Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues [1 ed.] 9781616680596, 9781604568080

Citation preview

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved. Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved. Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

TAIWAN: ECONOMIC, POLITICAL AND SOCIAL ISSUES

No part of this digital document may be reproduced, stored in a retrieval system or transmitted in any form or by any means. The publisher has taken reasonable care in the preparation of this digital document, but makes no expressed or implied warranty of any kind and assumes no responsibility for any errors or omissions. No liability is assumed for incidental or consequential damages in connection with or arising out of information contained herein. This digital document is sold with the clear understanding that the publisher is not engaged in Taiwan: Economic, Political and Sociallegal, Issuesmedical : Economic, Political andprofessional Social Issues, services. Nova Science Publishers, Incorporated, 2008. ProQuest rendering or any other

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved. Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

TAIWAN: ECONOMIC, POLITICAL AND SOCIAL ISSUES

LACHLAN N. ROBERTS AND

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

COOPER R. LEWIS EDITORS

Nova Science Publishers, Inc. New York

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Copyright © 2008 by Nova Science Publishers, Inc. All rights reserved. No part of this book may be reproduced, stored in a retrieval system or transmitted in any form or by any means: electronic, electrostatic, magnetic, tape, mechanical photocopying, recording or otherwise without the written permission of the Publisher. For permission to use material from this book please contact us: Telephone 631-231-7269; Fax 631-231-8175 Web Site: http://www.novapublishers.com NOTICE TO THE READER The Publisher has taken reasonable care in the preparation of this book, but makes no expressed or implied warranty of any kind and assumes no responsibility for any errors or omissions. No liability is assumed for incidental or consequential damages in connection with or arising out of information contained in this book. The Publisher shall not be liable for any special, consequential, or exemplary damages resulting, in whole or in part, from the readers’ use of, or reliance upon, this material. Any parts of this book based on government reports are so indicated and copyright is claimed for those parts to the extent applicable to compilations of such works.

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Independent verification should be sought for any data, advice or recommendations contained in this book. In addition, no responsibility is assumed by the publisher for any injury and/or damage to persons or property arising from any methods, products, instructions, ideas or otherwise contained in this publication. This publication is designed to provide accurate and authoritative information with regard to the subject matter covered herein. It is sold with the clear understanding that the Publisher is not engaged in rendering legal or any other professional services. If legal or any other expert assistance is required, the services of a competent person should be sought. FROM A DECLARATION OF PARTICIPANTS JOINTLY ADOPTED BY A COMMITTEE OF THE AMERICAN BAR ASSOCIATION AND A COMMITTEE OF PUBLISHERS. LIBRARY OF CONGRESS CATALOGING-IN-PUBLICATION DATA Roberts, Lachlan N. Taiwan : economic, political, and social issues / Lachlan N. Roberts and Cooper R. Lewis. p. cm. ISBN H%RRN 1. Taiwan--Economic policy. 2. Taiwan--Politics and government. 3. Political parties--Taiwan. I. Lewis, Cooper R. II. Title. HC430.5.R63 2008 330.95124'9--dc22 2008033023

Published by Nova Science Publishers, Inc.

New York

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

CONTENTS

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Preface

vii

Chapter 1

Political Parties and Electoral Politics in Taiwan Christian Schafferer

Chapter 2

Can the Deviation between Ownership and Control Explain Bank Performance in Taiwan? Chung-Hua Shen, Chien-An Wang and Meng-Wen Wu

Chapter 3

Predicting Bank Performance: A Distributional Approach Chiang Kao and Shiang-Tai Liu

Chapter 4

Corporate Financial Performance and Market Reaction to Employee Stock Ownership Plans in Taiwan Wei-Ning Chen, Dar-Hsin Chen and Po-Wei Lu

Chapter 5

Taiwan, Human Rights and Trade Ulric Killion

Chapter 6

Comparing the Structure and Competitive Priorities of Taiwan and Korea TFT-LCD Industry Pao-Long Chang, Han-Chi Chiang and Ying-Chyi Chou

Chapter 7

Taiwan Ida M. Conway

Chapter 8

A Study of Technical Efficiency of International Tourist Hotels in Taiwan Kuo-Liang Wang and Tai-Sen He

1

25 47

67 85

99 113

125

Chapter 9

School Reform: Initiatives & Issues Jenq-Jye Hwang, Chia-Chen Yu and Chia-Yu Chang

139

Chapter 10

Taiwan Gangs James O. Finckenauer

151

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

vi Chapter 11

Contents Assessing the Psychometric and Language Equivalency of the Chinese Versions of the Index of Nausea, Vomiting, and Retching and the Prenatal Self-Evaluation Questionnaire Fan-Hao Chou, Kay C. Avant, Shih-Hsien Kuo and Han-Fu Cheng

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Index

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

165 175

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

PREFACE Taiwan's economic, political and societal issues have become major points of interest to countries all over the globe. There has been an ongoing debate on Taiwan independence within the mainstream of domestic politics on Taiwan; political liberalization and the increased representation of opposition parties in Taiwan's legislature have opened public debate on the island's national identity; a broad popular consensus has developed that the island currently enjoys sovereign independence and - whatever the ultimate outcome regarding reunification or independence - that Taiwan's people must have the deciding voice; public opinion polls consistently show a substantial majority of Taiwan people support maintaining Taiwan's status quo for the foreseeable future; advocates of Taiwan independence oppose the stand that the island will eventually unify with mainland China. This new book presents in-depth analyses of issues such as foreign policy, political reform, and overall economic developments. Chapter 1 - There have been three stages of development in the history of Taiwan’s political party system since the Nationalist Chinese takeover in 1945: (1) Limited pluralist authoritarian party system (1945-1986), (2) pluralist party system with one dominant party (1986-2000), and (3) polarized pluralist party system. A limited pluralist authoritarian party system developed in Taiwan under the KMT regime between 1945 and 1986. During that time the KMT government allowed opposition candidates to take part in local and national elections, and to form loosely connected political organizations, but restricted the political power of the opposition. A pluralist party system with one dominant party evolved in liberalizing and democratizing Taiwan between 1986 and 2000. During that time, the regime tolerated the formation of political parties, and a number of parties took part in the political process. Nevertheless, the KMT still held an absolute majority in key decision-making bodies due to its continuing control over important institutions, such as the media, the military, and the judiciary. In the December 2001 national election, however, the KMT failed to gain an absolute majority of seats in parliament, ending its dominance in the lawmaking body. The election marked the beginning of a polarized pluralist party system. Chapter 2 - This chapter uses ownership data of Taiwan banks to investigate whether the deviation between ownership and controlling power is informative enough at an earlier stage to allow prediction of bank failure and bank performance. The gap between ownership and controlling power is proxied by voting rights (=voting rights/cash flow rights), as suggested by Claessens et al. (2000). This ratio is also referred to as moral hazard I (MH I here). For comparison and to reflect the unique local banking environment, the authors develop and test

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

viii

Lachlan N. Roberts and Cooper R. Lewis

a seat-control rights ratio (=seat-control rights/cash flow rights), which they refer to here as moral hazard II (MH II). Both moral hazard indices are useful [precursors] in terms of predicting returns (return on assets), but MH II outperforms MH I in predicting risk (nonperforming loans). In other words, in the case of Taiwan, MH II provides more useful information than MH I. Chapter 3 - Data envelopment analysis (DEA) has been used as a tool for evaluating past accomplishment in the banking industry. However, due to a time lag, the results usually arrive too late for the evaluated banking institutions to react. This study makes advanced predictions of the performance of twenty-four commercial banks in Taiwan based on their financial forecasts. The forecasts based on uncertain financial data are represented in ranges and treated as stochastic data. A simulation technique is adopted to obtain the efficiency distribution of each commercial bank. The authors’ finding is that all the efficiency scores calculated from the data published fall within the corresponding predicted ranges of the efficiency scores calculated from the financial forecasts. More importantly, the results show that the bad performance of two of the banks, which were taken over by the Financial Restructuring Fund of Taiwan, had been predicted in advance using the approach of this study. Chapter 4 - This study examines whether the financial performances of companies in Taiwan differ when the firms adopt employee stock ownership plans (ESOPs). The results of analyzing the market reactions, as reflected in the stock returns when an announcement is made in a board meeting regarding the adoption of an ESOP, indicate that the electronics and non-electronics industries differ significantly on terms of the ROE, profit margin and equity multiplier during the pre- and post-event periods. The total asset turnover rate in the nonelectronics industry, however, does not differ significantly during the pre- and post-event periods. Reactions toward the adoption of an ESOP are observed in the market before the event occurs, and the electronics industry accounts for the most significant reaction. Moreover, the authors find that a negative relationship exists between the CAAR and firm size, and that there is a positive relationship between the CAAR and the market-to-book ratio and the debt ratio. Chapter 5 – The bid for resumption by the People’s Republic of China (China) of membership in the General Agreement on Tariffs and Trade (GATT 1947), and pre-accession to the World Trade Organization (WTO) was burdened by many international political concerns. China was one of the original signatories to GATT. Primary problems concerning China’s accession to both the GATT and WTO were especially issues of Taiwan and human rights. Other issues concerned labor rights, environmental pollution and other concerns. The controversy of China’s accession to the WTO was primarily an issue of whether human rights, labor rights, environmental protections and other issues should be linked or de-linked with international trade. However, despite both a deluge of political issues surrounding China’s pre-accession to the multilateral world trading system and China’s non-GATTmember status, China in the 1980s had already emerged as a major player in the global economy.At the close of the infamous “Great Cultural Revolution” (Wenhua Da Geming) (1966-1976), China’s trade grew dramatically faster than world trade. China’s foreign trade rose from about $20 billion in the 1970s to about $475 billion in 2000, or, comparatively, from 0.6 percent of world trade (1977) to 3.7 percent of world trade (2000).

