This project is co-funded by the
Seventh Frame work Programme for
Research and Technologic al
Development of the Europe an Union
EU Grant Agreement number: 290529
Project acronym: ANTICORRP
Project title: Anti-Corruption Policies Revisited
Work Package: WP3, Corruption and governance improvement in global
and continental perspectives
Title of deliverable: D3.2.13. Background paper on
Due date of deliverable: 28 February 2014
Actual submission date: 28 February 2014
Author: Christian Göbel
Editors: Thomas Richter and Sabrina Maaß
Organization name of lead beneficiary for this deliverable:
Hertie School of Governance
Project co-funded by the European Commission within the Seventh Framework
Restricted to other programme participants (including the Commission Services)
Restricted to a group specified by the consortium (including the Commission
Confidential, only for members of the consortium (including the Commission
Department of East Asian Studies
GIGA German Institute of Global and Area Studies
17 January 2014
Corruption has been on the top of Taiwan’s political and social agenda since at least the
early 1980s. In many opinion surveys over the years, people have named it the most
pressing political issue. Taiwan’s democratization in 1992 did not improve the situation -
some observers even argue that corruption has worsened because of the need to finance
election campaigns, to win votes and to gain influence in the now-powerful legislature.
Since the first change in ruling parties in 2000, the situation has gradually improved. The
Democratic Progressive Party (DPP) initiated tough anti-corruption regulations, strengthened
anti-corruption organizations and cracked down hard on corruption and organized crime. The
Kuomintang (KMT), which came to power again in 2008, continued this policy.
Several high-profile corruption scandals in the last years mask the fact that Taiwan’s
governance has improved markedly in the last decade. Not only have anti-corruption
regulations been passed and are rigorously enforced, but also anti-corruption units in the
government were strengthened. However, cultural factors such as the importance of
personal relations in Chinese society and the habit of giving gifts not only to friends, but also
to strategically important persons like doctors, teachers or business partners make it difficult
to completely root out corruption.
Corruption, Taiwan, Democratic Progressive Party, DPP, Kuomintang, KMT, democratization
Department of East Asian Studies
© 2014 GIGA German Institute of Global and Area Studies.
All rights reserved. This document has been published thanks to the support of the European
Union's Seventh Framework Programme for Research - Socio-economic Sciences and
Humanities theme (EU Grant Agreement number: 290529).
The information and views set out in this publication are those of the author(s) only and do
not reflect any collective opinion of the ANTICORRP consortium, nor do they reflect the
official opinion of the European Commission. Neither the European Commission nor any
person acting on behalf of the European Commission is responsible for the use which might
be made of the following information.
Taiwan: Marked Improvements in the Last Decade
AAC Agency Against Corruption
BOI Bureau of Investigation
BTI Bertelsmann Transformation Index
CPI (Transparency International) Corruption Perceptions Index
DPP Democratic Progressive Party (Minjindang)
KMT Kuomintang (Nationalist Party)
MOJ Ministry of Justice
NTD New Taiwan Dollars
TI Transparency International
WGI World Governance Indicators
Figure 1: Lawsuits filed with local prosecutors (number of cases and people involved) ......... 8
Figure 2: Percentage of local lawsuits involving certain status groups (primary axis, in %) and
amount involved (secondary axis, in billion NTD) ......................................................... 10
Figure 3: Guilty verdicts in asset declaration lawsuits (primary axis) and sums involved
(secondary axis) .......................................................................................................... 10
Figure 4: Convictions for Violating Conflict of Interest Avoidance Law ................................. 10
Figure 5: Number of administrative penalties imposed on public servants because of
corruption .................................................................................................................... 11
Table 1: Particularism and Ethical Universalism in Taiwan ...................................................17
Corruption has been on the top of Taiwan’s political and social agenda for the last three
decades. In public opinion surveys, it was named the most pressing problem even during
authoritarianism (Huang 1991), and the promise to eradicate corruption is the cornerstone of
nearly every political campaign (Fell 2002). Nevertheless, corruption remained entrenched in
Taiwan’s political system until 2000, when the Democratic Progressive Party (DPP) was able
to seize the presidency from the Kuomintang (KMT), who had ruled Taiwan uninterrupted
since 1947. The DPP victory might be called an accident: a rift in the KMT caused two of the
party’s heavyweights competing against each other, splitting the party’s vote pool. Owing to
the fact that a relative majority suffices to be elected president, Chen Shui-bian won the race
with just 39,3% of all votes. Fighting corruption had long been the most important issue in the
DPP’s campaign agenda (Fell 2002), and the new leadership delivered on its promise, not
the least because fighting corruption weakened the KMT and helped ensure the DPP’s
political survival (Göbel 2004). In the subsequent presidential election, the DPP
administration was rewarded for its performance, and Chen was able to defend his
presidency by a narrow margin of 50,11%.
Eight years later, however, the DPP lost power again, ironically amid allegations of severe
corruption against the presidential family. In 2010, former President Chen Shui-bian and his
wife were convicted of bribery and money laundering and sentenced to 20 years in prison
(BBC News 2009). A short time after that, three High Court judges were detained on charges
of having accepted bribes for clearing a former KMT legislator from corruption charges (Chao
2010). In April 2013, the former secretary general of the Executive Yuan was convicted for
corruption (AFP 2013). Besides these high-profile incidents, whistleblowers report dozens
and sometimes even hundreds of alleged corruption cases (MOJ, 2013d; MOJ, 2013e; MOJ,
2013f; MOJ 2013g). These cases encompass allegations of corruption in the judiciary, the
health sector, the police administration, and public procurement (MOJ 2013g). Corruption is
a major concern for nearly all actors in Taiwanese society (Zhongyang Yanjiuyuan Minzuxue
Yanjiusuo 1995; Yu et al. 2008).
Although the overwhelming majority of people in Taiwan are perhaps never personally
confronted with corruption, Taiwanese people believe that politics are ruled by “organized
crime and money” (heijin) (Chin 2003). However, many do perhaps not realize that their
expectations of politicians, which frequently include personal favors, obtaining pork for
organized interests, or even personal recognition by a famous person, are a fertile soil on
which corruption can bloom (Yang 1994; Kastner 2010). Rather than being dominated by a
corrupt class, their own values are a major reason that political corruption can continue.
