Taiwan recently adopted a series of administrative reform laws designed to make the bureaucracy more transparent and allow public participation in regulatory policies. Because administrative reform limits the executive's power, it is clear why legislatures would favor strict administrative procedures. But it is less dear why presidents would support them. The passage of these laws begs the question why presidents support administrative procedural reforms designed to restrict their abilities to act freely. I argue that in Taiwan, President Lee Tenghui's control of his party deteriorated as factional disputes within his own party increased over time. Lee ultimately concluded that the Kuomintang's political survival depended on major reforms. Consequently, the status quo--oriented bureaucracy--hitherto an important source of support for Lee and his key constituencies--became an impediment. Lee supported Taiwan's Administrative Procedure Act in order to reduce the bureaucracy's capacity to impede reform. More generally, I argue that administrative procedures designed to open up the bureaucracy to the public, including previously excluded groups, can serve politicians' goal of redirecting the bureaucracy. Archival data, secondary sources, and interviews with key presidential advisers, senior career bureaucrats, and politicians support my argument.
KEYWORDS: administrative procedures, black-gold politics, bureaucratic transparency, corruption, KMT factions, public participation, Taiwanese politics
Now under the administrative reform by the KMT, there is a second track for appeals: They [citizens] can start at the Executive Yuan [branch]. When people fail there, then they can appeal through the judicial system. In turn, they can face each other and have a debate between the government and the people. In other words, we have created a situation where the third party, or the judiciary, can arbitrate between the people and government. (1)
--Yao Eng-Chi, Vice President, Legislative Yuan (July 15, 2000)
During the past decade, scholars have tried increasingly to understand the origin of the administrative procedures that govern policy implementation. The nature of the procedures used by government agencies to implement statutory mandates not only determines what groups are able to participate in rule-making but also ultimately affects the nature of policy. In the absence of procedures designed to offset it, there is a common tendency for agencies to become "captured," often by the interests they are intended to regulate. One primary effect of procedural openness is that broader interests and those with fewer political resources are enfranchised. Another is that status quo policies become harder to change. Particularly powerful in this regard are statutes that establish procedures to be used by all agencies, across all policy areas. In the United States, the primary statute of this type is the Administrative Procedure Act (APA) of 1946. Many other countries have APA-like laws, but many do not.
Taiwan recently adopted a series of administrative reform laws, including an APA, designed to make the executive branch more transparent and allow public participation in regulatory policies. For example, agencies are required to notify the public, incorporate public comments, and hold public hearings during the policymaking process. These procedural requirements empower individual citizens and organized interest groups to actively voice their opinions about virtually every government decision. By enfranchising new segments of society, the APA increases monitoring, predictability, and public influence over agency decisions. Yet, they also inhibit presidents' freedom of action. Thus, the passage of these laws suggests a puzzle. Why would a president voluntarily tie his or her own hands by supporting an APA, thereby limiting his capacity to unilaterally pass his preferred policies? In particular, when do the incentives of presidents and legislators cause them to codify procedural openness?
One possible explanation is that APAs are adopted when politicians want to make it hard to change status quo policies. This would occur in a system of separated powers--for example, when both the president and the legislative majority expect to lose the next election. This theory has been proposed by McNollgast (2) to explain the passage of the 1946 APA in the United States. This is not, however, the only possible explanation. An alternative theory is that the 1946 APA was passed to help politicians monitor and influence the decisions of bureaucrats within the current administration. A third possibility, implicit in much legal research, is that APAs are passed whenever the size of the regulatory state becomes sufficiently large, as in the post-New Deal United States. However, while a large regulatory state may be necessary for an APA, it is clearly not sufficient. Many countries with a large regulatory bureaucracy do not have APAs.
Taiwan is a particularly appealing country for this research question, since an APA passed after some unsuccessful attempts and some periods in which passing an APA was not on the agenda of those in power. By treating the administration as the unit of observation, I am able to get multiple observations within a single country, beyond the single case of the United States. In fact, an analysis of the passage of the Korean APA has produced evidence contrary to the predictions of McNollgast's theory that APAs are designed to make it harder for future politicians to change status quo policies. (3) I argue that the Taiwan case exemplifies a more general explanation for why chief executives support administrative procedural reform.
Theoretically, any executive--a president or a cabinet minister, for example--can have control problems within his or her own branch. In Taiwan, President Lee Teng-hui's control of his party deteriorated as intraparty factional disputes increased over time. Eventually, Lee determined that the political survival of the Kuomintang (KMT) depended on major reforms. At this stage, the status quo-oriented bureaucracy--hitherto a key source of support for Lee and his primary constituencies-became an impediment. Consequently, for Lee, the immediate benefit of resolving current delegation problems--that is, preserving the KMT's political dominance--outweighed the potential longer-term cost of tying his successor's hands. (4)
Specifically, I argue that the reform wing of the KMT was electorally strengthened by a shift in voter preferences, creating a new KMT median legislator. The KMT reformers wanted to change the party's image from one of entrenched corruption and of favoritism toward wealthy private interests to one of a "clean government" party. To do so, the party had to rein in the very bureaucracy that had long served as the cornerstone of its power. Under the status quo, bureaucrats' regulatory decisions consistently favored corrupt private interests. This bias both tarnished the party's image with a large segment of the electorate and produced policy outcomes that alienated the nation's fastest-growing electoral demographic: younger voters. This provided reformers within the KMT an incentive to manage current delegation to the bureaucracy with strict procedural requirements.
This article proceeds as follows. I begin, in the next section, by providing an overview of my theory. I argue that presidents who con front current control problems within the executive branch are likely to support an APA. In the subsequent section, I discuss the origins of Taiwan's APA (hereafter TAPA), its legislative history, and the regulatory requirements it contains. In the next section, I substantiate my theory by reviewing the institutional and political environment during two presidential administrations in Taiwan. I argue that TAPA was designed to promote change in the status quo rather than protect it, as the competing "lock-in" hypothesis (5) implies. The evidence from Taiwan--consisting of archival data, secondary sources, and interviews with key presidential advisers, senior career bureaucrats, and politicians--supports my argument.
A Theory of Intrabranch Delegation
The literature on delegation in democracies is voluminous. Among the most notable insights from this literature are how legislative strategies such as monitoring, (6) oversight,(7) and sanctions (8) induce bureaucrats to serve legislators' interests. Other scholars have shown how information asymmetries between legislators and agencies affect the optimal amount of delegation. (9)
Mathew McCubbins, Roger Noll, and Barry Weingast (10) suggest a mechanism for bureaucratic control by the legislature--the strategic use of administrative procedural mandates to "stack the deck" such that the legislative intent of the enacting coalition is preserved. Such procedures as subsidizing the participation of favored interest groups, adjusting the burden of proof, and enfranchising new interests are means by which the procedural deck may be stacked in favor of the interests of the enacting coalition. These procedural requirements serve as valuable tools for elected officials in governing information collection and dissemination by agencies.
While the scholarly literature has mostly emphasized the U.S. Congress delegating to a unified executive branch, the logic of interbranch delegation should apply to delegation within the executive branch as well. The ability of presidents to control the agencies and bureaus nominally under their authority is arguably of equal importance for democratic governance. Yet, with several noteworthy exceptions (11)--focused primarily on the United States--the latter topic has been understudied. (12) Executives play a significant role in many new democracies, and they would seem to have easier ways to solve delegation problems than adopting formalized administrative procedures. After all, why not simply appoint heads of agencies that agree with them and fire them if they do not behave as expected? Yet, executives do not always have the constitutional authority to appoint and dismiss freely. And even if they do, there are consequences to firing the appointees of a coalition partner. (13) First, executives will likely have problems passing legislation during their terms in office. And, second, executives who dismiss appointees from their coalition partners' parties are likely to see their own parties lose credibility in future coalition negotiations. Hence, for chief executives, relying on constitutional means to control their agents may, in practice, be too costly.
Variations in presidential power can arise either from constitutional or legal restrictions on a president's ability to appoint ministers or bureaucrats, or from having to share political power with allies who have different preferences. The less autonomous presidents are, in either sense, the greater the risk that their underlings will act contrary to their interests, and hence, the more interested presidents will be in instituting procedural controls to mitigate this risk.
I argue that the more intrabranch conflict presidents face while in office, the more likely they will be to support an APA, or similar law. Thus, while previous theories (14) view APAs as tying the hands of future administrations in order to protect current policies favored by legislators, I argue that presidents can also use APAs to help them overcome current control problems. Current control problems--that is, the inability of presidents to effectively manage their executive agencies--in turn, can arise in several different ways, any of which can induce presidents to support an APA or similar law. I discuss each of these circumstances in the following paragraphs.