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Preface

ix

Chapter 6 - The TFT-LCD industry started in Japan, where during 1992-1994 companies invested in eight 2~2.5 generation TFT-LCD production lines. At the time, the main application was in notebook panels. Once Korean firms entered the fray against their large Japanese competitors, the market became overly competitive and supply exceeded demand, leading to repaid price drops. During 1995-1996, the three largest Korean firms Samsung, LG. Philips and Hydis set up four 3~3.5- generation TFT-LCD production lines within the period of one year, following which switched their LCD focus from notebook computers to the monitor product market. Yet fluctuations in international economic trends led to the 1997 Asian economic meltdown, which seriously harmed Japanese and Korean firms. Taiwanese firms were less seriously affected, and although they lacked the large dimension panel production experience of their Japanese and Korean counterparts, Taiwan firms had their advantages. They had the capital, and the downstream industries to support the development of the notebook and monitor industries. Taiwan firms engaged in technology partnerships with Japanese firms and in 1998-1999 established seven 3~3.5-generation TFT-LCD production lines. This was their formal entry into the TFT-LCD arena. Before 2000, Japan was the dominant producer in the large size TFT-LCD industry, holding 49.5% of the global market, with Korea second with 36.8%. Taiwan only had 13.7%. Yet Taiwanese and Korean firms invested more aggressively in new factories while Japanese firms remained more conservative, which allowed Korea to overtake Japan in the large size TFT-LCD industry in 2001, with a 40.5% share internationally. Japan fell to 33.5% while Taiwan enjoyed 26.1% [6]. As Japanese firms were withdrawing from production of large size TFT-LCD panels and switching to high resolution and low power consumption LTPS (low temperature poly-silicon) display development, Taiwanese firms wrestled 38% of the world market in the first quarter of 2004, while Korea still holding over 45% [7]. In light of the fact that the three main producers in the large size TFT-LCD product market are Taiwan, Japan, and Korea, the commercial behavior of firms in these countries will intimately influence the development of this industry. Moreover, Taiwan and Korea occupy a process technology niche position in the large size TFT-LCD industry; both nations focus on panel manufacturing. Japan, on the other hand, has a system specification niche position and has emphasized technology R&D, thus taking control of the development of components and manufacturing equipment. This allows Japanese firms to avoid the price slashing competition with Taiwan and Korea and to increase their production technology royalty earnings. Taiwan and Korea firms are Japan’s largest technology transfer customers, and the strategic alliances between Japanese and Taiwanese firms are very close. Therefore, there is now more cooperation than competition between Japan and Taiwan. This chapter focuses on the large size TFT-LCD industry in Taiwan and Korea, investigating the structures these two countries possess in the industry and the differences in their manufacturing strategies. Future development suggestions for Taiwanese firms will also be presented. Chapter 7 – The U.S. goods trade deficit with Taiwan was $15.2 billion in 2006, an increase of $2.4 billion from $12.8 billion in 2005. U.S. goods exports in 2006 were $23.0 billion, up 4.3 percent from the previous year. Corresponding U.S. imports from Taiwan were th

$38.2 billion, up 9.7 percent. Taiwan is currently the 11 largest export market for U.S. goods. U.S. exports of private commercial services (excluding military and government) to Taiwan were $6.4 billion in 2005 (latest data available), and U.S. imports were $6.4 billion. Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

x

Lachlan N. Roberts and Cooper R. Lewis

Sales of services in Taiwan by majority U.S.-owned affiliates were $10.2 billion in 2004 (latest data available), while sales of services in the United States by majority Taiwan-owned firms were $475 million. The stock of U.S. foreign direct investment (FDI) in Taiwan was $13.4 billion in 2005. U.S. FDI in Taiwan is concentrated largely in the finance, manufacturing and wholesale trade sectors. The United States and Taiwan continued to work together to enhance economic cooperation through our bilateral Trade and Investment Framework Agreement (TIFA) process. The TIFA, which was established in 1994, is an important mechanism for both parties to resolve bilateral trade issues and to address the concerns of the U.S. business community. The United States and Taiwan held a productive meeting of the fifth meeting of the TIFA Joint Council in Taipei, May 25-26, 2006, covering issues related to agricultural trade, intellectual property rights, pharmaceuticals, government procurement and investment, as well as other areas. Chapter 8 - Based on the survey data of Taiwan’s international tourist hotels in 2002, this chapter first uses the DEA to access technical efficiency of each international tourist hotel, and then applies the Tobit censored regression model to investigate the relationship between technical efficiency and firm-specific characteristics. The DEA evaluation results show that international tourist hotels in Taiwan could have reduced inputs by at least 15 percent, on average, and still have produced the same level of outputs. The mean scale efficiency measure implies that the inefficiency is mainly from wasting resources instead of inappropriate production scale. The regression results show that an international tourist hotel’s size has a positive impact on its technical efficiency. The impact of an international tourist hotel’s service concentration on technical efficiency is positive. An international tourist hotel’s degree of concentration in guest type is positively related with technical efficiency. International hotels located in Taipei are more technically efficient than those located in other areas. International hotels located in resort areas are more technically efficient than those located in other areas. Chapter 9 - Hargreaves and Fullan asserted, "Beyond the walls of our schools, the world is changing dramatically; and it is pushing and pressing down hard on those who work within" (Hargreaves and Fullan, 1998: v). There is always a close connection between schooling and social changes in a country. Educational authorities and educators should devote themselves to improving their schools and to coping with the change. Taiwan has made these same endeavors. In order to meet the great challenge of social-political change since martial law was lifted in 1987, Taiwan has been striving to launch many educational reforms to improve her schooling. The 1990s were a critical era for rapid reforms to education in Taiwan. Reacting to the conservative education policy and the chronologically accumulated dissatisfaction with the performance of schools, Taiwan not only has put decentralization of educational policy and deregulation into action, moving away from improper governance and domination of central authority, but also empowered schools and teachers to expand their professional autonomy. Especially, the Ministry of Education (MOE), the highest authority of education in Taiwan, has launched The Action Program for Educational Reform, which was approved by the Executive Yuan on May 14, 1999. A total of NT $157 billion was allocated to carry out the following twelve projects:

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Preface

xi

consolidation of elementary education; making pre-school widely accessible; consolidating existing systems for teacher training and continuing education; creating dynamic and quality technological and vocational education; achieving excellence within higher education; encouraging lifelong education and information education; promoting family education; improving education for disabled students; enhancing education for aboriginal students; making the paths of education more accessible; re-establishing counseling systems; and advancing education research and funding (MOE, 1999a, 1999b). Most of these are focused on elementary and secondary education, especially on school improvement projects. This chapter is arranged as three parts to elaborate the current measures of Taiwan’s improvement in elementary and secondary schools and it examines the practical consequences of implementation as well as the issues surrounding school improvement. Chapter 10 – This chapter provides information on the organizational structure, violence, economic resources, political resources, response of law enforcement agencies to organized crime and the external environment for the following gangs:

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

• • •

The United Bamboo Gang Heavenly Alliance Four Seas Gang

Chapter 11 - This study was an initial psychometric test of the Chinese versions of the Index of Nausea, Vomiting, and Retching (INVR) and the Prenatal Self-Evaluation Questionnaire (PSEQ) in pregnant Taiwanese women. Although there already is evidence that the English-language versions of the scales are reliable and valid, it is important to verify the proper psychometric characteristics of the Chinese versions. Forward and backward translation, and a multiphase instrumentation study describing internal consistency, test-retest reliability, and content validity of the translated versions were conducted. A convenience sample was recruited from prenatal clinics in the south of Taiwan. Three measurement instruments were used in this study: the demographic inventory (DI), the INVR, and the PSEQ. Thirty pregnant women participated in the study. Both the internal consistency and stability coefficients of the INVR and PSEQ were satisfactory. The indices of content validity (CVI) for the Chinese versions of these two instruments were both 1.0, indicating that they are acceptable for use among Taiwanese pregnant women. This was the first instrumentation study of the INVR and PSEQ applied to Taiwanese pregnant women. Researchers could use this study as a model for future translation and application of psychometric instrumentation.

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved. Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

In: Taiwan: Economic, Political and Social Issues Editor: L. N. Roberts and C. R. Lewis

ISBN 978-1-60456-808-0 ©2008 Nova Science Publishers, Inc.

Chapter 1

POLITICAL PARTIES AND ELECTORAL POLITICS IN TAIWAN Christian Schafferer

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Overseas Chinese Institute of Technology

There have been three stages of development in the history of Taiwan’s political party system since the Nationalist Chinese takeover in 1945: (1) Limited pluralist authoritarian party system (1945-1986), (2) pluralist party system with one dominant party (1986-2000), and (3) polarized pluralist party system. A limited pluralist authoritarian party system developed in Taiwan under the KMT regime between 1945 and 1986. During that time the KMT government allowed opposition candidates to take part in local and national elections, and to form loosely connected political organizations, but restricted the political power of the opposition. A pluralist party system with one dominant party evolved in liberalizing and democratizing Taiwan between 1986 and 2000. During that time, the regime tolerated the formation of political parties, and a number of parties took part in the political process. Nevertheless, the KMT still held an absolute majority in key decision-making bodies due to its continuing control over important institutions, such as the media, the military, and the judiciary. In the December 2001 national election, however, the KMT failed to gain an absolute majority of seats in parliament, ending its dominance in the lawmaking body. The election marked the beginning of a polarized pluralist party system.

