, under normal circumstances, relatively fixed, it can be expected that candidates from the same party will compete against each other for the same pool of voters.In fact, this kind of intraparty competition is more often than not fiercer than competition between the two parties.As voters make their choices, they often first determine which party to vote for, and then pick one out of several candidates from that party.Since the platforms of these candidates are likely to be similar, voters need to rely upon other cues to make their choices, including personal connections, pork-barrel projects, or even vote buying.Elections may become very personalized.In addition, since each party, in general, wants all its candidates to win, and often needs to show impartiality among its own candidates, these candidates may have to turn to other sources of support to compete against their co-partisans.Factions, big businesses, or even gangsters may be dragged into the process.Corruption may thus sneak in.Moreover, because a candidate may need only a small portion of the vote in the district to get elected, he or she may choose to take extreme positions to attract the support of certain groups of voters.In this way, radicalization may become a constant feature of political life.
There seems to be a parallel between Taiwan's constitutional and electoral systems, therefore.The constitutional system combined elements of both parliamentary and presidential governments in a somewhat incoherent system that was held together by authoritarian one-party rule.The election system was neither SMD nor PR and appeared to undercut the incentives that one or the other might have provided for establishing a particular type of party system.Democratization, as might well have been expected, exacerbated these problems.The incoherence and ambiguities of the constitutional system became increasingly apparent as competing political forces were given free rein to pursue conflicting interests and goals; and the growing importance of elections accentuated the dysfunctions of the SNTV system.

Ineffective Constitutional Reforms
In response to the problems sketched in the previous section, Taiwan has amended its Constitution quite significantly over the last two decades during the democratic era.Yet, the efficacy of these reforms for surmounting the institutional imbroglio facing the nation remains rather problematic.Shelley Rigger (2007) developed a model of four rounds of constitutional change in democratic Taiwan over the last two decades, extending the work of Lin Jih-wen (2002) for the 1990s.Lin and Rigger differentiated two major initial motivations for constitutional reform.One was the ideological perspective on the existing Constitution; and the other concerned pragmatic power considerations.Ideologically, the DPP challenged the legitimacy of the old Constitution, which the traditional KMT generally supported; and during the 1990s, President Lee Teng-hui and his Mainstream KMT took an intermediate position with their goal of solidifying Taiwan's sovereignty within the existing constitutional structure.After the turn of the century, according to Rigger, a new justification for constitutional reform emerged because of growing dissatisfaction with the dysfunctional operations of the country's political institutions.
Round #1 (1991-97) Goals: Facilitate Democratization and Re-Balance Powers among Political Institutions 1. Direct election of Legislative Yuan, National Assembly, Provincial Governor, Special Municipality Mayors, and President; 2.More power to President (e.g., Legislative Yuan lost power to confirm Premier); and 3. Phasing out of Provincial Government.

Round #2 (1999-2000)
Goal: Define the Role of the National Assembly 1.Initial attempt to extend terms draws major public backlash; but 2. Ultimately, powers downgraded substantially.

Round #3 (2001-05)
Goal: Make Government More Effective 1. Legislative Yuan cut in size; 2. New election system strengthening major parties; 3. National Assembly abolished; and 4. Changing Constitution made much harder without very broad-based support.