Appointment and Dismissal Powers
Presidential powers vary across different separation-of-power constitutions. (15) In particular, presidents cannot necessarily perfectly control who can be in their cabinets. While most presidential systems do not require legislative approval of appointees, many countries give the assembly some influence over cabinet formation. Consequently, divergent preferences within the executive are more likely to arise and persist in these latter cases, compared to countries where presidents have exclusive control.
If the president's policy preferences differ from those of the dominant forces in the legislature, a situation of divergent preferences between the president and the cabinet ministers could arise. This implies that the more the president's ability to appoint and dismiss cabinet ministers is constrained by a constitutional requirement for legislative approval and the greater the divergence between presidential and legislative preferences, the more the president will be likely to support administrative procedural constraints.
Presidential power can also vary by the size and cohesiveness of presidents' political support/coalition base. Variation in that coalition in the legislature affects presidents' ability to control the executive branch. Appointment powers are particularly critical for understanding the level of conflict that arises in the executive branch. Presidents who are constrained in their appointment powers or whose party does not have a legislative majority might appoint a multiparty coalition cabinet. Doing so would mean a greater likelihood of conflict arising from the heterogeneous makeup of the cabinet.
Intra- and Interparty Conflict
When ministers are selected on a partisan basis, presidents must share executive powers with coalition partners from other parties, most likely representing different constituencies and policy preferences. The other parties sharing the executive branch are likely to oppose at least some of the president's preferred policies, which may differ significantly from their own policy agenda. When this situation arises, there is a greater possibility that the president's party will be unable to achieve its preferred policies because it will have less influence over executive agency activities. Therefore, in order to reduce a cabinet minister's ability to move policy toward his or her own party's ideal point, (16) the president has a greater incentive to favor ex ante versus ex post controls. Ex ante controls, such as extensive administrative procedures, reduce coalition partners' informational advantage. Ironically, it is precisely because of this two-way benefit of reducing hidden information and hidden action that, in a coalition cabinet, the partners would also tend to support an APA.
Factionalized single-party cabinets also threaten presidents' agendas. Within such cabinets, presidents are more likely to face ministries whose preferences differ from their own and thus have an incentive to institutionalize their control, ex ante, by supporting administrative procedures. As a consequence, I anticipate that presidents are likely to prefer more extensive procedural constraints on bureaucrats when their support coalition is either a multiparty coalition or a factionalized ruling party.
Civil Service Autonomy
Meritocratic recruitment and promotion of the civil service restrain political patronage. At one end of the continuum, the civil service might be highly professionalized with a merit-based recruitment and promotion system, giving presidents less control over their bureaucracies. At the other end of the continuum, the bureaucracies might be highly politicized, with a personalistic, patronage-based system, giving presidents greater control over their agents. Because promotions are politicized, civil servants in this system worry more about satisfying their superiors or influential politicians.
If civil servants are protected under the law (e.g., cannot be easily fired), there is a greater possibility of bureaucrats not complying with elected politicians' policy preferences. From the civil servant's point of view, knowing that firing cannot result from his or her actions, there is little risk in resisting the president's preferred policies. Therefore, presidents who cannot unilaterally fire civil servants need alternative mechanisms to constrain their behavior. Hence, I anticipate that presidents will prefer more extensive procedural constraints if they confront bureaucrats they cannot replace who do not share their preferences. (17)
Taken together, these several circumstances under which current control problems are likely to arise suggest that presidents are more likely to prefer APAs when they are weak. Whether their powers are constrained by constitutional requirements (appointment and dismissal power of the cabinet), coalitional support with heterogeneous preferences, or bureaucrats who cannot be fired, APAs can mitigate agency slippage by keeping track of bureaucrats' activities. In the next section, I argue that Taiwan's APA resulted from a combination of intraparty conflict and a recalcitrant, yet institutionally protected, bureaucracy.
Origins and Contents of TAPA
The impetus for the APA in Taiwan came from the KMT's nonmainstream faction. In September 1989, Premier Hau Pei-tsun, a military general under President Lee, created a research panel under the Council of Economic Development to draft TAPA. (18) A grand justice of the Judicial Yuan chaired the panel with eight legal scholars as drafters. The panel considered APAs from the United States, Germany, and Australia, as well as Japan's draft at the time (1993 Administrative Procedure Law) and Korea's draft at the time (1994 APA). The legislative debate and the actual passage of the APA spanned ten years.
One reason for the delay was that, following Premier Hau's departure in February 1993, subsequent premiers were not as active in pushing the bill. For example, Hau's successor, Lien Chan, was a member of Lee's faction rather than of the rival faction. During this delay, two legislators from the Democratic Progressive Party (DPP) (1993) and KMT (1994) introduced alternative drafts of an APA. Interestingly, the opposition DPP legislator introduced the panel's version. They sent both drafts to committee (jointly reviewed by Organic Statutes, Interior, and Judicial Committees). In April 1995, the Executive Yuan introduced its draft and sent it to committee. Over the next year and a half, the committee held five hearings, one of which was public.
Legislators from the New Party (NP)--composed of rightist former KMT members who defected in 1993--and KMT legislators with clean--that is, uncorrupt--reputations chaired all hearings. Shortly thereafter, the KMT legislator who introduced the alternative APA draft invited a legislator from the NP to promote TAPA. After reviewing the various versions, the NP legislator drafted her version and introduced it with another legislator from her party in December 1996. Of the thirty-one sponsors or cosigners, ten were NP, eleven were DPP, eight were KMT, and two were independents. Overall, this list is notable primarily for the absence of the local faction members and representatives of big business typically associated with corruption and money politics. The NP legislator commented:
After I proposed my version, the Executive Yuan tried everything they could to stop the passage of such a law. The Executive Yuan ... of course, understandably opposed the passage of replacing administrative orders or regulations with formal law. I can understand their situation but I totally disagree. What I mean is that if I was a bureaucrat, I would want as much discretion as possible and not be constrained by procedures mandated by law. (19)
After the NP introduced its draft bill, the reformers held five committee hearings, again all chaired by NP, DPP, and KMT legislators with clean reputations. In May 1998, the committee passed the bill after nine committee hearings and one public hearing. By comparison, a typical bill in Taiwan passes after only one or two committee hearings.
According to several interview subjects, the Executive Yuan's opposition to the substance of TAPA afforded the legislature more opportunities to manipulate the legislative process. For example, the NP legislators were able to invite many scholars to participate in the process, thereby increasing the act's legitimacy and broadening support for a more comprehensive TAPA. Prior to the final resolution, legislators conducted a closed session, involving interparty negotiations. On January 14, 1999, the floor considered the bill and passed the second and third readings without any vote or opposition.
The basic provisions include a public notice and comment process of all regulations, public access to agencies' information, notification of all public hearings, and judicial review of administrative actions. These provisions are designed to increase the degree of public information, participation, and accountability of agency decisions. Bureaucrats interested in continuing corrupt practices were the clear losers.
TAPA's Requirements and Their Effects
Table 1 provides a list of TAPA's key administrative procedural requirements to be used by all executive branch agencies for promulgating regulations across all policy areas. In addition, I describe the political impact(s) of each change in the policy process.
TAPA includes extensive technical detail. For instance, the notice and comment and judicial review provisions are the most central requirements of Taiwan's administrative law and are quite detailed. A notice and comment process requires agencies to notify the public and allows them to comment before publishing regulations. Upon receipt of public comments, agencies must respond to every comment. Theoretically, this can take years. So, whereas prior to an APA agencies wrote regulations without having to jump through a series of procedural hoops, today a final regulation can take years to promulgate. In the case of judicial review, the public can go to court to try and convince a judge to reverse a regulation they oppose. Again, this process can take years to resolve. It is also worth noting that, unlike the U.S. APA, the judicial review provision is included in the 1998 Administrative Litigation Act (ALA). The ALA authorizes a separate administrative court to review agency actions.
The major political effects of these arduous procedural requirements concern enfranchising new groups so as to increase monitoring, predictability, and control of agency decisions. These effects relate to intrabranch conflict because they solve delegation problems for executives. Each requirement is a tool for executives in governing information collection and dissemination by agencies. The political implications are important for several reasons. First, the notice and comment process forces bureaucrats to reveal information, and their activities become increasingly observable to the public. This forces agencies to consider the political costs and benefits of their actions. Second, the public nature of the process decreases the likelihood that decisions made between agencies and their favorite constituencies will conflict with the preferences of the legislature or the president. Third, each step of the process gives executives multiple opportunities to intervene if an agency deviates from their preferred policies. (20) To the extent that executives conflict with their underlings on issues of public policy, these tools reduce agents' capacity to deviate from the policy preferences of their principals.