LIMITED PLURALIST AUTHORITARIAN PARTY SYSTEM (1945-1986) On 17 April 1895, the Treaty of Shimonoseki ended the Sino-Japanese War. According to the treaty, China ceded Taiwan, the Pescadores Islands and the Liaotung Peninsula in southern Manchuria to Japan.1 Under Japanese rule, Taiwan’s infrastructure was 1

Edwin O. Reischauer, The Story of a Nation (Tokyo: Tuttle, 1988), 152-53.

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

2

Christian Schafferer

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

tremendously improved, its food production industrialized and an export-oriented economy created. It was at that time that the foundations of the Taiwan’s economic miracle were laid. During World War II, the United States reached an agreement with Chiang Kai-shek providing that Taiwan would be returned to China after the war. The agreement was confirmed in the Potsdam Declaration of July 1945. Soon afterwards, Chiang Kai-shek appointed a committee headed by Chen Yi to take over the island’s administration. The KMT’s strategy was to infiltrate Taiwan’s society by the means of public participation in state-controlled elections. At the early stages, the plan failed to materialize because of the widespread dissatisfaction with the new regime that culminated in the 2-28 Incident (1947), in which mainland Chinese troops killed several thousand Taiwanese. After the incident, the regime continued its strategy of infiltrating Taiwan’s society by holding state-controlled elections. There were two basic types of state-controlled local elections throughout the martial-law period: Those held at the provincial level, and those at the sub-provincial level. At provincial level, there were direct elections of members to the Provincial Assembly, and the special municipality councils of Taipei and Kaoshiung. The chief executives at provincial level were appointed, though. Sub-provincial elections can be separated into four types depending on their administrative levels: 1) County level: At county level, there were direct elections of magistrates and provincial municipality mayors, and the members of county councils and provincial municipality councils. 2) County municipality level: At this level, members of county municipality councils, rural township councils and urban township councils, and the mayors of rural and urban townships were directly elected. 3) Borough level: At the borough level, the chiefs of boroughs and villages were directly elected. 4) Neighborhood level: The neighborhood was the smallest electoral unit. At this level, wardens were indirectly elected. Voter turnout in local elections was between 60 and 70 percent. Since the formation of new political parties was illegal at that time, opposition candidates could only register as independents. In 1950s, only one out of ten councilors at county municipality level was nominated by the KMT. In 1986, three quarters were affiliated with the party. At county level, the situation is similar but less dramatic. At provincial level, the KMT could not significantly increase its share of councilors. Although the average number of successful independents was the lowest in the elections of county chief executives, Taipei City councilors and provincial assembly members, there was a significant increase in the number of elected independent candidates (Table 1). Most of these independents were true opposition figures: The group of unaffiliated politicians can be divided into two categories, the first of which comprising those who were indeed critical of the KMT government and the latter of those politicians who were supporters of the KMT regime. Most of the candidates belonging to the latter group registered as independents after having failed to obtain the party’s nomination. It was not before the 1977 local elections that the former group gained momentum. The elections brought about the highest number of non-KMT provincial assembly members, and chief executives of counties and provincial municipalities. At least 14 of the 21 independent assembly members were genuine opposition figures, eight of whom

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Political Parties and Electoral Politics in Taiwan

3

were the highest vote getters in their constituencies. Moreover, four out of twenty chief executives of Taiwan’s twenty counties and provincial municipalities were anti-KMT activists. The election was also different from previous ones in the way that it was for the first time that the opposition run on a common platform. Moreover, in thirteen counties and provincial municipalities the number of votes cast for anti-KMT activists almost equaled the KMT votes, which took the party by surprise and encouraged opposition figures to continue their struggle for democracy within the system.2 A year later, the strengthened opposition movement tried to continue its electoral success at the national level by taking part in the supplementary parliament elections scheduled to be held at the end of the year. Table 1. Local elections in postwar non-democratic Taiwan (1946-1987)

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Average Number of number of Administrative Level elections held candidates per seat Provincial Level Provincial Assembly 9 1.9 Taipei City Council 5 1.43 Kaoshiung City Council 2 1.81 County Level Council members 11 1.76 Chief Executives 10 2.32 County Municipality Level Council members 13 1.69 Chief Executives 10 1.95 Borough Level Chief Executives 11 1.61 b

Average elected First direct unaffiliated election held candidate

Number of seats

Voter turnout

57-77 ↑ 48-51 ↑ 42

73.48 65.72 76.22

20.62 ↑ 16.8 ↑ 23.81

1954 1969 1981

814-1024 19-21

79.96 74.32

28.94↓ 13.66

1950/51 1950/51

3700-9778 ↓ 309-360 ↓

66.37 a 70.67 ↑

54.5 a ↓ 12.25 ↓

1946 1950/51

5105-6608 ↓

64.71 ↓

28.53c ↓

1951

Source: Author’s own calculation based on data provided by the Central Election Commission. a data on 1946 and 1948 election not available. b data on 1946 and 1948 elections not available. c data on elections held before 1973 not available.

FROM LOCAL TO NATIONAL LEVEL Apart from local elections, limited direct parliamentary elections were held during the martial law period. Members of the three chambers of parliament were first elected in 1947 and 1948 respectively in all provinces of China including Taiwan. Delegates to the National Assembly (NA), the constitution-drafting body, members of the Legislative Yuan (LY), the lawmaking body, were elected by universal suffrage, whereas members of the Control Yuan (CY), the watchdog organ, were elected indirectly, namely by local councils. According to Article 65 of the Constitution, LY members should serve a three-year term and elections of new members be completed three months prior to the expiration of each term. The term of 2

Lin Jia-long, “Taiwan difang xuanju yu guomindang zhengquan de shichanghua” [Taiwan’s local elections and the marketization of KMT’s political power], in Liangan jiceng xuanju yu zhengzhi shehui bianqian [Grassroots elections, political and social changes in China and Taiwan], ed. Chen Ming-Tong and Cheng Yong-nian (Taipei: Yuandan, 1998), 236.

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

4

Christian Schafferer

CY members was restricted to six years by Article 93, whereas the term of office of NA delegates should terminate on the day on which the next NA convened. Article 28 of the Constitution rules that NA elections be held every six years. Due to the Communist rule over most parts of China, new parliamentary elections were, however, impossible and thus members of parliament were frozen in office: The Council of Grand Justices ruled that the then members of parliament should continue to function until nation-wide elections could be held.3 However, population growth, the ageing of parliament members, and the increasing doubt over the KMT’s legitimacy to rule Taiwan made supplementary parliamentary elections necessary. In its fourth plenary session held in March 1966, the National Assembly, thus, enlarged presidential powers by amending the Temporary Provisions affixed to the Constitution. The amendment added two important clauses, one of which accorded the President the right to make adjustments to the administrative and personnel organs of the Central Government and promulgate regulations providing elections to fill the elective offices at the Central Government level, which had become vacant, or for which additional representation seemed necessary due to the growth in population. That amendment paved the way for supplementary elections of all three parliamentary organs. In his message to the nation held during the 1969 Chinese new-year festivities, President Chiang Kai-shek announced that national elections would take place later that year to fill vacancies and add new members to all the three parliament bodies.4 In March 1969, the government promulgated the rules and regulations concerning the supplementary national elections. The first such election was finally held in December 1969. Fifteen National Assembly members, eleven Legislative Yuan members and two Control Yuan representatives were elected. There were further supplementary elections in the following years (Table 2).

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Table 2. Number of elected seats supplementary parliament elections (1949-1987) 1969 1972 1973 1975 1980 1983 1986 1987 National 15 (1.06) 53 (3.76) – – 76 (6.19) – 84 (8.71) – Assembly Legislative Yuan 11 (2.30) 36 (7.91) – 37 (8.62) 70 (17.24) 71 (19.29) 73 (22.53) – Control Yuan 2 (2.82) – 10 (12.82) – 22 (29.73) – – 22 (31.88) Total 28 98 10 37 168 71 157 22

Compiled by the author based on date provided by the Central Election Commission. Numbers in brackets are percentage of total members.

The 1969 election was a by-election of five National Assembly members, and an election of 23 additional members of parliament due to population growth and the elevation of Taipei to a special municipality. The term of those members who represented the province of Taiwan and had been elected in 1947 did not expire before the 1972 election. Since then, all members elected to represent Taiwan had to stand for re-election, whereas all other members remained frozen in office. The number of additional seats had been increased over the years due the changes in the calculation formula of seats per district, the elevation of Kaoshiung to a special 3 4

Interpretation Shih Tzu No.31, Council of Great Justices. Central Election Commission, Zhong hua min guo xuanju shi [Republic of China election history] (Taipei: Central Election Commission, 1986), 389.