Hence, in Taiwan there still exists a huge grey area between the customary cultivation of
personal relations by means of gifts and favors, and illegal practices such as bribery and
money laundering. The boundaries between the public and the private converge as one
moves down from the political center to the local administrations. In society, political
corruption is worst where people are uneducated and immobile. In contrast, educated
professionals in urban areas seem to have become resistant to the lures of clientelistic
politics (Göbel 2012).
Politicians seem to be bent on removing corruption by heavily fining bribery, embezzlement,
money laundering and other abuses of public office for private gain, and measures are taken
to sensitize the public to this gray area (Fawubu 2012). After all, as resolutely as nearly all
Taiwanese condemn corruption; many people are not aware that by accepting favors in
exchange for a vote, they are participating in the very act they condemn.
Recent policies cannot but be interpreted as serious efforts to turn Taiwan into a polity where
ethic universalism prevails over the particularistic values that underlie clientelistic politics. As
corruption is a hot topic in society, and as Taiwan’s political parties customarily accuse each
other of corruption, the pressure is huge to make true on their promise to clean up
Taiwanese politics. Supporters of both parties, who grew up in a political environment
dominated by practices that are today labeled “corruption”, skeptically observe, evaluate and
discuss the heavy-handed anti-corruption regulations that are formed, implemented and
Observers are justifiably skeptical: for the KMT, a party that had based its rule on
particularism during Taiwan’s long authoritarian spell, it is excruciatingly difficult to change
the institutions that had made it so resilient (Göbel 2004). The DPP, which had come to
power promising to rid Taiwan of corruption, but whose president provoked the largest
personal corruption scandal in the history of Taiwan, needs to work hard to regain its
Although Taiwan has still a long way to go, this report illustrates that no small feat was
achieved in the last decade: to develop from a polity where politics was based more on
informal relations than codified rules to one where the rule of law is becoming the norm. Two
interlinked factors are found to be instrumental: personal values and tough rules. The
Taiwanese example confirms that attitudes change only gradually, but “changes in
government can prompt a critical juncture in which a viable anti-corruption regime is forged”
(Göbel 2013). This is especially the case if the institutions that incentivize corrupt behavior
are fundamentally incompatible with the interests of the dominant political actors. The DPP,
when it came to power, was faced with “sophisticated, complex and deeply embedded
structures” of political corruption that benefited the KMT, but could not be altered to fit the
DPP. For the DPP, anti-corruption was a viable strategy for weakening the continuing
political, social and economic influence of the KMT. Although a viable anti-corruption legal
framework now exists, more time is needed until a truly open access order can be
established. Most importantly, there is still a blind spot in the minds of those people who
condemn corruption in principle, but at the same time cultivate personal relations to influence
the distribution of public resources in their favor. Establishing a truly open access order
requires people to resolve this paradox, which, given the importance of cultivating personal
relations in Chinese culture, might take years, if not decades.
II. Main Part
1. State of Governance: Better than the Worldwide Governance
Corruption is a fuzzy concept that denotes behavior which can, but does not need to be
illegal, and which can, but does not need to be morally questionable (Heidenheimer and
Johnston 2002). Much of what constitutes corrupt behavior takes place in a moral and legal
grey zone. This has two implications for the study at hand: first, corruption tends to be a
secret affair, which makes it difficult to obtain reliable data on the prevalence of corruption.
Second, people might judge as corruption behavior that is not illegal, and condone activities
that are illegal, but morally permissible according to prevailing cultural standards. In Taiwan,
these factors produce distorted images of how corrupt the country is. Although the
government has fought systemic corruption resolutely for more than a decade, perception-
based indicators of corruption have not improved much.
1.1 Perceptions of Corruption
The traditional corruption indicators reveal a contrasting image of corruption in Taiwan.
Perceptions-based indices of corruption suggest that Taiwan has largely stagnated in the last
decade. Taiwan’s CPI score leaped from 5.0 in 1996 to 5.9 in 2001, one year after Chen
Shui-bian became Taiwan’s president, but oscillated between 5.6 and 5.9 in the years
thereafter. When the Agency Against Corruption (AAC) was established in 2011, the score
increased to a historical high of 6.1 in 2011 (Transparency International 2011). Though
displaying the same trend, the oscillations in the World Bank Data are more marked.
Taiwan’s “control of corruption” score increased from 0.59 in 1996 to 0.86 in 2004, declined
to 0.48 in 2008, and increased to 0.90 between 2008 and 2011 (World Bank 2013). The data
indicates that respondents evaluated the effect of democratization and the first years of the
DPP positively, became disenchanted when allegations of corruption against Chen Shui-bian
surfaced, and supported the anti-corruption policies of the current Ma Ying-jeou
1.2 The Government´s Anti-corruption Record
Data on anti-corruption activities in the last decade stands in marked contrast to the
perception that Taiwan is trapped in a particularistic governance regime. For example, the
BTI country report 2013 states that corruption “is prosecuted rigorously under criminal law”
(Bertelsmann Foundation, 2013 (forthcoming), and official statistics on anti-corruption in the
public sector support this judgment. Taiwan’s Agency Against Corruption (AAC), which
began operating on July 20, 2011, provides rather detailed data on anti-corruption going
back to the year 1998. The data includes measures on the number and impact of activities
devoted to fighting and preventing corruption, especially in the ethics units of various
government departments and the AAC itself; figures on the violation of financial disclosure
and conflict of interest laws; statistics on complaints made by whistle-blowers; and data on
corruption indictments. The data is partly supplied by several ministries and government
departments, most notably the Ministry of Justice (MOJ), local prosecutors, the Control Yuan,
and partly generated by the AAC itself. As the AAC works, and in some instances overlaps
with, several bodies that had been tasked with fighting and preventing corruption in earlier
administration, it provides rather detailed data going back to the year 1998. Thereby, it allows
us to grasp the development of corruption over three different administrations between 1996
and 2013 (Lee Teng-hui, 1996-2000; Chen Shui-bian, 2000-2008; Ma Ying-jeou, 2008 until
October 2013, when this report was completed).