Why Taiwan Passed an APA
In this section, I examine two episodes of administrative procedural codification in the Taiwanese case: the 1999 passage of TAPA and the prior "nonevent." My dependent variable is the change in the degree of administrative procedural control during a particular administration. I begin by discussing the first Lee administration, where no APA passed. I then turn to a discussion of the institutional and political changes during Lee's second administration, which provided the impetus for passing TAPA. In the latter discussion, I first explain the KMT's declining electoral performance and the increasing influence of the reform-minded members within the party. Next, I explain how these institutional changes led Lee and his party to view his bureaucracy as a potential detriment to the survival of the KMT. Finally, I provide evidence that APA enactment resulted from the president's concern about controlling the bureaucrats, who could not be fired, and a coalition of reform-minded legislators.
Intraparty Conflict and Defection of the Nonmainstream Faction
President Lee did not have "current control" problems with the bureaucracy during the 1988-1996 period. Since he did not have to concern himself with managing delegation to like-minded bureaucrats, he lacked an incentive to constrain the bureaucracy through an APA. After all, the KMT had ruled Taiwan since 1949 and controlled all branches of government until Chen Shui-bian (DPP) won the 2000 presidential election. However, some intra-KMT factional conflict arose in the early 1990s. In 1991, for strategic reasons, Lee chose as his premier a prominent general, Hau Pei-tsun, from the nonmainstream faction. Prior to 1993, the main factional conflict involved Lee's "mainstream" faction and the "nonmainstream" faction of mainlanders representing the party's old guard. (21)
Shortly after the December 1992 elections, "factions within the party used the KMT's poor performance as a base to launch their attacks against the party mainstream." (22) Internal party conflict continued:
By the Spring of 1993, the battle lines of inner-party strife became especially clear, with the creation of the Chinese Democratic Reformers Alliance. Factional splits within the party now threatened the future of the KMT's electoral success. More importantly, nonmainstream factions came to view a KMT loss in upcoming year-end county elections to be in their favor because the mainstream would have to call on the mainlander factions for help. (23)
In August 1993, portions of the nonmainstream faction broke off and formed the New Party. (24) Moreover, factional differences led to serious rifts with important consequences for party organization:
Leaders were unable to stop the diminution of power from the center, and the growing influence of local factions. Factional infighting led to multiple party vice chairmen and a more diverse central committee and central standing committee, thus weakening the largely singular role of party chairmen like Chiang Kai-shek and Chiang Chingkuo. By the 15th party congress, even the central committee and central standing committee had lost the importance they held in the past. Power had shifted almost completely to the elected bodies.
With the defection, the Legislative Yuan was now composed of three types of legislators, representing three political parties: the KMT, DPP, and New KMT Alliance/NP. And, post-1995, with the KMT commanding only slightly more than half of the seats in the legislature, it became increasingly critical for the party to maintain its unity. Otherwise, the party--both mainstream and proreform factions--was vulnerable to a no-confidence vote. Indeed as Cheng and Haggard point out, Lee had the support of the new KMT legislators from the mainstream faction, but the rise of the opposition and the threat of outright factional defection--which ultimately came to pass--meant that he needed to appease nonmainstream legislators as well. (25)
Institutional Changes and KMT's Declining Electoral Performance
Table 2 summarizes the institutional changes that affected the powers of the legislature and the presidency. These changes, in effect, altered both the president's and legislators' expectations about the future, and thus their incentives. For my purposes, the key change, in terms of weakening the powers of the chief executive, is the lowering of the no-confidence vote threshold from two-thirds to a simple majority.
The KMT's electoral performance has also deteriorated over the past two decades. Figure 1 shows this pattern for both the KMT's vote and seat shares. These data indicate that in 1995, the KMT's majority reached an all-time low of 52.3 percent (85 seats, representing a razor-thin three-seat majority). Though the party recovered somewhat in 1998, capturing a 54.7 percent share (123 seats), under the post-1997 simple majority no-confidence rule, Lee clearly had reason to worry not only that his party might lose its capacity to control the policy agenda, but also that he could lose power. In fact, the DPP, who captured 33.2 percent of the popular vote and 32.9 percent of the seats (54), and the NP, who captured 13 percent of the vote and 12.8 percent of the seats (21), began discussing the possibility of a "grand reconciliation." Hence, the KMT justifiably felt intense pressure to reverse its electoral slide. (26) Indeed, the coalition of legislators pushing for the APA consisted of a group of DPP, NP, and KMT members with reputations for opposing corruption.
[FIGURE 1 OMITTED]
The 1997 county and city magistrates election results gave the KMT increasing cause for concern about its capacity to retain power. Not only did the DPP surpass the KMT vote for the first time (43 percent of the votes), but they won twelve districts and already controlled the Taipei City mayorship. (27) Combined, these thirteen districts held over 70 percent of the total population of Taiwan. "With control of Taipei as well, this gave the DPP executive power in local governments containing more than 70% of Taiwan's population." (28)
Much of the so-called black-gold corruption (discussed in a subsequent section) took place at local levels. Consequently, this electoral outcome was particularly important. These results, along with the 1998 Kaoshing mayoral race loss by the KMT (48.13 percent votes for the KMT versus 48.71 percent for the DPP), gave the KMT ample reason to worry that black gold would hamper the party's chances of winning at the national level.
Breaking Corrupt Practices and the Preferences of TAPA's Support Coalition
Anticorruption as an emerging priority. With democratic transition and the implementation of popular elections, the political survival of elected officials, including the president, depended on winning popular votes and therefore on satisfying the demands of their constituents. This entailed changing specific status quo policies, especially in those issue areas most salient to the public. And, in recent years, inadequate public infrastructure and environmental protections have become among the most visible concerns in Taiwan, particularly among younger voters. Voters, in turn, closely associate institutional corruption with these policy areas. (29) This is because corruption is increasingly seen as a primary cause of Taiwan's environmental problems and lax enforcement of public health and safety standards in public infrastructure projects. (30) In fact, "since 1992, the DPP has sought to make "money politics" and "mafia politics" the most salient issue in Taiwan's elections. The muckraking efforts of the opposition lawmakers were assisted by a growing number of independent-minded prosecutors and the market-oriented mass media." (31)
Public opinion data attest to the success of the DPP's efforts. Between 1992 and 1996, survey data show a clear trend toward a worsening KMT image among the general population. For instance, when asked whether the KMT "accommodate[s] or fight[s] corruption," the percent answering "accommodate" increased from 18 percent in 1992 to 42 percent in 1996. When asked whether the KMT "fight[s] or cooperate[s] with special interests," the percentage selecting "cooperate" increased from 21 percent in 1992 to 33 percent in 1995. Finally, when asked whether the KMT "represent[s] rich and powerful or average people," the percentage choosing "rich and powerful" increased from 25 percent in 1992 to 44 percent in 1996. (32)
Overall, according to public opinion polls, corruption has emerged in recent years as the Taiwanese public's top concern. (33) For instance, a poll taken several months before the 2000 presidential election showed that the public perceived the ultimately victorious opposition DPP candidate as having the greatest integrity. Twenty-six percent of respondents named the DPP candidate as having the most integrity, compared to 16 percent who named the People's First Party (PFP) candidate, (34) and 11 percent who mentioned the KMT candidate. (35) Indeed, popular perceptions in Taiwan held that "under the KMT, corruption and cronyism thoroughly infested government at all levels." (36) On this point, President Chen Shui-bian of the DPP asked the public to "join him in reforming government and ridding it of the 'black-gold,' or corruption money, endemic in [Taiwanese] politics." (37) Indeed, Chen's signature campaign promise was to "dump the KMT gang and money political corruption." (38) Moreover, Taipei mayor Ma Ying-jeou, a popular KMT figure, commented that "the KMT was in danger of 'bursting like a bubble' and is proposing a string of changes to clean up the party and make it more democratic and accountable." (39)
In turn, according to Taiwan politics expert Shelley Rigger, "The  election turned on the issue of corruption and reform. And apparently Chen had a little more credibility as a reformer than [independent candidate] Soong." (40) Chin observes, even more starkly, that "black-gold politics developed into a major problem during the past fifteen years and ... it might have ended KMT rule in Taiwan during the 2000 presidential election." (41)
Black-gold politics--including KMT ties to organized crime--thus mattered to the KMT reform faction because these policies alienated large numbers of voters, who were increasingly concerned with the consequences of corruption for public policy. This, in turn, represented a tangible threat to KMT political dominance in Taiwan. Indeed, younger voters (representing over half of the electorate) increasingly demanded improved public infrastructure and greener environmental policies. (42) Reflecting this perspective, Ray Chen, CEO of Compal Electronics Company commented, "We need the government to create transportation and other infrastructure.... We need to make environmental regulations that are strict and clear, with no gray areas." (43)
As consumers and environmental advocates began to recognize the public health and safety risks posed by the KMT's corrupt practices, these issues, along with the problem of corruption itself, became highly salient. (44) The KMT routinely won elections when the only noteworthy issue was relations with China. But, as these other issues became more and more conspicuous to voters, the KMT's advantage eroded. Political entrepreneurs--like the DPP--determined that by identifying themselves as the party of clean government (as well as the party of all the people, a clean environment, high-quality public infrastructure, public safety, and so on), they could use these newly salient issues as a means of challenging KMT dominance. (45)
This, at least in part, explains why former Taipei mayor Chen Shuibian, DPP candidate for the 2000 presidential election, pledged "to end money politics, cozy ties between government and big business and other 'abnormal' practices entrenched after half a century of KMT rule." Chen continued to point to the "KMT's stranglehold on the entire government machinery, underworld ties, one-sided pro-business policies ... environmental destruction" while adding, in classic populist overtones, "Only if the DPP well mobilizes the Taiwan people will we be able to win final victory." (46) In addition, during the 2000 presidential election, Chen's campaign strategy--including TV commercials "featuring young people in a bid to drum up support among young voters" (47)--demonstrated the increasing importance of younger voters, particularly for proreform candidates.