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Political Parties and Electoral Politics in Taiwan

5

municipality in 1979, and population growth. Since 1972, seats reserved for overseas Chinese representation were appointed by the president. The number of appointed members equaled about half of the elected members in 1972. This ratio decreased significantly, but remained almost unchanged in the composition of new Control Yuan members, that is about 45 percent. Although the number of parliament members elected in Taiwan increased over the years, there had been no chance of any law being passed without the support of the senior members—those elected on the mainland in 1948—and the newly appointed parliamentarians. Even in the late 1980s, the total number of newly elected members remained far below 50 percent of the total number of parliamentarians (Table 7.2). Nevertheless, these elections gave the opposition the opportunity to challenge the KMT government not only in local elections but also in national ones. Although the ratio of newly elected members to the total membership was anything but impressive, these elections were de-facto national ones since they were not in any way limited to only a small part of the area under the jurisdiction of the KMT government. At that time, Taiwan had already emerged as the de-facto leftover of Sun Yat-sen’s Republican China. After the election victory in 1977, the opposition was determined to expand its success from the local to national level by taking part in national elections. The 1978 parliamentary elections were hotly contested by candidates of both the opposition and the ruling party. In order to maximize the support of opposition candidates and to co-ordinate their campaign efforts throughout the island, a special assistance committee was founded by opposition leaders such as Huang Hsin-chieh and Shih Ming-teh, and named Tangwai Campaign Assistance Committee (dangwai zhuxuan tuan).5 Tangwai (also dangwai) was a term used to refer to those independent candidates who opposed the KMT and literally means “outside the party.” Since it was illegal to form political parties, collectivism among opposition candidates could only exist without official institutions. The 1978 election was expected to be a turning point in Taiwan’s political history. The opposition had high expectations since the KMT regime appeared to be less suppressive than in the past. In early December, non-KMT candidates were even allowed to make public a common platform for the elections and to introduce it to the local and foreign media, which let the opposition and foreign observers even more consider the election crucial for Taiwan’s political future. However, US President Carter’s decision to recognize the regime in Beijing and de-recognize Taipei as the only legitimate government of China just a few days before the elections should have taken place was a shock to the KMT government, and elections were postponed indefinitely. A move most likely caused by the regime’s uncertainty about the outcome of the election. Immediately after the elections were cancelled, democracy once again appeared to be a myth, since US pressure seemed gone. But, as a matter of fact, US pressure on Taipei increased as a result of the Taiwan’s Relations Act, which was passed shortly afterwards by the US Congress, and restored Taipei-Washington relations. The opposition tried to preserve some of the momentum gained during the interrupted election campaign by staging a number of indoor and outdoor meetings. At the beginning of June, the opposition established two organizations that should provide organizational facilities to opposition members of parliament and the provincial assembly respectively, namely the Joint Office of Tangwai Representatives (Taiwan dangwai minyi daibiao lianhe 5

Jürgen Domes, “Political Differentiation in Taiwan: Group Formation Within the Ruling Regime and the Opposition Circles,” Asian Survey XXI, no. 10 (1981): 1012; Mab Huang, “Political Ko’tung and the Rise of the Democratic Progressive Party in Taiwan,” Soochow Journal of Political Science 5 (1996): 136.

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

6

Christian Schafferer

banshichu) and the Joint Committee of Tangwai Parliamentary Candidates (zhongyang minyi daibiao xüanjü houxüanren lianyihui).6 In August 1979, the opposition started to publish a magazine, Meilidao, which was the most important political publication at that time. The magazine aimed at forming a united front against the KMT leadership, regardless of the different ideologies and backgrounds of its supporters. Its board members comprised of sixty founding members of all circles and tendencies within the opposition. Prominent people such as Shih Ming-teh, Huang Shin-chie, Lin Yi-hsiung, Yao Chia-wen, Chang Te-ming, Huang Tian-fu, Hsu Shin-liang, Chang Chunhung and Lu Hsiu-lien, however, soon gained the upper hand. The dominant group around Meilidao set up several offices around the island and openly doubted the regime’s legitimacy to rule Taiwan. The group was of the opinion that the KMT could be forced into a fullfledged democratization through escalating series of mass meetings and demonstrations.7 Soon, this opposition group appeared to be a serious threat to the KMT regime. The government thus decided to discredit the activists by initiating what would later be called the Kaoshiung Incident. Without doubt, the cancellation of the 1978 election brought about temporary political disorder in Taiwan and put great pressure on both the KMT regime and the opposition. A year after the incident, the postponed 1978 election finally took place. The election was viewed as a watershed event in Taiwan’s political development, since it was the first time that the opposition agreed on a common platform and was able to openly criticize the government without facing harsh consequences. Moreover, the election was held under a new election law that was a compromise between the government and opposition figures. Both international observers and opposition figures were surprised and some even called the election a “political holiday.”8 In spite of garnering less than ten percent of the seats in this election, Copper and Chen claim that the opposition could still “boast of victory” for several reasons, such as the fact that several opposition figures succeeded in gaining the highest numbers of votes. Moreover, the public in general seemed sympathetic toward the opposition and did not negatively associate it with the Kaoshiung Incident.9 After the election, leading figures of the opposition noticed that the formation of a political party would be unavoidable if they wanted to gain legitimacy, channel financing, seek agreement on issues, arrange for co-operation during campaigns, and better negotiate the co-ordination of candidates.10 Thus, pressure from opposition circles on the authorities to legalize political parties increased constantly. Finally, on 28 September 1986, 135 members of the opposition established Taiwan’s first true opposition party, the Democratic Progressive Party (DPP). The party comprised both moderate and radical Taiwan independence activists; those who were anti-KMT but in favor of unification with mainland China also supported it, the most prominent of whom certainly was Zhu Gao Zheng. Although it was still illegal to form political parties in Taiwan, the authorities did not crack down on the newly formed DPP. In October, President Chiang Ching-kuo made public his intention to lift the martial law decree and allow the formation of political parties. In December, the DPP took part in parliamentary elections and succeeded in 6

Domes, Differentiation, 1012-13. Domes, Differentiation, 1012-13. 8 John F. Copper and George P. Chen, “Taiwan’s Elections: Political Development and Democratization in the Republic of China,” Occasional Papers/Reprint Series in Contemporary Asian Studies 64, no. 5 (1984): 59-67. 9 Fu Hu and Yun-han Chu, “Electoral Competition and Political Democratization,” in Political Change in Taiwan, ed. Cheng Tun-jen and Stephan Haggard (Boulder: Rienner, 1992), 183; Copper, Political Development, 69. 10 This co-ordination is decisive in elections, since the SNTV-system is applied in national elections in Taiwan. 7

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Political Parties and Electoral Politics in Taiwan

7

gaining substantial support (10 out of 73 seats in the Legislative Yuan and 10 out of 84 seats in the National Assembly). The KMT government was taken by surprise when it learned that some of their ‘iron seats’ such as those reserved for representatives of labor unions were taken by the DPP. In the following year, the KMT regime lifted martial law. Substantial political and social changes followed and led to a pluralist party system with a dominant KMT.

PLURALIST PARTY SYSTEM WITH ONE DOMINANT PARTY (1986-2000)

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

After the lifting of martial law in 1987, political parties mushroomed in Taiwan. In December 1989, the first national election after the lifting of martial law took place. Thirteen political parties participated in the election of 130 parliament members. The KMT nominated 140 candidates, the DPP 57, the two KMT satellites, the CDSP and CYP, nominated 2 and 3 respectively, and other parties 20. Although a number of different parties contested in the election, the DPP was the only opposition party that could gain substantial support and put pressure on the KMT government to reform Taiwan’s political system. One of the DPP’s key demands was the retirement of the senior parliament members— those elected on the mainland and frozen in office. Only through massive street protests did the KMT regime finally agree to pass a law requiring the “voluntary retirement” of these members. Constitutional amendments in April 1991 paved the way for elections of all members of parliament (National Assembly and Legislative Yuan).11 Throughout the 1990s, Taiwan’s party system remained unchanged. There were several interesting developments, though: 1) The number of political parties contesting in national elections decreased. 2) The KMT lost popular support, but no opposition party succeeded in preventing the party from obtaining an absolute majority in parliament. 3) There was a tendency towards a three party system during the first half of the 1990s, and another towards a two party one during the second half. There was much enthusiasm in 1989 to participate in national elections.12 Within a few years, however, most of that enthusiasm was gone and after the 1992 parliamentary election most people predicted the emergence of a two party system. When the New Party (NP), a 11

Amendments in May 1992 stipulated that the members of the Control Yuan be nominated and, with the consent of the National Assembly, appointed by the president. Further constitutional amendments were made in April 2000. Article 1 of these amendments turned Taiwan’s parliament into a semi-bicameral one. The term of all National Assembly members expired on 19 May 2000. After that day, members to the assembly will only be elected by proportional representation within three months of the expiration of a six-month period following the public announcement of a proposal by the LY to amend the constitution or alter the national territory, or within three months of a petition initiated by the LY for the impeachment of the president or the vice president. Elected members have to convene of their own accord within ten days after the confirmation of the election result and have to remain in session no longer than one month, with the term of office expiring on the last day of the convention. 12 In this election, a total of 101 members of parliament were elected: Taiwan Province: 58; Aborigines: 4; Special Municipalities (Taipei, Kaoshiung): 20; Fukien Province: 1; functional seats (workers, farmers, businesspeople, industrialists, fishermen and teachers): 18

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

8

Christian Schafferer

KMT splinter, was founded in August 1993 there suddenly seemed to be optimism that the KMT would lose its majority in parliament and a three party system would emerge soon. The NP’s official election campaign strategy (san dang bu guo ban) in the 1995 national election was to win enough popular support as to break the KMT’s monopoly in the Legislative Yuan, the lawmaking body of government.13 The party almost succeeded: The KMT’s share of seats dropped from 60 percent the party obtained in 1992 to 52 percent in the 1995 national election. The NP captured 13 percent of the total number of seats and the DPP 33 percent, two percentage points more, compared with the previous election. In the 1998 national election, the party nominated 51 candidates but only eleven (3 percent of total number of all elected legislators) got elected. The ruling KMT captured 55, the DPP 31, and the remaining seats were taken by minor parties and independents (see Table 3). These results seemed to have put an end to speculations about a three party system in Taiwan.