One indicator that illustrates the development and impact of Taiwan’s anti-corruption
programs well is the number of lawsuits filed with local prosecutors. As Figure 1 shows, the
number of cases, and the number of people involved in such cases, increased after the DPP
implemented its anti-corruption policies. Several of these cases involved a large number of
people, as the more sudden increase in indicted individuals between 2000 and 2001
illustrates. More marked, however, is the increase of the sums involved. This suggests that
the prosecutors singled out major cases. Interestingly, the increases and decreases in cases,
people and sums corresponds to electoral cycles: they pick up in 2000, the year Chen Shui-
bian was elected president, and wind down as his first term in office ends. They pick up
again as the Chen’s second term begins, but peak in 2008, his last year in office. The
numbers decrease as Ma Ying-jeou takes office, registering a slight increase in 2012, one
year after the AAC had been established. This finding suggests that anti-corruption policies
were an important element in electoral politics. They served to demonstrate resolve in the
fight against corruption and, perhaps more importantly, to weaken the KMT. As political
corruption was an important factor in the operation of the authoritarian KMT regime, and as
many civil servants remained in office after the change in government in 2000, a large
proportion of those affected by the DPP’s anti-corruption activities were officials of the old
regime. Data on the party affiliation of those targeted by the KMT’s anti-corruption drive since
2008 is unfortunately not available. In any case, the data vividly illustrate that anti-corruption
was conducted in the form of political campaigns at crucial moments in time, and more
research is necessary to establish which factors prompted the beginning and the end of such
Figure 1: Lawsuits filed with local prosecutors (number of cases and people involved)
Source: MOJ (2013a).
As can be seen in Figure 2, the years 2000-2004 saw an especially high %age of lawsuits
against civil servants and legislators, and 2001 stands out as the year when most officials
were sued: altogether, 1293 politicians were targeted, 122 of which from the highest grades,
and a staggering 706 from the lowest grades.
Another sweep targeting high- and medium level officials was conducted before and during
the presidential and legislative elections in 2007 and 2008, when nearly 500 high- and
medium level officials were sued. Finally, the figure also shows that although the number of
people involved in individual cases was high, the financial stakes were considerably lower
than during Chen’s first term, when anti-corruption work commenced. While the cases
involved an overall amount of 6 billion Yuan NTD in 2002, this sum lay below 2 billion Yuan
NTD after 2007. This result allows at least two interpretations: Chen Shui-bian’s anti-
corruption campaigns have contributed to reducing corruption in Taiwan, or the KMT was
much more lenient in its fight against corruption. The available evidence would support the
1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
first interpretation: many corrupt officials were removed, and a clean record became an
important precondition for entering public service.
Figure 2: Percentage of local lawsuits involving certain status groups (primary axis, in
%) and amount involved (secondary axis, in billion NTD)
Source: MOJ (2013a).
62 52 62 122 50 75 51 64 84 150 140 84 80 62 88
60 73 146 120
61 65 68 36 65 49 64 45 40 48 28
270 206 148
161 265 325 359
234 177 197 278
596 436 390 706
339 406 307
352 445 361 401 433 297 250 281
388 436 461 416
558 524 346 649 805 977 968 811 615 506 444
1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
Highest grades Legislators Middle grades
Lowest Grades Ordinary People Involved sum (bn Yuan)
Figure 3: Guilty verdicts in asset declaration lawsuits (primary axis) and sums
involved (secondary axis)
Source: MOJ (2013b).
Figure 3 highlights the number of cases where officials were declared guilty of deliberately
forging their asset declarations. Again, we see an increase-decrease cycle between 2000
and 2004, and a large increase in Chen Shui-bian’s second term, peaking at 412 cases and
28 million NTD of undeclared assets. Once more, the first years of the Ma administration
register a decrease, whereas the establishment of the AAC is followed by another increase.
Figure 4: Convictions for Violating Conflict of Interest Avoidance Law
Source: MOJ (2013c).
57 79 106
1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
Guilty total Total sum falsely declared (Mio. NTD)
7 4 4
13 14 12
31 31 29 28
0 3 3 3
8 5 8
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
Total Cases Guilty verdicts Involved amount (mio. NTD)
As for convictions for violating the conflict of interest avoidance law, the number is low during
Chen’s tenure, but increases steeply during the Ma administration (Figure 4).
Figure 5: Number of administrative penalties imposed on public servants because of
Source: MOJ, 2013h.
The same is true for administrative penalties imposed on public servants declared guilty of
corruption (Figure 5), which increase steeply under the Ma Ying-jeou administration.
1.3 Popular Attitudes towards Corruption
Much of the difference between popular perceptions of continued corruption and the
impressive anti-corruption track record of the present and the previous administration can be
explained by a lack of popular understanding of what constitutes corruption. For example,
people tend to read individual acts of high-level corruption as signs that the whole system is
fundamentally corrupt. Furthermore, the increase in indictments for corruption is interpreted
as an increase in systemic corruption, and not as a necessary measure to reduce corruption.
Finally, and perhaps most importantly, the fluid boundary between customs and corruption
allows contrasting explanations of the same behavior. That many people see no conflict in
nurturing personal networks to receive preferential treatment, but condemn those political
officials who grant such favors to others serves as a good example.