Hence, the DPP saw the changes in Taiwan's electorate as an opportunity to undermine the KMT. But they faced competition from the KMT reformers who also believed they could capture young reform-minded voters. (48) After all, they had the advantage of the KMT brand name--albeit a somewhat tarnished brand name--and they too could campaign for an end to corrupt practices and, in doing so, promote better public health and safety standards and enhanced environmental protection. If they could steal these issues from the DPP and change the KMT's image from the "party of corruption," to "the party of clean government and national security," then they might be able to retain power.
In fact, James Soong, a KMT nonmainstream faction leader who sought but failed to win the party's nomination and so ran as an independent presidential candidate in 2000, pursued precisely this strategy. (49) Along these lines, the Times of London thus observed that Soong's "main constituency consists of voters who favor ending corruption but fear that Mr. Chen, the DPP candidate, could provoke a Chinese invasion." (50) In an effort to accomplish their goal of resuscitating the party's image, the reformist faction of the KMT vowed to begin a purge of party elements responsible for corrupt practices. Chu, a party youth leader and reformist, commented, "This is a turning point for the KMT. We are not stupid. We need the votes--not the money. This reformist faction will strive to be the dominant force in the KMT. We would like the party to be more clean." He added that if the proreform forces succeed in gaining control of the party, they might return KMT assets to the public. (51) And Lin Feng-cheng, secretary-general of the KMT Central Committee, commented shortly after assuming office that his goal was to transform the party into a "young, localized, pluralistic, grass-roots, volunteer-based and democratic" party, more in tune with the times. He added that the KMT is "sincere in its reforms, and that the thoroughness of the reforms will not disappoint the expectations of the people." (52)
Three high-profile cases in 1996 and 1997--which drew public attention to black-gold problems just before the KMT's big loss in the 1997 county and city magistrates elections--offer further evidence of the KMT's critical need to respond decisively to the issue of rising crime and money politics. First, in November 1996, the head of the Taoyuan county government was murdered. Gang members broke into his residence, killing him and several associates. The motive was reputedly a dispute over fixing land prices. Second, in late 1996, Peng Wanlu, head of the DPP's women's bureau, was abducted, raped, and brutally murdered. The official investigation revealed no political motives, but the murders were widely reported as being politically motivated. At minimum, the connection of a politician with violent crime was bad for the government's image. Third, in April 1997, the daughter of actress Pai Ping-ping was abducted and murdered. (53)
"Operation Chih-ping," in 1996, involved a major crime sweep targeting organized crime members who were elected politicians. Though not particularly effective, it had a high public profile (e.g., media images of crime lords being whisked away by helicopter to the prison on Green Island). By conducting this high-profile anticrime sweep, the government intended to show that it was serious about fighting money politics, crime, and corruption. (54)
Courting younger voters. Most voters under age 50 were born on the island of Taiwan, and hence lack their parents' emotional ties to the mainland. (55) The following comment by a twenty-two-year-old Taiwanese voter, in a preelection interview, reflects changing of the political guard in Taiwan: "Taiwan is independent now in reality. I don't identify myself with mainlanders because I don't have any life experience similar to theirs." (56) Another twenty-three-year-old voter added, "I think I'm Chinese--but not mainland Chinese. When I travel abroad, I say I'm from Taiwan, not from China." (57)
As younger voters who identify themselves as native Taiwanese have emerged as a dominant population demographic, politicians in Taiwan have become increasingly concerned with winning their votes. And a growing number of these voters have become fed up with the KMT's corrupt practices in the government. (58) Not only do younger voters recoil from the KMT's perceived cozy relations with organized crime, but they are increasingly alienated by the policy outcomes such illicit relationships produce. (59) Under the status quo regulatory environment, the primary "winners" in Taiwan were big business and the KMT-dominated bureaucrats who benefited from the distribution of public works contracts. These public works contractors cut corners and built infrastructure projects out of inferior materials that led to crumbling buildings and bridges. Dexter Roberts et al. (60) summarize the relationship between corruption in Taiwan's contracting industry and rising concerns over public safety as follows:
The construction industry illustrates how corruption also has imperiled public safety. Faulty work was responsible for many of the 2,321 deaths in last September's earthquake, for example. Investigators allege government inspectors were paid off by contractors to overlook flaws in apartment buildings.
In addition, proenergy industry policies led to relatively unregulated construction of chemical and nuclear power plants. Many of these projects raised serious environmental concerns among the public. These concerns came to a head in 1997, when a nearly $2 billion chemical plant project was brought to a halt following accusations of corruption and inadequate environmental protections. (61) The corruption problem even extended to the military:
A three-star general who headed the body responsible for handling military supplies and provisions resigned to accept responsibility for a bribery scandal over the construction of ammunition dumps for the Air Force. Nine other military officers, including a two-star general, have received various penalties for dereliction of duties relating to the case. (62)
As a result of this and other similar incidents, the KMT gained a reputation as the party of corruption, money politics, and organized crime. (63)
At the same time, as Taiwan grew wealthier, younger voters became increasingly concerned with quality-of-life issues, such as public health and safety and environmental protection. (64) Lin I-Hsiung, a leader of the DPR summarized the priorities of Taiwan's younger voters as follows: "We want to live on our island, protect the environment and have good relations with the mainland." (65) No longer focused solely on the issue of national survival, younger voters began to demand more from government than merely national security. Increasingly, they also wanted a better quality of life. And the corrupt deals of the KMT put public safety and the environment at risk. In other words, the "losers" under KMT rule were consumers and environmental groups who were excluded under the corrupt status quo regulatory decisionmaking process.
For the KMT, one potential means of altering its public image in order to court younger voters would be to appoint younger party members to senior posts. To be effective, such a strategy would need in turn to focus on mainlander party members who represented the party's "old guard." Figure 2 presents the trend in average ages of mainlander and native Taiwanese cabinet ministers from 1988 to 1998.66 In fact, the average age of cabinet ministers belonging to the mainlander ethnic group did indeed decline substantially--from 68 between 1988 and 1989 to 58 between 1997 and 1998 period (p < .01). Over the same period, the average age of nonmainlander ministers--who tended to be far younger at the outset--changed hardly at all. In the former period, mainlander ministers were, on average, eleven years older than their nonmainstream counterparts (68 versus 57, p < .01). By the 1997-1998 period, the age gap had disappeared almost entirely (58 versus 57.8). These data suggest that Lee may indeed have sought to present a more youthful face to the electorate, in part by appointing younger mainlander ministers.