THE RISE AND DECAY OF MINOR PARTIES During the period of pluralism in Taiwan’s party system (1986-2000), there were a total of 32 political parties competing against the ruling KMT in parliamentary elections. Apart from the DPP and the NP, none of the other parties succeeded in gaining significant popular support, that is 5 five percent of the total votes cast for candidates nominated by political parties. These parties are referred to as minor parties in the following.14 Table 4. Minor parties in parliament elections (1989-2004)

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Number of parties Candidates

LY 1989 NA 1991 LY 1992 LY 1995 NA 1996 LY 1998 LY 2001 LY 2004

11 14 12 3 4 5 7 4

25 81 45 4 21 24 52 7

Elected

Total votes received

% of total votes cast

0 0 1 0 1 1 2 1

125,258 222,976 182,332 7,456 129,859 155,443 289,265 18,068

1.3 2.54 2.25 0.08 1.24 1.55 2.78 0.18

Table compiled by the author based on data provided by the Central Election Commission

As illustrated in Table 4, political parties competing in national elections proliferated at the end of the 1980s and the beginning of the 1990s. A total of fourteen minor parties (sixteen parties in total) took part in the 1991 National Assembly election—the highest number of 13 14

New Party, Qin xiu nei zheng ai Taiwan [Change the political system, love Taiwan] (Taipei: New Party, 1995), 16-30. Apart from minor parties, four political alliances took part in parliamentary elections. Although these alliances are registered as political parties they are not in practice. Alliances have merely been formed so as to get access to free TV advertising and to be eligible for nationwide seats allocated by proportional representation. See Christian Schafferer, The Power of The Ballot Box: Political Development and Election Campaigning in Taiwan (Lanham, Md.: Lexington, 2003), 72.

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook Central,

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Table 3. Candidates and elected officials nonsupplementary parliament elections (1947-2004)

Date

Legislative Yuan 21-23 January 1948a 19 December 1992 2 December 1995 5 December 1998 1 December 2001 11 December 2004 National Assembly 21-23 November 1947a 21 December 1991 23 March 1996 Control Yuan 10 January 1948a

Seats

Candidates DPP NP

DPP

Elected NP Otherc

Female

96 85 123 68 79

50 54 70 87 89

21 11 1 1

6 15 4 21 69 56

2 17 23 43 50 47

5 76 103

254 183

66 99

49

27 5 6

4 42 61

3

-

-

-

5

1

Otherb

Female

45 51 32 1

13 167 120 174 243 259

2 46 50 86 81 96

144 153

83

78 179 92

-

-

8

Turnout

Total

KMT

8 161 164 225 225 225

71.41 72.02 67.65 68.09 66.31 59.16

13 403 397 498 455 492

158 139 161 97 103

78 93 112 83 129

27 325 334

73.24 68.32 76.21

78 627 591

304 263

5

81.08

8

-

KMT

Table compiled by the author based on data provided by the Central Election Commission a Taiwan Province only b LY: 1948: no data on partisanship; 1992: CSDP(1), LP(1), SCP(1), WP(1), CUP(4), ACDP(1), CCP(1), RP(1), TP(1), ChSDP(25), CPWP(1), CPAP(2), independents(122); 1995: LP(2), CTDAP(1), CP(1), independents(115); 1998: CYP(1), CTDAP(1), NDNA(5), GP(1), NDP(1), TAIP(20), DU(25), NNA(12), independents(108); 2001: PFP(61), TSU(39), NDNA(1), TN1(3), WAP(1), TAIP(3), GP(1), CTDAP(1), GCBLU(1), independents(132); ); 2004: PFP(65), TAIP(4), TSU(40), WAP(1), NPSU(32), WA(1), independents(116); NA: 1947: no data on partisanship; 1996: LP(2), CTDAP(1), GP(1), PP(1), CYP(1), independents(69) c LY: 1948: no data on partisanship; 1992: ChSDP(1); 1995: independents(4); 1998: NDNA(3), TAIP(1), DU(4), NNA(1), independents(12); 2001: PFP(46), TSU(13), TN1(1), independents(9); 2004: PFP(34), TSU(12), NPSU(6), independents(4); NA: 1947: no data on partisanship; 1991: NDNA(3), independents(2); 1996: GP(1), independents(5)

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

10

Christian Schafferer

participating political parties ever. Three years later, only three minor parties (six parties in total) contested in national elections. After 1995, the number of (minor) parties participating in national elections increased slightly (Table 4) but never reached the 1992 climax. Between 1986 and 2000, there were only five minor political parties of political significance (the Labor Party, Workers Party, Chinese Social Democratic Party, Green Party, and the Taiwan Independence Party), and only three that obtained enough popular support as to get one of their hopefuls elected. In 1991, the Chinese Social Democratic Party (CSDP) captured on seat in the Legislative Yuan. The Green Party (GP) was successful in one constituency in the 1996 National Assembly election, and the Taiwan Independence Party (TAIP) in the 1998 Legislative Yuan election. Most of the remaining minor parties nominated one or two candidates who hardly garnered more than a few hundred votes in constituencies where at least 30,000 votes were necessary for a victory. With the exception of the CSDP, all of the politically important minor parties were single-issue oriented and part of the various social movements that emerged at the end of the 1980s. Among these, the Labor Party was the first to contest in national elections. The party was founded in November 1987 by socialist intellectuals and labor activists under the leadership of former DPP legislator Wang Yi-hsiung. At the end of the 1980s, labor disputes were on the rise and Wang Yi-hsiung thought that a party representing the interests of Taiwan’s 3.4 million industrial laborers would have a political future in democratizing Taiwan. The 1989 Legislative Yuan election was the first national election after the lifting of martial law. Wang Yi-hsiung had great expectations and his party nominated candidates in eight constituencies.15 Wang himself ran in the industrial city of Kaoshiung. The party, however, garnered only about one percent of the total votes cast, and none of the hopefuls was elected. Compared with the other minor parties, the results were impressive, though. In 1992, the party took part in elections for the last time. Hope was vested in artist Hsu Hsiao-tan, who contested in the city of Kaoshiung. The party adopted a rather different way of attracting voters in that election campaign: The candidate, Hsu Hsiao-tan undressed publicly on various occasions and promised an open campaign. The party's strategy almost worked out: Hsu Hsiao-tan succeeded in getting 32,349 votes and would have needed another 108 votes to have been elected. Instead of her, another, less open, female candidate (KMT) captured the parliamentary seat. The defeat of Hsu Hsiao-tan ended the party’s existence. The Labor Party was not the only political vehicle of the labor movement at that time. In March 1989, Luo Mei-wen founded together with a number of other socialist intellectuals the Workers Party (lao dong dang). Most of its 3,000 members had been former political prisoners, who belonged to branches of the Chinese Communist Party, workers and socialist intellectuals.16 Before establishing the Workers Party, Luo Mei-wen was the vice-chair of the Labor Party. There were several differences between Luo Mei-wen and the chair of the Labor Party, Wang Yi-hsiung, which led to Luo’s decision to quit the party: Firstly, Wang had no interest in grassroots activities, he only wanted to win elections and provide legal advice to workers. Luo argued that a party has to achieve more than just providing services to workers. In his view, the party has to be a symbol, a revolutionary force that struggles fiercely against 15

Two in Kaoshiung City (1st and 2nd electoral district); one in 12th electoral district of Taiwan Province (Kaohsiung County); one in the 14th electoral district of Taiwan Province (Taidong), one in the 7th electoral district, of Taiwan Province (Chang Hua), and three in the functional constituencies representing workers (lao gong tuan ti). 16 The party’s vice chairman, for instance, was 34 years and 7 months in prison. Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Political Parties and Electoral Politics in Taiwan