Round #4 (2005-07)
Clashing Perspectives: Promote Governmental Efficiency vs. Create New Constitution (Country) 1. Major DPP push ended with no proposal, while KMT not interested in major constitutional change.1991, 1992, 1994, and 1997. Rigger (2007) ) sees these as facilitating democratization and balancing the powers among Taiwan's basic political institutions.The first three clearly were essential to establishing a democracy since they allowed the direct election of the Legislative Yuan, National Assembly, Provincial Governor, Mayors of Taipei and Kaohsiung, and finally the President.Ideologically, these reforms conformed to the DPP critique of the old Constitution, helped increase the power of the DPP and of Lee's Mainstream faction against the KMT old guard, and were overwhelmingly supported by the general public.The 1997 amendments, in contrast, focused on the relationships among governmental bodies.The power of the President was strengthened by removing the Legislative Yuan's power to confirm the Premier; and the Provincial Government was drastically downsized.According to Rigger, this represented a trade-off between President Lee's ambition to increase presidential power and the DPP's ideological goal of removing links to China from the Constitution (see Yeh 2002 for a somewhat similar interpretation).This first round of constitutional reform was based on an overlap of the DPP's ideological interests and the Mainstream KMT's power considerations.It left the DPP still advocating the need for major constitutional change but satisfied the Mainstream KMT, thereby making it hard, if not impossible, to create sufficient support for more reform (Rigger 2007).A second round emerged, though, in a battle over the status of the National Assembly which had lost its major function of electing the President.In 1999, it passed an amendment that extended the term of its members which led to a great popular uproar and which was then declared unconstitutional by the Council of Grand Justices.The widespread opposition to the National Assembly's self aggrandizement led to a new amendment in 2000 that substantially downgraded its status as the DPP's long-standing ideological opposition to the body dovetailed well both with the KMT's desire to avoid an election that would benefit a new party founded by KMT defector James Soong and with the general popular revulsion against the National Assembly (Rigger 2007;Yeh 2002).
President Chen Shui-bian's presidential victory in 2000 was soon followed by a vicious confrontation with the KMT-controlled Legislative Yuan (Clark 2006).The public seemingly agreed with the DPP that the legislature had become overly obstructionist and bore the brunt of the blame for the polarized gridlock that was afflicting Taiwan.In addition, there was a widespread perception that the SNTV electoral system promoted political polarization and corruption in public affairs.Thus, the new push for constitutional reform during the first Chen administration focused on making what had become a chaotic government more effective and efficient.While the Kuomintang was far from enthusiastic about this third round, it did not feel that it could defy overwhelming popular support for reform.Consequently, a new set of amendments was passed in 2005 that cut the size of the Legislative Yuan in half, replaced the SNTV system with a combination of mostly single-member districts supplemented by some proportional representation of major parties (which ironically worked to the KMT's favor in the next elections in 2008), abolished the National Assembly, and made further constitutional change very difficult without broad-based support (Hsieh 2009;Rigger 2007).
The actual constitutional change produced by the third round focused on making Taiwan's government work better.President Chen and many others in the DPP, though, had ambitions for more drastic reform in line with the DPP's ideological goal of creating a new Constitution.The DPP, thus, renewed a campaign for revising the Constitution that combined Chen's periodic proposals for creating a "Second Republic" with initiatives to make the structure of the government more effective.However, this fourth round was essentially abandoned in mid-2007 when the DPP switched its focus to supporting a referendum at the time of the 2008 elections on whether the nation should join the United Nations under the name Taiwan, while the Kuomintang showed little interest in major constitutional reform (Rigger 2007).
While not a major constitutional issue necessarily, the potential use of referenda in Taiwan politics also has some relevance for the question of how to make the political system work better.The Democratic Progressive Party had initially advocated a referendum for declaring Taiwan Independence in the early 1990s.Thus, the idea of adopting legislation to allow referenda and of holding referenda strongly appealed to the DPP "base."Referenda, of course, can be held on many issues that have nothing to do with Taiwan Independence and the island's status and sovereignty (e.g., a township referendum that was held on whether it should get a freeway exit).Indeed, when Chen Shui-bian began to push for a referendum law in 2003 with the goal of holding a referendum simultaneously with the 2004 presidential election, he took more than a little care to choose issues that did not involve a direct change in Taiwan's status or declaration of Independence (e.g., whether Taiwan should be granted membership in the World Health Organization or WHO which appealed to the presumably large majority of citizens who were frustrated and angered over the PRC's ability to deny Taiwan status and "face" in international affairs).As Shelley Rigger (2004, 186) argues, this certainly appears to have been politically motivated in terms of the upcoming election: The theory is that referendums, especially symbolic ones like that on the WHO, will help the DPP politically by mobilizing the party base and perhaps even exciting patriotic emotions that will draw votes beyond the DPP's traditional supporters.Holding the referendum together with the presidential election would allow enthusiasm for the referendum to spill over into the presidential race.
Chen Shui-bian's advocacy of establishing referenda turned out to have two quite distinct and separate appeals.It certainly appealed to supporters of Taiwan Independence among the Pan-Green forces.It also had wide support among the general public who rejected Independence as too radical and provocative, because referenda were seen as a way of surmounting the ongoing gridlock in Taiwan's politics and as deepening Taiwan's democracy by allowing the people to directly determine policy (Mattlin 2004;Rigger 2004).Consequently, the politics of the referendum issue during 2003-2004 turned out to be quite convoluted.The legislative enactment involved a three-sided struggle among Chen, more radical advocates of using the referendum to achieve Taiwan Independence, and the narrow KMT-led majority in the Legislative Yuan who initially opposed passing a referendum law but came to support the idea when its strong popular support became apparent.The final law set a high threshold for approval, support by 50% of all registered voters, which has greatly limited its effectiveness.For example, Chen was able to hold referenda simultaneously with the 2004 and 2008 elections, but they were defeated by boycotts of the referenda organized by the Kuomintang (Clark 2006;Gold 2009).
The 1947 Constitution of the Republic of China, to sum, created an unwieldy government whose institutional incoherence was covered up by the KMT's authoritarian rule.The shortcomings of the constitutional system became increasingly apparent as the democratic era progressed; and a series of constitutional reforms were enacted between 1991 and 2005 which sought to redress these problems.These reforms were certainly successful in some regards.The institutional basis for a democratic polity was constructed; several bodies of dubious effectiveness (e.g., the Provincial Government and the National Assembly) were eliminated; and, more recently, the SNTV election system was reformed.Still, the continuing gridlock and polarization in Taiwan demonstrate that the country's institutional imbroglio has yet to be satisfactorily resolved.These shortcomings are often attributed to the fact that the constitutional reforms have been incremental, piecemeal, ad hoc, and opportunistic, both by those who want a more effective government and by those who advocate a new constitutional order (Rigger 2007;Yeh 2002).