[FIGURE 2 OMITTED]
"Heijin (Black-gold) Politics" and the Bureaucracy
Changing status quo policy outcomes in Taiwan required controlling the bureaucrats. To see why, it is necessary to consider in greater detail the relationship between bureaucratic corruption and public policy. So-called black-gold politics is the central corruption issue in Taiwan. (67) In Taiwan, black (hei) means the underworld; gold (jin) means money or business. "Black-gold politics consisted of the penetration into politics of violent underworld figures and greedy business tycoons and the inevitable subsequent social ills such as vote buying, political violence, insider trading, bid rigging, and official (and unofficial) corruption." (68)
The term black gold evokes the image of bribery and gangsters associated with the influence of big business in contemporary Taiwanese politics. Under this system, "public works contracts are routinely let out to unqualified but well-connected companies." (69) The result, according to Rigger, is that much of the industry "is controlled by government and gangsters who build buildings out of tin cans." (70) Moreover, Chin explains, black-gold politics is a direct consequence of a steadily declining ruling party adopting extreme measures to maintain its hold on power:
The KMT hoped to counter the growing challenge from the indigenous political party, the DPP, which had been encroaching on its power base by winning a significant number of seats in major elections. While facing this daunting challenge from outside the party, the split between two factions within the KMT, and the subsequent establishment in 1993 of a third party--the New Party--by former KMT members, all the more convinced KMT leaders that their alliance with local factions, business tycoons, and gangsters was [a] necessary evil if they wanted to remain in power. (71)
Black gold is also linked to a $6 billion slush fund from which the KMT reportedly drew money to buy votes. (72) The KMT's possessions, which in 1999 yielded $261 million in net profits from equity investments alone (see Table 3), consisted mainly of shares in seven major holding companies, listed in Table 4. The majority of these were highly profitable enterprises formerly owned by the state. "The remaining assets are in the form of stakes in dozens of listed and privately held companies in Taiwan and abroad, including mainland China. Key stakes in these concerns were held by prominent local families, which had grown rich and powerful under the patronage of the KMT regime." (73)
As Table 3 shows, the KMT's total assets more than tripled between 1993 and 1999. This increase occurred under the management of Liu Tai-ying, a close friend of President Lee. During this period, opposition parties, as well as the reformist faction within the party, criticized the KMT for using those assets more for vote buying than for improving the country. Prior to the 1980s, the KMT had funded infrastructure projects and started key businesses that the private sector would not support. But when the KMT began to implement political reforms in the 1980s, such as multiparty elections, it reportedly began redirecting its resources toward buying votes to ensure that its candidates won at the polls. (74)
Black-gold politics, however, became controversial when the media reported that the KMT used gangsters to buy votes. While the KMT denied these charges, Li Ao, 2000 presidential candidate of the NP, describes how it worked: "They would give you money in one hand and tell you to vote for so and so." (75) This issue reached a new level of controversy when some of the gangsters decided to stop functioning as intermediaries and instead run for office themselves under the KMT party label. According to Eric Chu Liluan, a KMT reformist legislator and official campaign spokesperson for Lien Chan (the KMT's 2000 presidential candidate), at least 10 percent of the Legislative Yuan and 25 percent of the local assemblies are linked to gangs. Whether or not corrupt practices really reached this extent, it is clear that there existed at least some relationship between the KMT and organized crime. Public perceptions of that relationship, in turn, appear to be accurately reflected in the preceding quotations. (76)
Additionally, not all the money for buying votes came from party funds. According to a former legislator, corruption spread once a "corrupt" politician reached office:
The party will contribute only in key races, forcing many candidates to pay out of their own pockets. This leads to further corrupt behavior once the politician wins, as they may want to recoup their money by taking cuts in public works projects in their constituencies. In fact, many such projects are subcontracted out by companies linked to lawmakers. (77)
How did black-gold politics affect corrupt behavior in the bureaucracy? As noted, the KMT's assets are in the form of stocks in private companies. Hence, if the bureaucracy gives favorable regulatory treatment to companies in which the KMT is a prominent shareholder, the party benefits directly, as the capital valuation of those companies rises (i.e., their stocks go up), thereby increasing the party's net profits.
In addition, the bureaucracy is likely to award government contracts for regulatory projects, such as public works, to companies linked to the KMT, thereby further enhancing the profitability of such firms. Cal Clark (78) makes this point clear:
When Lee Teng-hui and his young Turks challenged the KMT old guard, they turned to the support of rich business people and local patronage-oriented political factions.... The burgeoning political corruption offended general public sensibilities in three major areas. First, many contracts for public projects became highly politicized. Corrupt legislators would form their own companies to bid on contracts and, in many instances, would form ties to organized crime who, in turn, would deter bids by legitimate businesses. Needless to say, quality suffered greatly, and public cynicism about government projects rose. Second, several prominent business leaders and conglomerates blatantly used political connections and legislative positions to help their corporations and influence the stock market, leading to a seemingly endless series of major scandals in the early and mid-1990s. Finally, local political factions, which had always been patronage-oriented, became increasingly blatant and tied to organized crime.
The end result was shoddy construction and delays in major public transportation projects such as Taipei's subway line, the world's most expensive, costing $13 billion and finished two years late. (79) Therefore, the dominant faction of the KMT and the bureaucracy had a powerful incentive to preserve the status quo and oppose policies that would break the existing method of delivering public works.
Managing Coalition Politics
As the previous discussion suggests, the increasingly youthful legislature was becoming more reform oriented and, hence, more hostile to Lee's less reform-minded faction of the KMT. The provision of the 1997 constitutional reform lowering the no-confidence vote threshold from two-thirds to a simple majority, in turn, may have provided additional incentive to Lee to support the reformers as well as to "buy off" the legislature in order to reduce the likelihood of a no-confidence vote. TAPA's timing is certainly suggestive, albeit by no means definitive, evidence that the 1997 amendment influenced Lee's decision to accept administrative reform. Along these lines, Professor Huang Jiashu, a senior analyst at the Taiwan, Hong Kong and Macau Research Center of the People's University in Beijing, commented in 1998, "The KMT is a union of interests.... Lee can no longer dictate. He must play coalition politics along with others." (80)
As Professor Yeh Jiunn-rong of National Taiwan University explains, the young KMT reformists and mainstream members' positions regarding the role of the APA was clear:
The young KMT reformists believed that by passing an APA, the bureaucracy would open up and effectively enfranchise new groups. But the mainstream KMT members and bureaucrats opposed the rule-making provisions. The KMT wants things to stay the same. In terms of the decision-making process, the KMT has more discretion under the status quo but has less after the implementation of TAPA. There was no opposition from the DPP or NP on rule-making provisions. Not at all. (81)
KMT nonmainstream faction leader James Soong implicitly acknowledged the potential problem of bureaucratic opposition to TAPA in the following comment, offered several months prior to TAPA's passage: "The country cannot afford to see further internal squabbling over government restructuring.... Government restructuring should ... be carried out in a gradual manner in order to reduce the backlash from public functionaries." (82)
Managing Intrabranch Delegation with an APA
Why an APA? The APA, in one stroke, broke the status quo deal making in the bureaucracy between the old wing of the KMT, entrenched bureaucrats, and big businesses (including organized crime bosses). The APA forced bureaucrats to change their behavior by enfranchising consumer and environmental groups in the regulatory decisionmaking process such that bureaucrats had to take into account their opinions about every regulation. It also shined light on the government contracting process, making it harder for bureaucrats to cut deals with crime bosses in back rooms.
The "winners" were consumers, environmental groups, and export-oriented or "internationalist" businesses that benefit from an improved climate for foreign investors, who would be attracted by a cleaner political process in Taiwan. An example of the latter group is the so-called Evergreen Group--an internationally oriented transportation conglomerate-who favored reform, including an APA, because increased transparency in the government would facilitate increased foreign investment, thereby helping their bottom line. Hence, not surprisingly, the Evergreen Group was "a major financial backer of Chen's [from the DPP] political career." (83) The losers under the APA system, in turn, were the building contractors and other public works business contractors--some of whom had ties to organized crime--who were cozy with the old-school KMT members.
KMT reformers wanted to control the bureaucracy, through an APA, in order to rob the DPP of its primary issue, thereby effectively stealing the rival party's political oxygen. For instance, in order to win support among youthful, proenvironmental voters, the KMT reformists sought to thwart a major nuclear reactor project:
The most celebrated problem in the late 1990s was an attempt by the opposition and dissenting KMT deputies to block funding for Taiwan's fourth nuclear power plant, located at Kungliao. The project was stalled in parliament in May 1996, just ahead of the official announcement.... The episode also illustrated the potency of political and environmental opposition to key projects. (84)
The KMT refomers recognized that unless they undertook far-reaching measures to change the KMT's image, as well as public policy outcomes, the party was likely to lose power. Reining in the entire government bureaucracy represented a relatively drastic measure, designed to reform the KMT's image by producing different policy outcomes in a broad range of policy areas, particularly with respect to environmental protection and public infrastructure. The reformists hoped that such changes would effectively undercut the DPP's primary source of political strength. An APA, by improving executive control over the bureaucracy, helped to accomplish this goal. Because the regulation of public works projects was so poor, voters were concerned about public safety and the quality of the nation's infrastructure. And, like their contemporaries in other advanced industrial countries, young Taiwanese voters were wary of nuclear power and environmental degradation and were generally concerned with improving their quality of life. Essentially, the APA would undo the cozy relations between bureaucrats and contractors that had become the source of many of the KMT's political problems.