11

capitalist exploitation. A further difference was the attitude towards unification and socialism. Wang was of the opinion that unification with the People’s Republic of China would worsen the situation for the workers since a large number of mainland workers would influx Taiwan and lower wages. Wang was not interested in Marxism and saw in pragmatism a way to improve the lives of the workers. Luo, however, thought that only through unification could the working class get strong enough to fight effectively against capitalist, imperialist forces. Luo was pervaded with Marxist ideology and wanted to significantly reduce the influence of the capitalists even by revolution. Apart from the ideological differences, Wang Yi-hsiung and Luo Mei Wen belonged to different social classes. Wang studied Law at National Taiwan University and pursued further studies at academic institutions in Japan, France, and the UK. Although he had been a legislator representing the industrial city of Kaoshiung in the past, he had never been directly involved in the labor movement. Luo, on the other hand, was at that time a veteran activist of a powerful trade union in Hsinchu County (yuan dong cuan gong hui). The Workers Party filed three candidates in 1989 parliamentary election; none gained enough popular support. In 1991, the party concentrated its resources on the campaign of Luo Mei-wen, who ran for office in Hsinchu County, where he was quite popular due to his active role in the 1989 Hsin Pu-chen strike.17 Luo garnered 18,008 votes. With 10.35 percent of the votes cast, he ranked sixth in his electoral constituency. Luo Mei Wen would have needed another 5,613 votes (3.2 percent) for his victory.18 The failure of Luo Mei-wen in Hsinchu let even members of the party doubt about the possibility of having a powerful labor party in Taiwan. Others blamed the weak financial capabilities and the KMT-dominated media for the poor results.19 Popular support for the party diminished over the years, and after another defeat in 1996, the party decided to refrain from filing candidates in national elections. Apart from the Labor Party and Workers Party, there were two other important minor parties, which tried to become a vehicle of Taiwan’s social movements, namely the Green Party and the Taiwan Independence Party. Prof. Kao Cheng-Yan of National Taiwan University founded the first in January 1996. The party filed 16 hopefuls in the 1996 National Assembly elections. In total, the Green Party obtained about 1 percent, and only one hopeful was elected with 11 percent of the votes cast in Yunlin County. In the beginning, the party had a variety of objectives, such as abolishing the National Assembly and promoting social welfare legislation, and did not primarily focus on environmental issues.20 After the defeat in 17

In 1989, Luo Mei Wen was the leader of a trade union (yuan dong cuan gong hui) belonging to Far Eastern, one of Taiwan’s top 50 business conglomerates (total assets). The factory was situated in Hsin Pu-chen, Hsinchu county. Employees there were dissatisfied about the working conditions and payment; Luo Mei-wen demanded improvements from Far Eastern. However, as a result of the union’s activities, the union had its office ransacked by the police. That incident led to the union’s final decision to stage a protest against the company. According to him several thousand workers took part in the demonstration, which eventually proved unsuccessful. Tang Shu, a leading party official, blamed the government for the failure, claiming that government officials had put pressure on the families of the participants. Family members were told that there would be no work for them in future unless they withdrew their support for the union. (Interview with Tang Shu, Secretary-General, Workers Party, August 1996, Taipei). 18 There were nine candidates running for office in the same constituency. The district magnitude was four. 19 Interview with Tang Shu, Secretary-General, Workers Party, August 1996, Taipei. 20 With the president directly elected by popular vote, the Green Party as well as other opposition parties considered the National Assembly futile and tried to abolish it. Until its abolishment, the Green Party regarded it as “garbage” that should be recycled. Party candidates elected to the National Assembly were expected to use their seats to support constitutional changes allowing for the abolition of the chamber, and to use 80 percent of their salaries to financially support Taiwan's social movements. The Green Party initiated the formation of the Social Legislative Campaign Alliance (she fa lien) in September 1996. The alliance consisted of sixteen major social groups: the Taiwan Environmental Protection Union (taiwan huan jing bao hu lian meng), Animal

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

12

Christian Schafferer

1996, the party under the leadership of Kao Cheng-yan developed into a single-issue party. Kao ran a highly focused—though unsuccessful—campaign in 1998, and later in 2001. His objection to the construction of Taiwan’s fourth nuclear power plant became the only issue raised in his campaign speeches, pamphlets and newspapers ads. The Green Party was, thus, not able to repeat its 1996 success at the polls and disappeared from Taiwan’s political landscape after its defeat in 2001. In March 1996, the people of Taiwan could for the first time directly elect their president. Incumbent President Lee Teng-hui of the ruling party obtained 54.0 percent of the vote, whereas independence activist and former prisoner of conscience Peng Ming-min (DPP) only garnered 21.1 percent. The result was a setback for the independence movement. Their supporters blamed the DPP leadership for the disastrous outcome of the election. Independence activists believed that the perceived weakening of the DPP’s adherence to the founding principle of pursuing Taiwan’s independence had disappointed many core DPP supporters who either refrained from voting or cast their votes in support of Lee Teng-hui.21 The Taiwan Communiqué, a major news organ of the independence movement, wrote at that time: The attempts by DPP leaders Shih Ming-teh and Hsu Hsin-liang to play political power games, first by co-operating with the pro-unification New Party and recently with the ruling KMT, has deeply disappointed many of the DPP’s core supporters. In December 1995, thenchairman Shih Ming-teh orchestrated a “grand reconciliation” with the pro-unification New Party and aligned himself with the NP in an unsuccessful attempt to run for the presidency of the Legislative Yuan. In the spring of 1996, in a peculiar zigzag change of course, the new chairman of the DPP, Mr. Hsu Hsin-liang, offered to join the KMT in forming a coalition

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

government.

22

In October 1996, a group of leading members of the Taiwan Association of University Professors together with over two hundred other independence activists formed the Taiwan Independence Party. The party was supposed to become a platform for those dissatisfied with the current DPP polices regarding Taiwan’s status as an independent nation-state. The party exclusively focused on the independence issue. In its charter, the party declared that it was “the eternal, changeless and highest objective of the party to establish a new and independent Republic of Taiwan.” Three DPP legislators joined the party shortly after its formation. The Protection Association of the ROC (zhong hua min guo guan huai sheng ming xie hui), Alliance for the Disabled (can zhang lian meng), Taiwan Journalists' Association (taiwan xin wen zhi che xie hui), 410 Education Reform League (410 jiao yu gai cau lian meng), Social Workers' Association of the ROC (zhong hua min guo she hui gong zuo zhuan ye ren yuan xie hui), Foundation for the Encouragement of Social Welfare Institutions (li xin she hui fu li shi ye ji jin hui), Taiwan Labor Front (taiwan lao gong zhen xian), Legislative Action Committee of the Workers (gong ren xing dong li fa wei yuan hui), The Eden Social Welfare Foundation (yi dian she hui fu li ji jin hui), League for the Promotion of Welfare for the Aged (lao ren fu li tui dong lian meng), Car Accident Rescue Association of the ROC (zhong hua min guo che huo jiou yuan xie hui), Taiwan Association of University Professors (taiwan jiao shou xie hui), and the Modern Social Welfare Association (xian dai she hui fu li xie hui). The alliance’s main purpose is to push welfare legislation. 21 Lee Teng-hui was very popular due to the fact that he was the first president born in Taiwan and that he was believed to be a disguised independence activist. See Ian Buruma, “Taiwan’s New Nationalists: Democracy with Taiwanese Characteristics,” Foreign Affairs (July/August 1996) < http://www.foreignaffairs.org/ 19960701 faessay4214/ian-buruma/taiwan-s-new-nationalists-democracy-with-taiwanese-characteristics.html> (1 August 2003). 22 “Towards the fourth party,” Taiwan Communiqué 72 (October 1996): 14.

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Political Parties and Electoral Politics in Taiwan

13

Taiwan Independence Party was thus represented in parliament prior to the 1998 Legislative Yuan election. The party leadership vested great hope in 1998 national election and filed 20 candidates. Only one was elected, though. In total, the party secured 1.45 percent of the vote. Without doubt, the election was disastrous and the Taiwan Independence Party, too, was bound to vanish. Apart from these three single-issue parties, there was one catchall party, namely the Chinese Social Democratic Party (ChSDP). Former DPP member Zhu Gao-zheng, who got famous because of his verbal and physical attacks of KMT legislators at parliamentary meetings, founded the party. Zhu modeled his party on the German CSU (Christian Social Union), and it addressed all parts of Taiwan’s society. It was indeed a catchall party. Zhu Gao-zheng expected the party to become a third force in Taiwan’s politics. 23 The ChSDP, therefore, nominated 58 candidates in the 1991 National Assembly election—the highest number of candidates ever nominated by a minor party. None of the candidates was elected, though. The ChSDP garnered 2.12 percent of the total vote. A year later, the party took again part in a national election. However, only one out of the 25 party candidates was elected. With about 65,000 votes—18 percent of the votes cast—Zhu Gao Zheng was the second highest vote getter in Yunlin County. This time, total popular support, however, dropped to 1.5 percent. After the 1992 defeat the party dissolved and Zhu Gao Zheng joined the New Party (NP).24 Why did all these political parties fail? There are common and indigenous factors that led to the downfall of all minor opposition parties. The first include the limited financial capabilities of minor parties, the KMT’s media monopoly, and the lack of prominent political figures. Indigenous factors include strategic errors made in the nomination process of party candidates. Local and national elections have been held ever since—with the exception of several small constituencies—in multi member constituencies with the application of a system known as the single non-transferable vote (SNTV). Under this system, more than one hopeful is elected in each constituency, namely those garnering the highest votes.25 Multiseats constituencies require each political party to assess how many votes it reasonable hopes to poll in a certain constituency. If the party nominates too many candidates in the constituency, party votes may be split to the extent that rival candidates take the seats away. By nominating too few candidates, the party runs the risk of wasting votes. The former led to the disastrous defeat of the Chinese Social Democratic Party in the 1991 National Assembly election. The party nominated forty-three candidates in Taiwan province and the special municipalities of Taipei and Kaoshiung. Only the KMT and DPP had more hopefuls. Although the party garnered over two percent of the total vote, none of its candidates was

23

Interview, Zhu Gao Zheng, August 1996, Legislative Yuan, Taipei. See also Chinese Social Democratic Party, zhonghua shehui minzhu dang: jiben gangling [Chinese Social Democratic Party: Basic Principles] (Taipei: ChSDP, 1991). 24 In an interview with the author, Zhu Gao Zheng claimed that the NP merged with his party. Interview, Zhu Gao Zheng, August 1996, Legislative Yuan, Taipei. 25 The number of electoral districts has changed several times since 1945. In Legislative Yuan elections, for instance, provinces and special municipalities were electoral districts in the 1948 election Eight members were elected in the province of Taiwan in that election. In the last election of legislators (2001), the province of Taiwan was divided into twenty-seven constituencies. With the exception of two thinly populated counties— Penghu and Taitung—an average of seven hopefuls were elected in each of these constituencies.