Data from the TI Global Corruption Barometer illustrate that perceptions are not a good
indicator of actual levels of corruption. For instance, more than 60% of those surveyed for the
same report find parliament, police and civil servants to be corrupt or extremely corrupt, only
seven % of all respondents in the 2010/2011 TI Global Corruption Barometer reported to
have paid a bribe (Transparency International 2011). Interestingly, the perceptions of
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
corruption are above, but the actual encounter of corruption in these institutions far below the
The fact that only a minority of cases reported by the general public qualifies for in-depth
investigation lays additional credibility to the assumption that corruption is perceived worse
than it actually is. Out of 1.632 allegations of corruption filed by members of the general
public in the second half of 2011, only 91, less than 6%, were substantial enough to qualify
for in-depth investigation (Fawubu, 2012). In Taiwan, high profile acts of corruption, such as
those committed by former president Chen Shui-bian, by the former general secretary of the
Executive Yuan Lin Yi-shih or by three High Court judges, are misinterpreted for endemic
and systemic corruption. This impression is reinforced by partisan politics and a highly
politicized media. The DPP and its supporters frame every instance of abuse of office by a
member of the KMT as evidence that the party is unable to change, whereas the KMT blows
up corruption that involves DPP office holders as proof that the DPP is even more corrupt
than the KMT. The media, which in Taiwan lean to one or the other side, amplify these
accusations and thereby reinforce the impression that Taiwan is hopelessly corrupt.
As the line between customs and corruption is difficult to draw, people display ambivalent
attitudes towards corruption in their personal lives. On the one hand, the general public is
very active in reporting suspected cases of corruption. Each year, thousands of cases are
reported by whistleblowers. In addition, people express their discontent in opinion surveys.
On a scale from zero to ten, 32.7% of all respondents in a survey mark that they have zero
tolerance for corruption. On the other hand, 22.3% of all respondents rate themselves slightly
over the median. People are also ambivalent in what they would do when faced with
corruption: in 2012, 62% of all respondents stated that they would be willing to report
corruption, but 31.8% said that they would not (Fawubu 2012).
In the general public, resentment is especially widespread about the assumed necessity to
give “red envelopes” containing money to teachers, doctors and sometimes justices to
influence outcomes. Such presents are seldom demanded by those who have the power to
give good grades, speed up procedures or save lives, but they are nevertheless customarily
given by parents and patients (Kastner 2010). Such bribery is rooted in customary behavior
that has existed for centuries. Although current anti-corruption policies not only severely
punish such behavior, but also educate the population, such changes take time to get
accepted. Customary bribery represents a classical prisoner’s dilemma: to continue giving
red envelopes represents the best strategy if one is not certain that this practice has
completely stopped by all actors.
Conversely, people do not consider themselves corrupt when they observe customs that are
an integral part of Chinese culture: nurturing personal networks and rewarding followers
(Yang, 1994). For a long time, many of these activities were neither forbidden by law, nor
deemed morally repulsive. As will be shown below, the expectation that political support
should be rewarded by money, a present or respect is still widespread especially among less
educated Taiwanese. The following sections illustrate how culture and political institutions
were interwoven in the KMT regime, and how the DPP, once it gained control of the
executive, tried to unravel this edifice.
2.1 Historical Roots
Taiwan’s governance regime is rooted in a particular arrangement that was fashioned after
the KMT forces had lost the civil war against the Chinese Communists in 1949 and fled to
Taiwan. Between 1945 and 1949, around 1.8 million soldiers, bureaucrats and other KMT
personnel moved to the island, which was already inhabited by a population of six million
people. Most of these people were Han Chinese, as were the immigrants. Apart from the
Han Chinese, half a million Japanese still lived in Taiwan, as did a minority of aboriginal
tribes that had inhabited Taiwan for thousands of years. Although the majority of the
“Taiwanese” were of the same ethnicity as the “Mainlanders” that arrived in the late 1940s,
relations between the two groups quickly soured. The slaughter of thousands of Taiwanese
by KMT forces on February 28, 1947 cemented a sub-ethnic cleavage between “Taiwanese”
and “Mainlanders” that exists to this day (Chu and Lin 1996). Hence, the émigré KMT
government had to assume that it was resented by at least 75% of the population. This put it
into the extraordinarily difficult situation of having to start its rule over Taiwan without a
tangible base of legitimacy.
Besides imposing Martial Law and ruthlessly cutting down on dissent, the KMT managed to
uphold stability by co-opting influential Taiwanese families into its regime. It did so by
opening local political offices to semi-competitive elections and providing incentives for local
elites to run as KMT candidates. In a policy of divide and conquer, the KMT made sure that
at least two families competed against each other in the same locality. Obtaining the
magistracy or mayoralty was very attractive, as the office came with tremendous financial
power, and allowed elites to skim additional profits from turning a blind eye to illegal
gambling or prostitution (Chen 1995; Chen and Chu 1992). Hence, the KMT leadership
tolerated the abuse of public office for private gain in exchange for political support by
powerful local families, which in turn mobilized various constituencies (Göbel 2012). As for
the governance regime that was formed at the time, it can be said that local corruption was
the linchpin that held the KMT regime in place.
It was not the scarcity of public funds that affected corruption in the early years of the KMT
autocracy, but the disproportionate influence of some individuals and groups at the county,
township and village levels. This is true despite the holding of local elections: elections were
semi-competitive and mainly served to integrate local elites into the regime, whose power
was augmented or reduced as a consequence of winning or losing an election. The system
was closed to outsiders, who seldom succeeded in obtaining political office.
2.2 Vote Buying and Clientelistic Local Politics
Individuals were tied into this exchange relation by receiving material and immaterial
gratifications for their political support of a particular candidate. Those who compete in local
elections establish support networks encompassing all administrative levels, various social
sectors, and local entrepreneurs (Göbel 2004). Rewards often take the form of paying for a
vote. Although vote-buying is prosecuted rigorously the practice persists; the assets of
legislators are monitored, and some legislators even lost their seat because they bought
votes. On the supply side, vote buying is difficult to discover. Vote buyers use increasingly
sophisticated means to hide their tracks, ranging from depositing untaxed revenue on the
accounts of willing collaborators to hiding bills in presents given to the voters. On the
demand side, many voters expect to receive a compensation for casting their vote, which is
often euphemistically referred to as “shoe money” (zoulufei) (Göbel 2012). Such rewards are
especially expected of KMT legislators, who find it difficult to distance themselves from
practices their party has long been associated with. As opposed to DPP legislators, who
used issue campaigning to win over voters, KMT legislators traditionally depended on long-
standing social networks to organize their votes. Personal contacts are as, if not more,
important than what a candidate stands for (Göbel 2004). As candidates explain, exchanging
money and other presents for a vote is not only a financial, but also an affective transaction,
and the thirty-odd dollars a vote can cost has as much symbolic as it has monetary value.