In sum, for the DPP, NR and KMT reformists, the APA would provide the necessary tools for reducing the level of corrupt practices of the KMT and the bureaucracy. By instituting such procedures as the public notice and comment process and judicial review of agency decisions, bureaucrats could be held accountable for their decisions. Under the notice and comment process, bureaucrats must take into account opinions of new groups (e.g., environmental advocates and consumer groups concerned with public health and safety--issues that younger voters cared about) previously excluded from the KMT's decisionmaking process.
Under the previous system, bureaucrats were not required to offer public notice or solicit public comment before they could write regulations. They were less likely to consider opinions of, say, environmental groups who might have opposed a project. Instead, the bureaucrats typically wrote regulations giving favorable treatment to the KMT-linked public works companies that were likely to receive the contracts, while paying less attention to environmental and other public interest-oriented groups. Hence, absent the APA's notice and comment provision, bureaucrats were less inclined to enforce meaningful environmental regulations on public works projects.
Why Lee supported TAPA. The question of why Lee went along with the APA begs the question of whether the lock-in explanation applies to the Taiwan case. The lock-in argument would predict an APA under Lee's first administration. This is because Lee faced the possibility of losing Taiwan's first popular election, in which he ran for president, as his party's electoral dominance had declined to an all-time low in 1995 (see Figure 1). Hence, the logic of the lock-in hypothesis would hold that a vulnerable president should seek to lock in his favored policies, through an APA, before losing office. Yet an APA did not pass.
During the second administration, in contrast, the lock-in argument would not predict an APA. This is because the KMT expected to win the next presidential election, especially after increasing its legislative seat share in 1998 (see Figure 1). In fact, almost none of the officials I interviewed believed the KMT would lose the 2000 presidential election. (85) Yet, in this instance, an APA did pass. Taken together, these two cases suggest that the lock-in hypothesis cannot explain Taiwan's APA passage. Instead, I argue that Lee went along with an APA in order to keep the KMT from losing more young voters.
In numerous interviews with scholars; legislators from the KMT, DPP, and NP; and senior bureaucrats, none believed the KMT thought to use TAPA to lock in the status quo. (86) Quite the contrary, the APA facilitated changing status quo bureaucratic practices by enfranchising new groups previously excluded from the KMT's decisionmaking process. For example, when I asked him how he would characterize the enfranchisement of different interest groups besides big business as a result of TAPA, Cheng Pao Ching, legislator and convener of the DPP caucus, replied:
Because we do not have a formal lobbying system, we see equal opportunity for all groups now. As a matter of fact, the DPP used to maintain better relations with social groups such as environment or other "weaker" groups. In terms of new relationships, I do not see much difficulty for these groups in having access to agency policy making.
In short, by enfranchising new groups, TAPA's supporters, including President Lee, clearly intended to facilitate altering, rather than preserving, the policy status quo.
In this article, I have sought to demonstrate how the passage of an APA in Taiwan is consistent with a current control explanation and inconsistent with lock-in. Democratic transition certainly pushed politicians in the general direction of political reform, but it does not explain why they supported the passage of an APA per se.
I have argued that TAPA passed during Lee's second administration because the KMT reformists wanted to change Taiwan's status quo policies regarding environmental protection and public infrastructure. To do so, they needed to break the corrupt relationship between entrenched bureaucrats and big businesses. An APA helped accomplish this latter goal.
Table 5 summarizes my argument, showing which of the key conditions for APA passage existed during each period in Taiwan. In each case, I note whether the president had problems with cabinet ministers or bureaucrats, and whether bureaucrats enjoyed civil service protection. The table also compares the actual outcome with the logical prediction if the goal of President Lee had been to "lock in" status quo policies at each point in time. (87)
These data support my argument that Taiwan's APA enactment resulted from the president's concern about reining in bureaucrats who could not be fired. During Lee's first administration, neither of the key conditions was present. So, even though the bureaucracy enjoyed civil service protection, Lee did not need to worry about constraining disloyal underlings. In Lee's second administration, however, the KMT reformers sought to reverse the KMT's declining electoral dominance by competing with the DPP for the increasingly critical bloc of young, native-born voters. Reining in the bureaucracy was thus a means of preventing the opposition DPP from replacing the KMT as the country's dominant political party, rather than a tool for locking in the status quo.
To win these voters, the KMT reformers advocated the same policies that led to the DPP's rise and popularity--clean government, environmental protection, and higher-quality (i.e., safer) public infrastructure. Their goal was to remake the party's image from the party of corruption and favoritism toward private interests to the party of "clean government" representing "all of the people." An APA was a tool for managing current delegation to the bureaucracy in order to break the status quo--consisting of cozy regulatory relationships between bureaucrats, wealthy business interests, and, in some instances, organized crime bosses--that had resulted in unsafe public construction and lax environmental protections. By breaking these ties, KMT reformers wanted to make government policy more proenvironment and more responsive to public health and safety concerns. It did so by forcing bureaucrats to change their behavior to take into account the views of younger voters and their representatives (e.g., consumer and environmental groups) prior to implementing policies. President Lee did not necessarily "prefer" to rein in the bureaucrats, who were, after all, his political allies. But he did so nonetheless because he prioritized maintaining KMT power over maintaining the relative free hand that his bureaucratic allies enjoyed under the pre-TAPA status quo. In other words, for electoral reasons, he concluded that bureaucratic discretion was a necessary sacrifice to preserve the KMT's political dominance in Taiwan.
Taiwan is an important case for my theory for two primary reasons. First, Taiwan experienced an episode in which a president had similar preferences as his bureaucracy and one in which he did not. Consequently, this case provides variation on the dependent variable. Second, Taiwan's transition to a president-parliamentary system made it a useful case to further show how APA enactment is consistent with a reining-in explanation and inconsistent with lock-in. (88)
My case study of Taiwan, combined with recent work on the passage of Korea's APA, (89) suggests that the source of institutional barriers to policy change can vary across institutional and political settings. One implication is that despite institutional variations across both presidential and parliamentary systems, as long as the source of institutional barriers gives rise to intrabranch conflict, similar principal-agent problems may arise and similar solutions may be pursued. I have argued that politicians in new democracies can resort to APAs as one solution to this delegation problem.
Portions of this research were funded by the National Science Foundation and the UCLA International Studies and Overseas Program. I wish to thank the following individuals for their valuable comments and suggestions: Matthew Baum, Richard Baum, Kathleen Bawn, Gary Cox, J. R. DeShazo, Stephan Haggard, John Londregan, Mathew McCubbins, Michael Thies, and two anonymous referees. I am particularly grateful to Nathan Batto for his valuable insights into Taiwanese politics. I also wish to thank Tsai-wei Sun for research assistance and Carl Chiaohao Chang, Jih-wen Lin, Michael T. W. Lin, Dennis T. C. Tang, and Roy Hsiang-jung Wu for facilitating my fieldwork.
(1.) Interview with author, Taipei, Taiwan, July 15, 2000.
(2.) McNollgast, "The Political Origins of the Administrative Procedure Act," Journal of Law, Economics, and Organization 15, no. 1 (1999): 180-217.
(3.) Jeeyang Rhee Baum, "Presidents Have Problems Too: The Logic of Intrabranch Delegation in East Asian Democracies," British Journal of Political Science, forthcoming.
(4.) It may appear that my theory contains an implicit assumption that presidents choose APAs in part because they enfranchise "the people" and so are likely to be popular. However, this follows only if a president cares about the previously excluded interest groups that an APA enfranchises. Indeed, important support groups for a president may represent a fairly broad spectrum of voters. Consequently, an APA may be popular, even if a president's motivation is not to enhance his or her popularity by empowering the people.
(5.) McNollgast, "Political Origins."
(6.) Jeffrey S. Banks, "Agency Budgets, Cost Information, and Auditing," American Journal of Political Science 33 (1989): 670-699.
(7.) Mathew McCubbins and Thomas Schwartz, "Congressional Oversight Overlooked: Police Patrols Versus Fire Alarms," American Journal of Political Science 28 (1984): 167-179; Arthur Lupia and Mathew McCubbins, "Learning from Oversight: Fire Alarms and Police Patrols Reconsidered," Journal of Law, Economics, and Organization 10 (1994): 96-125.
(8.) Barry Weingast and Mark Moran, "Bureaucratic Discretion or Congressional Control? Regulatory Policy by the Federal Trade Commission," Journal of Political Economy 91 (1983): 765-800.