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

14

Christian Schafferer

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

elected. Expectations of the party were too high and far too many hopefuls were nominated in as many as ten constituencies. Apart from nominating too few or too many candidates, the overall success of the party depends also on its ability to coordinate popular support. If the party network fails to support each candidate with the same degree of enthusiasm, one of the party’s candidates may take a disproportional number of votes away causing the defeat of the party colleague. As to ensure a proportional distribution of votes among the candidates contesting in the same constituency, the NP and the DPP practiced the so-called pei piao system (forced vote distribution). Pei piao is a rational system based on the fact that the chance of someone being born on Monday is the same as of someone being born on Tuesday. A party could, for instance, nominate five candidates in an electoral district, and give each of its candidates two single-digit numbers, that is zero and one to the first candidate, two and three to the second and so forth. Party supporters could then be urged to vote for the candidate whose number coincides with the last digit of their ID. If most party supporters followed the strategy, each candidate should receive an equal amount of votes. The pei piao system has been regarded as one of the key reasons for the success of the NP and the DPP in parliamentary elections.26 The ChSDP failed to adopt such a strategy, and consequently no even distribution could be achieved in those constituencies where the party filed more than one candidate. The WP and the LP made a similar mistake. Candidates opted to contest against each other in key constituencies instead of forming a coalition and nominating a common candidate. The split within the LP and the subsequent formation of the WP led to the downfall of both parties. The TAIP failed because it nominated rather unknown politicians as its candidates. Prominent party leaders, such as Lin Shan-tian, refused to take part in national elections since they questioned the legitimacy of the KMT government.27 The party’s influence thus diminished within a few years.

POLARIZED PLURALIST PARTY SYSTEM Taiwan’s party system was about to undergo significant changes at the turn of the century. The 2000 presidential election caused a deep split within the KMT and led to the formation of the People First Party (PFP). Its founder, Song Chu-yu, quit the KMT a year earlier after the party announced that it would not nominate him as the party’s presidential candidate. The national elections of December 2001 made Song’s party the third strongest force in parliament (20 percent of seats). Song’s move led to the end of a KMT majority in the legislative body. The KMT obtained 30 percent of the 225 seats at stake and the DPP 38 percent. The DPP thus became the largest political party represented in parliament. This 26

For the use of this system in the 1995 parliamentary election see Cheng-hao Pao, “Xin dang ping jun pei piao ce lüe jiao jie zhi yan jiu: yi ba shi si nian li fa wei yuan xuan ju wei li [The Effectiveness of the New Party’s Strategy of Forced Vote Distribution in the 1995 Legislative Election],” Journal of Electoral Studies 5, no. 1 (May 1998): 95-138. 27 Lin Shan-tian is one of Taiwan’s leading criminal law experts. He became famous as the leader of a movement calling for the abolition of §100 of the Criminal Law (also known as the Sedition Law). In an interview with the author, he emphasized that the KMT government is a foreign regime without legitimacy to govern Taiwan. He, therefore, could not take part in elections of representatives of an illegitimate institution. Interview with Lin Shan-tian, National Taiwan University, March 2000.

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Political Parties and Electoral Politics in Taiwan

15

constellation may suggest a three party system, but what has developed over the last few years in Taiwan instead is a polarized pluralist party system. There have emerged four political parties of great significance since the 2001 national election: the DPP, KMT, PFP and the Taiwan Solidarity Union (TSU). The latter is the smallest party of these four and captured about 6 percent of the 225 parliamentary seats in 2001. Former President Lee Teng-hui is the spiritual leader of the party that was founded in July 2001 with the aim of stabilizing Taiwan’s political system and of assisting the DPP to secure a majority in parliament.28 Currently, Taiwan’s political landscape is divided into the blue and the green camp. The terms ‘blue camp’ and ‘green camp’ derive from the main colour in the party flags of the KMT and DPP, respectively. The blue camp comprises the KMT, PFP and the NP, whereas the DPP and the TSU belong to the green camp. The blue camp favors unification with the People’s Republic of China under the formula “one country, two systems,” whereas the green camp is actively seeking international recognition as an independent nation-state with formal representation at the UN and other international bodies. The 2004 presidential election divided Taiwan’s society among party lines and seriously endangered Taiwan’s political and social development. Parliamentary elections held in December helped to improve the situation and will most likely depolarize Taiwan’s party system and society.

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

DEVELOPMENTS LEADING TO A POLARIZED PLURALIST PARTY SYSTEM AND SOCIETY When the KMT arrived in Taiwan in 1945, gentry’ politics dominated the island. With the introduction of local self-governance in 1950 under the 1946 constitution, powerful local factions (difang paixi) emerged and soon began to replace the power of the gentry. The nature of local politics in Taiwan changed with urbanization, industrialization, and liberalization. In the 1950s, the era of gentry and factional politics, Taiwan was an agrarian society. Most people lived in the countryside. In the 1960s, local politics changed in urban areas but remained dominated by gentry and local factions in rural areas. In urban areas, there was an increasing involvement of business groups in local politics. In the mid-1970, more than half of Taiwan’s population resided in urban areas and only 30 percent were employed in the agricultural sector. In urban and rural areas, a new era of local politics appeared. Apart from the traditional political forces, party politics came to light: the KMT versus the dangwai (nonKMT). With industrialization, wealth accumulated and money politics surfaced. Vote-buying (mai piao) emerged at this stage. Further significant changes occurred in the 1980s. In urban areas, money politics became an even more decisive factor in politics. More and more interest groups emerged. Social movements began to dominate the political scene. With the lifting of martial law in 1987, political parties mushroomed. It was about that time that a new phenomenon entered Taiwan’s political landscape—a phenomenon that led to the polarization of Taiwan’s pluralist 28

It is interesting to note that it is now the TSU that speaks out in favour of a fully independent Taiwan nation-state. The DPP has become pragmatic on various issues such as cross-straits relations and the official name of the island-state. Most of the TSU legislators and senior party staff are former members of the KMT, not of the DPP.

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

16

Christian Schafferer

party system and to the coup attempt of March 2004. Some analysts, such as Chao Yungmau, spoke of an over-politicization of Taiwan’s society.29 Perhaps it was more than just an over-politicization of Taiwan’s society. Perhaps it was the beginning of a new era in Taiwan’s political culture—the era of populism.

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

DEFINING POPULISM The existence of populism as a political ideology can be traced back to the early 1870s. Its founders, members of various agrarian organizations, engaged in a social analysis of contemporary American society and wanted to reform the societal structure. Lawrence Goodwin writes, “Populist reformers felt that business domination of the political process— through massive campaign contributions to friendly officeholders and persistently effective lobbying in the national Congress and the state legislatures—had proceeded to the point that the practice had begun to undermine the democratic idea itself.”30 In other words, populists at that time held that a common person is oppressed by an elite in society, and they considered it their mission to grasp the power from the self-serving elite as to use it for the benefit and advancement of the oppressed masses. By the late 1880s, the agrarian organizations had developed into broad networks that included thousands of black and white chapters of the Farmers Alliance. In 1891, these Populists established the People’s Party to contest the 1892 presidential election. The party was quite successful at the beginning, and joined with the Democratic Party to support William Jennings Bryan’s unsuccessful presidential bid in 1896. Thereafter, the party lost much of its own identity and gradually dissolved after the 1904 presidential election.31 At its early stages, populism was believed to be an ideology capable of solving the various societal problems that existed in the late nineteenth century America. It was, however, soon blurred with primitive and demagogic elements, and finally developed into a political instrument to instigate the masses. In other words, it developed into a vehicle of racism, antiSemitism or other similar ideologies. Lawrence describes it as “a behavioral manifestation of deep-seated prejudices and ‘status anxieties,’ not a sensible product aimed at correcting unbalanced or generally exploitative economic practices pervading American society.”32 The modern usage of the word populism mainly reflects Lawrence’s description. The Collins English Dictionary defines populism as “a political strategy based on a calculated appeal to the interests or prejudices of ordinary people.” According to this definition, a populist would then be a person, especially a politician, who appeals to the interests or prejudices of ordinary people as to reach a certain goal. It is important to note that a populist makes a “calculated appeal,” rather than just appeal to people. Populism can also descend into demagogy and cultism. West and Langone define a cult as “a group or movement exhibiting a 29

Chao Yung-myo, Taiwan di fang zheng zhi de bian qian yu te zhi [Changes in Taiwan Local Politics] (Taipei: Han Lu, 1998), 248-63. Lawrence Goodwin, “Populism,” in The Reader’s Companion to American History, ed. Eric Foner and John A. Garraty (3 August 2004). 31 Goodwin, “Populism.” See also Omar H. Ali, “The Making of a black Populist: A Tribute to the Rev. Walter A. Pattillo,” Oxford Public Ledger 121, no. 25 (28 March 2002) < http://www.geocities.com/salika4/Oxford PublicLedgerRevWAP.html> (1 August 2003). 32 Goodwin, “Populism.” 30

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Political Parties and Electoral Politics in Taiwan

17

great or excessive devotion or dedication to some person, idea, or thing and employing unethically manipulative techniques of persuasion and control . . . designed to advance the goals of the group's leaders to the actual or possible detriment of members, their families, or the community.”33 Henry Louis Mencken, one of the most influential American writers of the early twentieth century, defined a demagogue as “one who preaches doctrines he knows to be untrue to men he knows to be idiots.” Populism, cultism and demagogy can go hand in hand and are not mutually exclusive.