Explaining Taiwan's Polarization and Gridlock by Its Institutional Imbroglio
Taiwan's democratic era, welcome and successful as it has been in many regards, has been marked by growing political gridlock and polarization over the divisive national identity issue and by the politicization of economic policy.To some extent, this represents a cost of success from democratization because the authoritarian Kuomintang party-state papered over a variety of problems in the country's political institutions.First, there was substantial institutional incoherence, in particular, with unclear and overlapping authority among the President, Premier, and Legislative Yuan.Second, a Constitution designed for a "tutelary democracy" in continental China had what seemed like superfluous institutions (e.g., the National Assembly and the Provincial Government) when applied to democratic Taiwan.Third, the SNTV election system was widely perceived to promote corruption and extremism.Finally, democratization brought a much greater politicization of economic and financial policy-making.Taiwan has adopted a series of constitutional amendments over the last two decades to address these problems.Their effectiveness has been limited, however, both because most have been piecemeal and/or opportunistic and because the nation suffers from a major rift over the legitimacy of the existing Constitution.Figure 3 sketches the dynamics by which institutional legacies of the constitutional and electoral systems led to several untoward consequences.As indicated in the top part of the figure, the authoritarian tradition blurred governmental lines of authority and helped stimulate very personalistic politics.These factors, in turn, had several deleterious consequences for policy-making, such as never-ending conflict and gridlock, burgeoning corruption, and the politicization of supposedly technical policy realms, such as financial regulation.For its part, as diagrammed in the bottom part of the figure, the election system of the "single nontransferable vote" (SNTV) contributed both to personalistic and patronage politics and to the election of extremist candidates who led the polarization of elite politics on the highly emotional issues of national identity and cross-Strait relations (Fell 2005;Lee 2005;Makeham and Hsiau 2005;Wachman 1994).
John Fuh-sheng Hsieh (2006) puts this into a broader and more theoretical perspective.He argues that constitutional systems can be ranked along a continuum from the liberal objective of protecting human rights to promoting efficient policy-making, with the checks-and-balances of a presidential system promoting the former and the unified decision-making of a parliamentary system conducive to the latter.Election systems, similarly, can promote the populist value of individual representation or efficient policy-making by majority parties with PR systems providing the former and SMD ones the latter.He then uses these distinctions to create a typology of four different kinds of democracies.He classifies Taiwan as a presidential system in practice (though fairly parliamentary in constitutional design) and quite populist, at least under the old SNTV system.This creates a "hyperdemocracy" which, according to Hsieh, is the least desirable type because of its tendency for political stalemate and ideological polarization, exactly the problems facing Taiwan today.This might imply that the institutional imbroglio in the ROC is intractable.Yet, such problems should not obscure the substantial advancements that the nation has made politically.Taiwan is clearly a well functioning democracy.Elections are held regularly; and their results are accepted, albeit with a lack of good grace at times.Governmental transitions from one party to another are managed smoothly.Clearly, the country has had very significant democratic successes.Indeed, as described in the first section, democratic dynamics initially produced a deescalation of polarization on the volatile national identity issue.

Figure 3 :
Figure 3: How two political institutional legacies proved dysfunctional during Taiwan's democratization.