DPP candidates willing enough to engage in such an “organizational war” fight a losing battle
against a more superiorly organized KMT, but often see no alternative choice, because many
people with low levels of education do not want to be represented, but appreciated (Göbel
Of course, these descriptions do not represent a static situation, and they differ across
Taiwan. Younger and better educated people do not wish to be bothered by vote captains,
and cast their vote on issues rather than based on affection. This means that KMT
candidates are forced to change their strategies especially in urbanized and developed
regions (Göbel 2012). Conversely, DPP candidates have been known to engage in vote
buying as well in order to gain access to local influential networks in the more conservative
regions of Taiwan (Chin 2003). As a trend, clientelism is weakened as people become more
educated and leave their hometowns. It is no wonder then that the power of “local factions”
(difang paixi), as the well-organized local families are called, seems to wane. Within the next
decade, vote-buying might also become a thing of the past, and this process is reinforced by
the anti-corruption measures initiated by the DPP and continued by the KMT administration
As can be seen, improvements have been made, but corruption is difficult to root out in a
society where large sums are habitually invested in cultivating personal relations. The
institutional sources of corruption are gradually removed, but social norms and values are
much more difficult to change.
3.1 Tipping Points
The most important tipping point in Taiwan’s governance regime was the DPP’s ascend to
power in 2000. In many ways, this transfer of power completed the regime change that was
set in motion when Taiwan democratized between 1991 and 1992. Unlike most other
transitions to democracy, in Taiwan there was no convergence between regime change and
change in ruling parties. The KMT was so embedded in Taiwan’s economy and society that it
managed to survive democratization unscathed. It has remained the dominant force in
parliament until today, and scored a clear victory in the first direct presidential elections in
1996. Although the Taiwanese population deeply resented political corruption in Taiwan, and
although the eradication of corruption had been a prominent campaign topic for both parties
in these elections, the autocrat-turned-democrat regime could not afford to fight corruption.
Electoral campaigns were costly, as candidates had to rent campaign vehicles, host
banquets in urban and rural neighborhoods, distribute presents and buy votes. The fact that
the number of elections increased after democratization confronted political parties and their
candidates with additional financial burdens. Hence, politicians answered to public
dissatisfaction by vowing to fight corruption, but at the same time engaged in corrupt
practices to gain access to or keep political office.
Anti-corruption was the most important issue on the DPP’s campaign agenda when it won
the presidential elections in 2000. Although it seems fair to say that most of those who cast
their vote for the DPP supported this agenda, this was not enough for securing victory. Chen
Shui-bian and Annette Lu, his running mate, received only 39,3% of the popular vote.
However, the DPP profited from a split in the KMT. Soong Chu-yu, the well-connected former
Governor of Taiwan, ran as an independent in the 2000 presidential election when the KMT
refused to nominate him, capturing 36,8% of the vote. Had the KMT stayed united, they
might well have won the election by absolute majority. In any case, the DPP administration
lived up to its promise to fight corruption and organized crime and initiated an “action
program to sweep out black gold” under the auspices of the MOJ. A number of anti-
corruption laws were drafted and legal revisions to existing laws proposed. Given the strong
base of the KMT in personal patronage, “fighting corruption and changing the rules of the
game was not only a campaign promise, but necessary for the DPP’s political survival”
3.2 Institutional Changes
As for revisions, the Public Functionaries Election and Recall Law was to bar organized
criminals from running for political office in the future, the Civil Servant Services Act was to
be amended to forbid civil servants from accepting gifts or donations, and the Criminal Law
was to punish repeated offenses more severely than before. In addition, amendments to the
Public Functionary Assets Disclosure Law made the assets of government officials and their
immediate relatives transparent; the amended Money Laundering Control Act required
financial institutions to report large money movements and establish a citizens’ bank account
database; and the proposed revisions of the Anti-Corruption Penal Statue allowed the
authorities to presume corruption where an official’s wealth exceeded his income level and
confiscate funds the official cannot account for (Göbel 2004: 18).
The new laws that the DPP administration proposed and that eventually were implemented
included three Sunshine Laws, notably the Conflict of Interest Law, which forbids public
officials from using their position for private gain; the Political Contributions Act, which forbids
“quid pro quo exchanges”, sets ceilings for donations to political parties, and limits the sum of
anonymous donations that can be accepted; and the Lobby Law, which requires lobbyists to
apply for a lobbying permit with the Ministry of Interior and publicize lobbying activities (Göbel
2004: 23). These laws sparked considerable controversy, were resisted by KMT and
independent lawmakers, but nevertheless passed the legislature in 2001, 2004 and 2008,
In addition, the DPP caucus proposed a Political Party Law and a Statute Regarding the
Disposition of Assets Improperly Obtained by Political Parties, which targeted the ample
resource flows the KMT enjoyed from the state enterprises previously under its control
(Göbel 2004: 23). As can be expected, these laws were never implemented.
Other measures taken by the DPP administration were the proposed establishment of an
anti-corruption agency, which also failed due to KMT resistance, and efforts to abolish
sources of irregular revenue that legislators, mayors, county commissioners and local
factions tapped to finance their campaigns and reward their followers. Among others, highly
leveraged local financial institutions were closed down or merged, and access to funds better
regulated. Besides, the allocation of public expenditures was centralized. For example,
legislators had been able during the KMT era to “recommend” the financing of small-scale
construction projects to the Directorate General of Budget, Accounting and Statistics, which
in fact amounted to a sizeable pork barrel. Amid great protest, the DPP ended this
prerogative (Göbel 2004).
Interestingly, after recapturing the executive in 2008, the KMT implemented the anti-
corruption measures it had so vigorously opposed during its previous tenure, and it is
instructive to examine how this change of hearts came about. An important factor in this
outcome is the controversy around the heavy prison sentence for the former president.
Reacting to heavy accusations that this sentence was politically motivated, the KMT invested
significant resources to present itself as a party that had rid itself of is inglorious past and
was now serious about fighting corruption. This narrative was shattered when not only a
number of KMT legislators, but also three high court judges were convicted for corruption.