(9.) Randall Calvert, Mathew McCubbins, and Barry Weingast, "A Theory of Political Control and Agency Discretion," American Journal of Political Science 33 (1989): 588-611; David Epstein and Sharyn O'Halloran, Delegating Powers (Cambridge: Cambridge University Press, 1999); Kathleen Bawn, "Political Control Versus Expertise: Congressional Choices About Administrative Procedures," American Political Science Review 89 (1995): 62-73. Kathleen Bawn, "Choosing Strategies to Control the Bureaucracy: Statutory Constraints, Oversight, and the Committee System," Journal of Law, Economics, and Organization 13 (1997): 101-126.
(10.) Mathew McCubbins, Roger Noll, and Barry Weingast, "Administrative Procedures as Instruments of Political Control," Journal of Law, Economics, and Organization 3 (1987): 243-277.
(11.) See for example, Terry Moe, "The Politicized Presidency." In John E. Chubb and Paul E. Peterson, eds., The New Direction in American Politics (Washington, D.C.: Brookings Institution, 1985).
(12.) For a related study on how ministerial instability affects bureaucratic behavior in parliamentary delegation, see John D. Huber and Arthur Lupia, "Cabinet Instability and Delegation in Parliamentary Democracies," American Journal of Political Science 45, no. 1 (January 2001): 18-32.
(13.) Partners in a coalition can either be from multiple parties or from factions within a single party.
(14.) McNollgast, 1999.
(15.) Matthew S. Shugart and John M. Carey, Presidents and Assemblies: Constitutional Design and Electoral Dynamics (Cambridge: Cambridge University Press, 1992).
(16.) Michael F. Thies, "Keeping Tabs on Partners: The Logic of Delegation in Coalition Governments," American Journal of Political Science 45, no. 3 (July 2001): 580-598.
(17.) The inability of presidents to fire bureaucrats might not stop them from reassigning unresponsive bureaucrats to outposts where they can do less harm. Yet, this tactic is of limited value to presidents, as it is impractical to reassign large numbers of bureaucrats (beyond political appointees).
(18.) In 1974, the Committee of Research, Development and Assessment of the Executive Yuan (branch) appointed Professor Lin Chi-Tung of National Taiwan University to chair a committee charged with conducting a comparative study of APAs in the United States, Japan (1964 draft), Germany (1966 draft), Austria, Italy (draft), and Spain (Kuo-Tsung Huang, "Introduction of the APA," Journal for the Lawyers [April 1999]: 49-51). Although this academic study never materialized into law at the time, it became the theoretical foundation for TAPA.
(19.) Interview with author, Taipei, Taiwan, July 17, 2000.
(20.) Mathew McCubbins, Roger Noll, and Barry Weingast, "Administrative Procedures as Instruments of Political Control," Journal of Law, Economics, and Organization 3 (1987): 259.
(21.) Tun-jen Cheng and Stephan Haggard, "Democracy and Deficits in Taiwan: The Politics of Fiscal Policy 1986-1996." In Stephan Haggard and Mathew D. McCubbins, eds., Presidents, Parliaments, and Policy (Cambridge: Cambridge University Press, 2001), p. 200.
(22.) Steven J. Hood, The Kuomintang and the Democratization of Taiwan (Boulder: Westview Press, 1997), p. 108.
(23.) Ibid., emphasis added.
(24.) Ibid., p. 144; Cheng and Haggard, "Democracy and Deficits in Taiwan," p. 200. The NP also supported strong business ties with mainland China in order to facilitate reunification and rejected a proindependence policy.
(25.) Cheng and Haggard, "Democracy and Deficits in Taiwan," p. 201.
(26.) Yu-Shan Wu, "Comparing Semi-Presidentialism in the ROC and the Russian Federation," Chinese Political Science Review (December 1998): 123-186.
(27.) The R.O.C. Election Study Center, National Chengchi University. Results of recent elections available at http://www2.nccu.edu.tw/%7Es00/eng/ data/data02.htm.
(28.) Larry Diamond, "The Anatomy of an Electoral Earthquake: How the KMT Lost and the DPP Won the 2000 Presidential Election." In Muthiah Alagappa, ed., Taiwan's Presidential Politics: Democratization and Cross-Strait Relations in the Twenty-first Century (Armonk, N.Y.: M. E. Sharpe, 2001), p. 54.
(29.) "Soong Laments Downsizing," Asia Intelligence Wire, September 2, 1998; Henry Chu, "Taiwan Youth Drifting Toward Independence," Los Angeles Times, March 12, 2000.
(30.) Dexter Roberts, Stuart Young, Mark L. Clifford, and Stan Crock, "The Reformer in Charge," Business Week, April 3, 2000.
(31.) Tun-jen Cheng and Yung-han Chu, "State-Business Relations in South Korea and Taiwan." In Laurence Whitehead, ed., Emerging Market Democracies: East Asia and Latin America (Baltimore: Johns Hopkins University Press, 2002), p. 48.
(32.) Based on 1991-1996 survey results from Fu Hu and Yun-han Chu at National Taiwan University.
(33.) Cal Clark, "Taiwan Elections," Asia Society (March 2000), available at http:www.asiasociety.org/publications/taiwan_elections.html; James Conachy, "Corruption and China Policy Dominate Taiwan Presidential Campaign," World Socialist, January 14, 2000, available at http://www.wsws.org/articles/ 2000/jan2000/taiw-j14.shtml; Shih-Chung Liu, "The Real Potential for Political Decay," Taipei Times, January 23, 2000, available at http://www.taipeitimes. com/news/2000/01/23/story/0000021197; Sofia Wu, "Taiwan Goes to Polls Amid Sense of Uncertainty," Central News Agency, March 17, 2000. Not surprisingly, the other primary public concern registered in most opinion polls is the question of reunification with China.
(34.) On July 16, 1999, James Soong, a rebel Nationalist ruling party leader, declared his candidacy for the 2000 presidential elections ("Rebel Leader Soong Announces Candidacy for 2000 Elections," Dow Jones Newswires, July 16, 1999). Soong was expelled from the KMT in November 1999 (EIU Country Profile: Taiwan, 1999-2000, London, Economist Intelligence Unit, p. 7) and formed the People's First Party (PFP).
(35.) Clark, "Taiwan Elections."
(36.) Roberts et al., "The Reformer in Charge."
(37.) Henry Chu, "New Era in Taiwan," Los Angeles Times, March 19, 2000, p. 2.
(38.) "Tougher Tests Face Chen Shui-bian After 'Dream' Win," Deutsche Presse-Agentur, March 18, 2000.
(39.) Hilfe Country Report, "Taiwan Politics," Janet Matthews Information Services, January 20, 2000, p. 4.
(40.) Sangwon Sub, Alejandro Reyes, and Allen T. Cheng, "Seismic Changes," Asiaweek, March 31, 2000.
(41.) Ko-lin Chin, Heijin: Organized Crime, Business, and Politics in Taiwan (Armonk, N.Y.: M. E. Sharpe, 2003), p. 5.
(42.) Roberts et al., "The Reformer in Charge."
(43.) Ibid., p. 2.
(44.) Tony Frassrand, "Newsroom Worldview," CNN transcript no. 98042700V05, April 27, 1998.
(45.) Laurence Eyton, "A Lesson in Democracy." Asiaweek, December 12, 1997; James Flanigan, "Asia's Improbable New Strongman," Los Angeles Times, April 26, 1998; Susanne Ganz, "Chen Shui-bian Announces Bid for Taiwan Presidency," Kyodo News Service, Japan Economic Newswire, May 27, 1999.
(46.) Ganz, "Chen Shui-bian Announces Bid."
(47.) "From Poverty to Power: Chen's Rise to the Presidency," Financial Times Information, November 9, 2000.
(48.) Hilfe Country Report, "Taiwan Politics."
(49.) Ironically, however, James Soong rose to become a formidable contender during the 2000 presidential election precisely because be was able to use Taiwan governorship, and the fiscal and financial resources under the control of his office, to nurture an extensive network with local factions, most of them key players in black-gold politics.
(50.) "Taiwan's 50-Year Regime Senses Electoral Defeat," Times (London), March 16, 2000.
(51.) Allen T. Cheng, "The Curse of 'Black Gold': Lee Teng-hui's Darker Legacy May Be a Political Culture Riddled with Corruption and Gangsters," Asiaweek, March 17, 2000, available at http://www.asiaweek.com/asiaweek/ magazine/2000/0317/nat.4taiwan.kmt.html 26(10).
(52.) "Lin Feng-Cheng Becomes KMT Secretary General," quoted in Central News Agency, March 27, 2000.
(53.) Ko-lin Chin, Heijin, pp. 7-8.
(55.) Chu, "New Era in Taiwan."