Support for the KMT plummeted just two years before the presidential and parliamentary
elections, and Ma Ying-jeou sought to restore his credibility by founding Agency Against
Corruption under the auspices of the MOJ in 2011 (Office of the President 2011) as well as
stepping up the implementation of the existing anti-corruption legislation.
4 Detailed Diagnosis
Table 1: Particularism and Ethical Universalism in Taiwan
High. Pluralism with evenly distributed and democratically legitimated
political influence. Only minor flaws in the system.
• Democracy since 1992: free and fair elections of representatives for legislature
and National Assembly; since 1996: change from indirect to direct presidential
• Diversified party landscape: four effective parties in two ideological camps: pro-
independence (DPP, TSU) and against independence (KMT, PFP). Many more
parties are registered, but have not obtained seats in the national parliament.
Political parties are stable and programmatic. Virtuall
y no party switching
occurs. But unequal access to party finance (The KMT is known as the richest
political party in the world), high cost of election campaigns.
Power changes have occurred in national (2000, 2008) and many local
executives as well as in local parliaments (since 1997). Uninterrupted KMT
majority in national legislature despite free and fair elections.
• Politicians are punished for corruption regardless of party affiliation.
• Central level: no oligarchic families. Subnational level: influence of local factions
in some counties.
• Power inequality is no more serious than in other developed democracies. No
access restrictions for public services, gatekeeping or monopolies.
Very active, but biased media that nevertheless guarantees a voice for
perceptions of particularism.
High. No state capture by interest groups. Political processes highly
• No state capture by interest groups. KMT enterprises were put into a trust,
overlaps between politics and business are now minimal. Officials have to
declare their assets publicly.
Administration and public sector characterized by a high degree of
professionalism. High degree of continuity in professional civil servant force.
• Bureaucracy merit-based, well-trained and sufficiently remunerated. Selection
criteria strict and impartial. Limited influence on policy formation and -
• Taiwan signatory to Government Procurement Act, state contracts awarded in
an open and transparent bidding process. Bid-rigging occurs (as in any other
country), but is penalized if discovered.
• A broad range of organizations monitor undue private influence: 1) the Control
Yuan, a branch of government that fulfills inspecting and auditing functions; 2)
ethics units within government departments; 3) the Bureau of Investigation of
the MOJ; 4) prosecutors imbued with a wide range of powers. Since 2011, this
work is concentrated in an Agency Against Corruption (de facto the Ethics
Bureau of the MOJ).
• The rules regulating lobbying and campaign financing are strict and enforced in
an impartial manner.
Impartial. Access to public services and goods is fair, highly equal and not
constrained by undue influence.
• Transfer payments centralized and allocated according to a fixed formula.
Nearly no funds available for pork barreling. No regions are favored, public
services are distributed impartially, and the quality of public services is high.
• With centralization of Small Construction Fund under the DPP government,
public revenue is not used as rents anymore. Public finances are very
transparent, and the auditing agencies are active in discovering and punishing
High. The use of public resources for private gain is conducted secretly. As
punishment is severe, the risk of embezzling public funds is very high.
• Public positions filled by means of elections or examinations. No inheritance of
positions. There are families where three or m
ore persons have influential
positions in central or local politics, but these persons are also elected.
• The abuse of public funds for private expenses is rigorously punished, public
scrutiny is thorough, and outrage over disclosures large.
• Nepotism is prohibited, information of relatives is disclosed.
Taiwan has a culture of informal practices that are in conflict with formal
institutions, but in the last decade, governments from both major parties
have made great strides in pushing back the importance of informal
• Dominant norm is now closer to formal institutions. Informality is still important at
local levels, but formal institutions take precedence.
• Court rulings are generally enforced, although the courts are often accused of
leaning towards the KMT.
• Gap between formal and informal rules has gradually narrowed since beginning
of DPP administration. Insider dealings have become infrequent.
and rule of law
High. Public accountability and the rule of law are strong and
• In the past ten years, several individuals belonging to chief status groups were
deposed or sentenced, including the former president, high level officials in
current administration, and local county commissioners, for example in Yunlin
and Taichung counties. Although the public is skeptical about the impartiality of
anti-corruption efforts, wrongdoings are followed up, and politicians are not
perceived to be above the law.
• Political parties are not perceived to be vehicles of success.
• Whistleblowing is encouraged both in the general public and within the
administration, and is the most important channel to detect wrongdoings.
• Accountability reports are filed regularly, detailed statistics on filed cases and
indictments even on a monthl
y basis. Annual reports analyze measures,
outcomes and public perceptions of anti-
corruption work, and measures to
improve work are discussed.
High levels of individual freedom, collective action capacity, and negative
attitudes towards corruption.
• Taiwan’s citizens enjoy freedom of information, organization and speech. The
media is free, but not impartial, yet all political blocks have an outlet. Equality
before the law is also guaranteed.
• Most people believe that voting can make a difference; corrupt practices are
widely denounced and greatly
covered by the media. However, people
sometimes allow themselves to be corrupted, for example in vote-buying.
• Several legislators have lost their seat, regardless of political affiliation, when it
was discovered that they bought votes. Rallies for good governance have taken
III. Summary and Conclusions
Taiwan has made impressive strides towards clean government. Although the secretive
nature of corruption makes it difficult to exactly measure progress, a number of indicators
back up this statement. First, it has become clearer what exactly constitutes corruption in the
Taiwanese case, and corruption has become far riskier than before. The previous DPP
administration is to be credited for specifying in legal terms which activities constitute
corruption, and which do not. Almost all laws and legal revisions that had been proposed in
the early 2000s have passed in the meantime, and the thorough documentation by Taiwan’s
Agency Against Corruption indicates that violations of the laws are indeed sanctioned.
Given the importance of corruption in public discourse, the existence of a critical media
sector and equally, if not more critical, the awareness of the general public sphere serve as
monitors for the government’s anti-corruption activities. It was shown in this analysis that this
awareness preceded the DPP’s anti-corruption agenda, and indeed contributed to bringing
the DPP to power. Urbanization, intimately connected with economic modernization and the
concurrent exposure of the Taiwanese public to “Western” values of fairness and impartiality,
played no small part in leveraging this change.
However, it seems that the general public is overly critical, because real improvements in
Taiwan’s anti-corruption regime are not reflected in the public opinion. One chief reason for
this is that the term is politicized to an extent that precludes a differentiation between
different kinds and different realms of corruption. Although the general state of governance
has improved and is not related to the embezzlement of state funds by powerful individuals,
the public seems to conflate these realms. In other words, the public does not believe that an
administration in which chief officers are convicted of corruption can be clean.
However, this report indicates that this seems to be exactly the case. This interpretation is
strengthened by the fact that these individuals, once discovered, are indicted and often
sentenced to pay heavy fines and serve lengthy jail terms. While this qualifies as an indicator
of impartiality, public actors still accuse the judiciary of favoring KMT members, because the
sentences the latter receive are perceived as too lenient. Given the differences in the crimes
committed, the differences in the laws that are applied, the leeway each law grants a judge,
and the different propensities of individual judges to impose the maximum penalty, this
accusation is difficult to verify. As long as a trial is conducted according to law, and as long
as the same laws are applied to the same crimes, the principle of impartiality seems not to be
A second issue pertains to cultural factors. Most important of all is the importance the
Chinese culture applies to cultivating interpersonal relations. This importance is manifested
in the comparatively large amount of money people are willing to invest in order to be
admitted into a network that is beneficial for their careers, but also in the fact that people give
monetary presents to obtain services they are entitled to in any case by law, as for instance
high-quality medical services and education. Of course, as in any other society, the
propensity to bribe and the amount of monetary capital that is exchanged for social capital
differs across social strata, and more research is needed to establish how large the impact of
cultural factors is in various social realms.
In sum, the most important mechanisms responsible for the improvement of Taiwan’s
governance are institutional changes following a critical juncture in Taiwan’s politics: the
change in ruling parties after the DPP’s electoral success. As the survival of the DPP
depended on a change of the rules of the game, which was to a large degree based on
political corruption, the regime elites were adamant in their fight against corruption.
Institutional change was followed up by crackdowns on bribery, embezzlement, money
laundering, and vote-buying. Another important factor was cultural change: corrupt practices
are more prevalent in remote locations and traditional communities. In contrast, well-
educated young people tend to be more averse to these practices. Hence, the improvement
of Taiwan’s state of governance is the result of a political project combined with social
change, which in turn is the result of successful economic development and the introduction
of “Western” values.
IV. Results beyond the empirical assessment
Previous research has found that corruption is “sticky” in the sense that rapid improvements
in a country’s governance regime are extremely rare (Uslaner 2008). Researchers have
proposed various hypotheses why this is so, but a definite answer is still found wanting. In
the Taiwanese case, two factors were identified as being crucial: institutional reforms and
cultural change. The fact that cultural change takes a long time can in part explain the
“stickiness” of corruption. Political or educational campaigns might help to speed up this
process, but it is far from certain if this is indeed the case. Social modernization, which is a
spin-off from economic development, takes generations to achieve, and is not uniform across
a society. Some places and social strata change fast, while others remain traditional for a
A more immediate factor that influences both attitudes to corruption and the prevalence of
the phenomenon itself is institutional and organizational reforms. The establishment of potent
anti-corruption agencies in Singapore and Hong Kong is often credited with improving
government in these polities. However, the crucial question is how such an organization can
be established, and how it can be made potent. Korea might present a counter-example to
Hong Kong and Singapore. Here, an anti-corruption agency was established, but it was not
given enough resources and clout to act independently (Göbel 2004). That governments who
have been operating on the basis of informal relations must change their own ways of
governing seems to be the chief problem. Our research suggests that this problem can only
be overcome at critical junctures, such as regime-threatening demonstrations against police
corruption in Hong Kong or a change in ruling parties in Taiwan. Absent regime-threatening
events or regime changes, it remains doubtful that corruption can be combatted effectively.
This finding needs further thought and should be tested for other countries as well. As the
case of Taiwan has shown, elections are instrumental for this process: it was not the regime
change per se that initiated the fight against corruption, but the opposition’s coming to power
a decade later. In most countries, regime change and change of the ruling party converge, so
these elements tend to be conflated. Developments in Taiwan, and for that matter Korea,
demonstrate that elections alone are not sufficient for the initiation of anti-corruption
programs – indeed, elections without changes in the ruling party worsened corruption only
when free and fair elections brought the opposition into power, was corruption fought in
A final issue concerns the notoriously difficult concept of culture and the question of how to
measure, for example, the importance of informal personal relations in a country. More
research is needed to make “culture” more tangible.
As for the concept and indicators used in this study, they are a definite improvement over
perceptions-based indicators, which are not a credible yardstick of the quality of government.
Apart from this, the diagnosis questions left much room for context-specific analysis, but it
remains to be seen how a meaningful comparison can be achieved over idiosyncratic country
reports. This is even truer as the indicators did not seem to match the questions in Table 1.
Still, I believe that a large number of country reports that take a fresh look at partial behavior
are a good basis to kick-start a fruitful re-evaluation of the corruption concept that might, in a
second step, lead to the formulation of a model that can be broadly tested.
AFP. (2013). “Taiwan's former top cabinet official jailed for graft”, 30 April.
BBC News. (2009). “Taiwan ex-leader jailed for life”, 11 September.
Bertelsmann Foundation. (2013). Bertelsmann Transformation Index 2013. Country Report:
Taiwan. Available at http://www.bti-
project.de/uploads/tx_jpdownloads/BTI_2014_Taiwan.pdf. [Accessed 04 July 2013].
Chao, V. Y. (2010). “Three judges, prosecutor held on graft charges”. Taipei Times, July 15,
Chen, M.-T. (1995). Paixi zhengzhi yu Taiwan zhengzhi bianqian, Taipei, Yuedan.