(58.) Ibid.; Sofia Wu, "Taiwan Goes to Polls Amid Sense of Uncertainty," Central News Agency, March 17, 2000.
(59.) Wu, "Taiwan Goes to Polls."
(60.) Roberts et al., "The Reformer in Charge," p. 5.
(62.) "Counting the Cost of Corruption," Janet Matthews Information Services, April 1, 1998.
(63.) Asia Intelligence Wire, "Soong Laments Downsizing"; Hilfe Country Report, "Taiwan Politics," Janet Matthews Information Services, July 17, 2000, available at http://www.agora.stm.it/elections/election/taiwan.htm; Suh et al., "Seismic Changes."
(64.) Asia Intelligence Wire, "Soong Laments Downsizing."
(65.) Flanigan, "Asia's Improbable New Strongman," p. 3.
(66.) I code as mainlanders nine cabinet ministers who claim they are from Fu-Chien province. While these individuals consider themselves to be mainlanders, other Taiwanese view them as natives of Taiwan. Since, however, the most pertinent issue for my purposes is self-identification, I have elected to classify these ministers according to their self-identification. A second test, recoding these individuals as native Taiwanese, had no material effect on the results.
(67.) Cheng, "The Curse of 'Black Gold'"; Chu, "Taiwan Youth Drifting Toward Independence."
(68.) Tsai Shi-yuan, A Complete File of Black-Gold (Taipei: Tsai Shi-yuan Research Office in the Legislature, 1998), quoted in Ko-lin Chin, Heijin, p. 5.
(69.) Roberts et al., "The Reformer in Charge."
(70.) Quoted in ibid.
(71.) Ko-lin Chin, Heijin, p. 128.
(72.) Cheng, "The Curse of 'Black Gold.'"
(75.) Quoted in ibid.
(76.) Oliver August, "Taiwan's 50-Year Regime Senses Electoral Defeat," Times (London), March 16, 2000; Chu, "Taiwan Youth Drifting Toward Independence"; Ganz, "Chen Shui-bian Announces Bid for Taiwan Presidency"; Wu, "Taiwan Goes to Polls"; "DPP Launches New Campaign TV Ad Focused on Young People," Financial Times Information, February 10, 2000; Tim Healy and Laurence Eyton, "Old Party, New Face," Asiaweek, December 18, 1998; Hilfe Country Report, "Taiwan Politics," Janet Matthews Information Services, July 17, 2000, available at http://www.agora.stm.it/elections/election/ taiwan.htm; Roberts et al., "The Reformer in Charge."
(77.) Quoted in Cheng, "The Curse of 'Black Gold.'"
(78.) Clark, "Taiwan Elections."
(79.) Roberts et al., "The Reformer in Charge."
(80.) Healy and Eyton, "Old Party, New Face," p. 22.
(81.) Interview with author, Taipei, Taiwan, July 17, 2000.
(82.) Asia Intelligence Wire, "Soong Laments Downsizing."
(83.) "DPP Launches New Campaign," p. 2.
(84.) Hilfe Country Report, "Taiwan Politics."
(85.) Interviews conducted in Taipei, Taiwan, July 2000.
(86.) Interviews conducted in Taipei, Taiwan, July 2000.
(87.) The evidence in support of Table 5 is derived from both archival research and interviews conducted during fieldwork in Taipei, Taiwan, during summer 2000.
(88.) I consider Taiwan a president-parliamentary system after the 1997 constitutional amendments. A president-parliamentary system is defined as follows: (1) the president is popularly elected; (2) the president appoints and dismisses cabinet ministers; (3) cabinet ministers are subject to parliamentary confidence; and (4) the president has the power to dissolve parliament or legislative powers, or both (Shugart and Carey, "Presidents and Assemblies," p. 24).
(89.) Baum, "Presidents Have Problems Too."
Jeeyang Rhee Baum is assistant professor, Department of Political Science, University of California, San Diego. Her current research focuses on delegation and accountability in new democracies, particularly in East Asia.
Table 1 The 1999 Taiwan Administrative Procedure Act Procedural Requirement Political Impact * Requires public notice and * Since agencies must consider comment period for proposed relevant political interests rules and agency reorganizations prior to making a new decision, (Articles 9, 11). sufficiently organized interest groups are more likely to * Requires agency investigation influence or veto agency's of evidence and facts regarding decision. appeals. Must record and notify relevant parties of decision * Procedural requirements have justifications (Articles 36-43). the benefit of causing the most controversial issues to generate * Allows public access to the most complete information. agencies' information (except national security matters) * Since process is public, we are (Articles 44-46). less likely to observe decisions made between agencies and their * Agencies must notify or announce favorite constituencies that in a government document (or conflict with preferences of any other method) the relevant president and legislature. Slows parties of a public hearing down agency decisionmaking well in advance. Notice should process and gives principals include: purpose, names of multiple opportunities to participants, date and location, respond if agency deviates and procedure of hearing from their policy preferences. (Articles 54-56). * Chairman of public hearing should be chief executive of agency or any other appointee. Procedures allow relevant parties to state opinion, evidence, and with chairman's permission, ask for testimony by witness, investigator, and relevant parties. Relevant parties can claim dissent to chairman's decision. Chairman can override dissent. Entire process must be recorded (Articles 60-66). * Entitles any party whose "rights or legal interests are infringed upon through an unlawful administrative act by a central or local government agency" to "initiate an administrative action ... in the high administrative court" (Article 4). Delegates review of administrative actions to courts (Articles 13-18). Sources: Administrative Litigation Act of Taiwan, 1998; Administrative Procedure Act of Taiwan, 1999; Mathew McCubbins, Roger Noll, and Barry Weingast. "Administrative Procedures as Instruments of Political Control," Journal of Law, Economics, and Organization 3 (1987): 258-259. Note: The 1998 Administrative Litigation Act (ALA) authorizes judicial review of administrative actions. According to former legislator John K. C. Huang, the "ALA follows the German Administrative Court Act and is not intended to include TAPA" (e-mail exchange with author, March 12, 2001). Table 2 Institutional Changes, 1990-2000 Relevant Dates KMT/Legislative Branch Lee Teng-hui/Executive Branch June 1990 Long-term parliamentarians' tenure ends as of December 1991; elections of three reorganized representative bodies in 1991 and 1992. July 1995 Legislative Yuan approves first presidential election on March 23, 1996. July 1997 Authority given to appoint the premier without consent of Legislative Yuan. July 1997 Legislative Yuan's vote of Can dissolve Legislative no confidence requirement Yuan within 10 days reduced from two-thirds to following vote of no 50 percent plus one. confidence. July 1997 Recall of president or vice- president requires motion of one-fourth and two-thirds of National Assembly. Must pass national referendum by 50 percent plus one. July 1997 Presidential and vice- presidential terms limited to one consecutive four-year term. July 1997 Effective 2003, National Assembly has authority to confirm presidential appointment of grand justices (life to one eight-year term). Source: Constitution of the Republic of China. Table 3 Money Machine: Kuomintang Equity Investments (in U.S. dollars) Year Total Net Profit 1993 $712 million $173 million 1994 $1.36 billion $307 million 1995 $1.4 billion $240 million 1996 $1.6 billion $286 million 1997 $2.0 billion $542 million 1998 $2.23 billion $395 million 1999 $2.29 billion $261 million Source: Allen T. Cheng, "The Curse of 'Black Gold': Lee Teng-hui's Darker Legacy May Be a Political Culture Riddled with Corruption and Gangsters," Asiaweek, March 17, 2000, available at http://www.asiaweek.com/asiaweek/ magazine/2000/0317/nat.4taiwan.kmt.html 26(10). Note: Investments do not include real estate and other fixed assets owned by the party. Table 4 Big Seven KMT Holding Companies * Central Investment Co. * Kwang Hua Investment Co. * Central Motion Pictures Corp. * Hua Hsia Investment Co. * China Broadcasting Co. * Central Daily News * China Daily News Source: Allen T. Cheng. "The Curse of 'Black Gold': Lee Teng-hui's Darker Legacy May Be a Political Culture Riddled with Corruption and Gangsters," Asiaweek, March 17, 2000, available at http://www.asiaweek.com/ asiaweek/magazine/2000/0317/nat.4taiwan.kmt.html 26(10). Table 5 Factors Affecting APA Enactment in Taiwan Presidential Problem Problem Civil Service Cases Ministers? Bureaucrats? Protection? Taiwan 1: No No Yes Lee, 1988-1996 Taiwan 2: No Yes Yes Lee, 1997-1999 Presidential Did APA "Lock-in" Cases Pass? Prediction Taiwan 1: No Yes Lee, 1988-1996 Taiwan 2: Yes No Lee, 1997-1999…