Chen, M.-T. and Chu, Y.-H. (1992). “Quyuxing lianhe duzhan jingji, difang paixi yu
shengyiyuan xuanju: yi xiang shengyiyuan houxuanren beijing ciliao de fenxi, 1950-
1986”, Proceedings of the National Science Council C, Social Sciences and
Chin, K.-L. (2003). Heijin: organized crime, business, and politics in Taiwan, Armonk, N.Y.,
Chu, Y.-H. and Lin, T.-M. (1996). “The process of democratic consolidation in Taiwan: social
cleavage, electoral competition, and the emerging party system”. In: Tien, H.-M. (ed.)
Taiwan’s electoral politics and democratic transition: riding the third wave. Armonk/
London: M.E. Sharpe.
Fawubu L. (2012). Fawubu Lianzhengshu 100 niandu gongzuo baogao, Taipei, Fawubu
Fell, D. (2002). “Party platform change in Taiwan’s 1990s elections”. Issues and Studies. An
International Quarterly on China, Taiwan, and East Asian Affairs, 38, 31-60.
Göbel, C. (2004). “Beheading the Hydra: combating corruption and organized crime”, China
Göbel, C. (2012). “The impact of electoral system reform on Taiwan’s local factions”, Journal
of Current Chinese Affairs, 69-92.
Göbel, C. (2013). “Warriors Unchained. Critical Junctures and Anti-Corruption in Taiwan and
South Korea”. Zeitschrift für Vergleichende Politikwissenschaft,.
Heidenmeier, A. J. and Johnston, M. (2002). Political corruption: concepts & contexts, New
Brunswick, N.J., Transaction Publishers.
Huang, G. (1991). “Tanwu wenti”. In: Guoqu, Y. and Qizheng, Y. (eds.) Taiwan de shehui
Kastner, J. (2010). “Doubts over Ma's tough cure for corruption”. Asia Times Online, August
MOJ (Ministry of Justice) (2013a). Difang fayuan jianchashu zhenban tandu qisu anjian.
Available at http://www.aac.moj.gov.tw/public/Attachment/352920382877.pdf
[Accessed 04 July 2013].
MOJ (Ministry of Justice) (2013b). Gongzhi renyuan caichan shenbao shenyi qingxing.
Available at http://www.aac.moj.gov.tw/public/Attachment/352920364386.pdf
[Accessed 04 July 2013].
MOJ (Ministry of Justice) (2013c). Gongzhi renyuan liyi chongtu huibi shenyi jieguo.
Available at http://www.aac.moj.gov.tw/public/Attachment/352920372892.pdf
[Accessed 04 July 2013].
MOJ (Ministry of Justice) (2013d). Lianzhengshu xinshou anjian biduan xiangmu fenxi,
Zhonghua Minguo 102 nian 1 yue. Available:
http://www.aac.moj.gov.tw/public/Attachment/331517485238.pdf [Accessed 04 July
MOJ (Ministry of Justice) (2013e). Lianzhengshu xinshou anjian biduan xiangmu fenxi,
Zhonghua Minguo 102 nian 2 yue. Available at
http://www.aac.moj.gov.tw/public/Attachment/33191832435.pdf [Accessed 04 July
MOJ (Ministry of Justice) (2013f). Lianzhengshu xinshou anjian biduan xiangmu fenxi,
Zhonghua Minguo 102 nian 3 yue. Available at
http://www.aac.moj.gov.tw/public/Attachment/342911154275.pdf [Accessed 04 July
MOJ (Ministry of Justice) (2013g). Lianzhengshu xinshou anjian biduan xiangmu fenxi,
Zhonghua Minguo 102 nian 4 yue. Available at
http://www.aac.moj.gov.tw/public/Attachment/35292044171.pdf [Accessed 04 July
MOJ (Ministry of Justice) (2013h). Quanguo zhengfeng jigou qita chachu yewu qingxing.
[Accessed 04 July 2013].
Office of the President. (2011). Guojia lianzheng jianshe xingdong fang'an zong shuoming.
Available at http://www.president.gov.tw/Default.aspx?tabid=811 [Accessed 04 July
Transparency International. (2011). 2010/11 Global Corruption Barometer. Available at
Transparency International. (2013). Corruption Perceptions Index 2012. Available at
http://cpi.transparency.org/cpi2012/results/ [Accessed 04 July, 2013].
Slaner, E. M. (2008). Corruption, inequality, and the rule of law: the bulging pocket makes
the easy life, Cambridge; New York, Cambridge University Press.
World Bank. (2013). Worldwide Governance Indicators. Available at
http://info.worldbank.org/governance/wgi/pdf/wgidataset.xlsx [Accessed 04 July
Yang, M. M.-H. (1994). Gifts, favors, and banquets: the art of social relationships in China,
Ithaca, N.Y., Cornell University Press.
Yu, C., Chen, C.-M., Juang, W.-J. and Hu L.-T. (2008). “Does democracy breed integrity?
Corruption in Taiwan during the democratic transformation period”, Crime, Law and
Social Change, 167-84.
This project is co-funded by the
Seventh Framework Programme for
Research and Technological
Development of the European Union
ANTICORRP is a large-scale research project funded by the European
Commission’s Seventh Framework Programme. The full name of the project is “Anti-
corruption Policies Revisited: Global Trends and European Responses to the
Challenge of Corruption”. The project started in March 2012 and will last for five
years. The research is conducted by 21 research groups in sixteen countries.
The fundamental purpose of ANTICORRP is to investigate and explain the factors
that promote or hinder the development of effective anti-corruption policies and
impartial government institutions. A central issue is how policy responses can be
tailored to deal effectively with various forms of corruption. Through this approach
ANTICORRP seeks to advance the knowledge on how corruption can be curbed in
Europe and elsewhere. Special emphasis is laid on the agency of different state and
non-state actors to contribute to building good governance.
Project acronym: ANTICORRP
Project full title: Anti-corruption Policies Revisited: Global Trends and European
Responses to the Challenge of Corruption
Project duration: March 2012 – February 2017
EU funding: Approx. 8 million Euros
Grant agreement number: 290529
Project website: http://anticorrp.eu/