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

POPULISM IN TAIWAN In Taiwan, populism emerged in the late 1970s and became a common political strategy a decade later amid political, social, and environmental changes. Until the end of the twentieth century, populism was a predominantly local and single-issue oriented phenomenon. It surfaced as a byproduct of competitive local and national elections. As to win votes and attract the masses, an increasing number of politicians tried to understand the grievances of the electorate—especially that of the middle class and the various social movements that surfaced as a consequence of the worsening social and environmental conditions. There were at least eight major social movements promoting the interests of consumers, environmentalists, laborers, women, aborigines, farmers, students and teachers. Even though some of these movements emerged earlier than the 1980s, they all gained considerable political significance in that decade. The social movements of the 1980s differed not only in the degree of threat they posed to the state, but also in their capability of mobilizing internal resources. Some of the movements, such as those focusing on labor and environmental issues, did in fact pose a serious threat to the government, and caused the founding of cabinet-level agencies. Moreover, they succeeded in determining election issues and thus the political agenda of the regime. In the 1983 national election campaign, consumer protection emerged as the issue. Three years later, “environmental protection was the issue raised by almost every candidate.”34 It was this combination of powerful social movements and electoral competition that introduced populism to Taiwan’s political culture and at the same time speeded up the democratization processes. Populism of the 80s and early 90s was predominantly local and single-issue oriented. Politicians rather used populist rhetoric and campaign strategies than being full-fledged populists. At the end of the 1990s, Taiwan’s political system not only experienced a deep split within the KMT but also the birth of the nation-state’s first powerful populist leader and movement—a movement that radicalized Taiwan’s society and political culture. Populism was no longer local and single-issue oriented. It encompassed a large variety of issues and swept the whole island. The movement has its roots in the right wing of the KMT. After the death of President Chiang Ching-kuo, Lee Teng-hui took over the presidency. Lee was the first president who was born in Taiwan—a fact which contributed to his popularity among the population of Taiwan and caused discomfort among the more radical pro-China forces within 33

L. J. West and M. D. Langone, Cultism: A conference for scholars and policy makers. Summary of proceedings of the Wingspread conference on cultism, September 9–11(Weston, MA: American Family Foundation, 1985. 34 Mao-gui Zhang, She hui yun dong yu zheng zhi zhuan hua [Social Movements and Political Change] (Taipei: Institute for National Policy Research, 1994), 57. Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

18

Christian Schafferer

the KMT and society. Although Lee was officially praying unification his real intentions were no secret. In his view, Taiwan is not only a political entity separate from China but also a nation on its own. Inside the KMT, members of the non-mainstream faction, such as Jao Shao-kong and You Mu-ming, established the New KMT Alliance (xin guo min dang lian xian) in May 1990. Within three years, the faction developed into a political party of its own. The New Party (NP) gained significant support at the polls and soon became a third force in Taiwan politics (Table 8.3). At the end of the 1990s, however, the party lost most of its glamour. A number of internal conflicts and accusations of illegal activities carried out by some of its leaders harmed the party’s image as a ‘clean’ party. Moreover, democratization marginalized the pro-China forces in Taiwan whereas it strengthened the position of the pro-Taiwan politicians: The people of Taiwan, being isolated from the mainland, developed their own cultural and ethnic identity, despite the efforts of the former KMT government to assimilate the Taiwanese. At the end of 1990s, pro-China oriented political leaders could no longer win sympathy with radical views regarding ethnic identity and unification with Communist China. They, therefore, turned into populists and even demagogues. One of the most popular and successful populist leaders (demagogues) in Taiwan’s political history is Song Chu-yu, who was born in China in 1942. Song was educated in the US and held several leading position within the KMT apparatus. Between 1979 and 1984, for instance, he was Director of the Government Information Office and after that he was the head of the Department of Cultural Affairs, a central organ of the KMT. While assuming these positions, Song became one of the key figures behind the KMT’s attempt to “clean” Taiwan’s society from “illegal thoughts,” that is any critical assessment of KMT rule, and to prohibit the use of Taiwanese as a means of communication. In 1993, he was appointed governor of Taiwan and in the same year, he contested successfully in the first direct election of that post. At the end of the 1990s, Song was more and more marginalized by the KMT under the leadership of democracy-oriented Lee Teng-hui, especially when the KMT agreed with the DPP to streamline the Provincial Government and to suspend all provincial elections. In 1997, the constitution was amended accordingly. Song fiercely protested the amendment since it crossed his plans of a second term. A second term as governor would have given him the opportunity to improve his Taiwanese, which by that time had developed into the lingua franca of politics, and as to further extend his power base in southern Taiwan, which was one of the prerequisites for his ambitions to win the presidential race of March 2000. When his term as governor ended in 1998, he was still hoping to obtain the KMT’s presidential nomination. Lee Teng-hui, critical of Song’s understanding of democracy and passion for Taiwan, picked the less powerful bureaucrat Lien Chan as the presidential candidate. Noticing that the KMT leadership turned their backs on him, Song had no other option as to run as an independent. It was at that time when Taiwan’s first nationwide populist movement appeared. Song’s campaign strategy was to attack the political establishment, to blacken the image of the KMT and label the DPP anarchists. His campaign messages reached out to the ordinary people. He talked much about the economy and other social concerns, and often stressed that he wanted to serve the people of Taiwan. His political platform was named accordingly: New Taiwan People Service Team (xin taiwan ren fuwu tuandui). Song, finally, lost the presidential election by a small margin to Chen Shui-bian of the DPP. His defeat was, however, still a victory, considering the fact that he did much better in the election than his rival Lien Chan. Populism obviously was a key to political success. Song, therefore, wanted to extend his power by leading a populist movement based on his

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Political Parties and Electoral Politics in Taiwan

19

own political party. No other name would better describe the nature of Song’s political party (movement) than People First Party, which was eventually founded on 31 March 2000. Within a short period, it resembled more a nationalist movement similar to the Austrian Freedom Party led by populist leader Jörg Haider. Party staff and legislators turned into an army of sycophants dressed in orange vests. Whenever they appeared in public they would wear orange vests and convey the message that they were here to service the people of Taiwan, that they were the guardian angels—those who would help the ordinary people to deal with the corrupt government, those who would defend the rights of the laobaixing, the common people. Apart from the People First Party, Song founded an organization named the Friends of Song. According to its statutes, all members have to study Song’s philosophy, to identify themselves with Song’s beliefs, to follow him, to unquestionable support him, and never criticize any of his actions.35 This all leads to the question whether Song is a cult leader, a demagogue, or both.

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

THE 2004 PRESIDENTIAL ELECTION, POPULISM, AND THE LIEN-SONG REBELLION The March 2004 presidential election was different in many ways from the previous two direct presidential elections. Previous elections, for instance, had a number of presidential candidates representing a variety of views. This election, however, polarized society and led to violence and hatred among the population. The polarization was triggered off by an attempt of senior political leaders of the blue-camp, especially Song Chu-yu and Lien Chan to stop the marginalization of their political and economic power. They had one common enemy, namely Chen Shui-bian. The PFP and the KMT were aware of the fact that only if they agreed to run under a joint presidential ticket could they defeat Chen Shui-bian.36 Although Lien Chan received fewer votes in the previous presidential election than his more charismatic rival Song Chu-yu, he was nominated as the presidential candidate and Song ended up running for vice-presidency. The fact that Song’s party received fewer votes and seats in the 2001 parliamentary election than the KMT and performed poorly in the 2002 local elections may explain this awkward situation. Still the Lien-Song ticket itself is rather obscure: During the presidential race four years earlier, the two were worlds apart and made malicious accusation against each other. Lien Chan, for instance, described Song as cruel and unscrupulous (lang xin gou fei). Song, on the other hand, was of the opinion that Lien Chan had proved incompetent while serving as vice president and thus questioned Lien’s ability of becoming the head of state. Four years later, things appeared to be completely different. Another difference between the 2004 presidential election and the previous ones is the nature of the electoral campaign. In previous campaigns, candidates seemed to have their own platform and ideological background. During the first part of the election campaign, the DPP dominated the campaign by announcing major policy issues, such as the drafting of a new constitution. At that time it became obvious that the opposition lacked in consistency and own policies. The Lien-Song ticket appeared to be ideological bankrupt. The second part of the 35 36

These rules are summarized in a pamphlet published by the organization (song you hui xu zhi shou ce) Under the current law, the president and vice-president are elected on a single ticket by a plurality vote.

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

20

Christian Schafferer

campaign (about two weeks before the election) merely was a personal attack on Chen Shuibian. No more policy issues were discussed. The first campaign issues emerged at the beginning of September, when a group of one hundred-fifty thousand people gathered in front of the presidential office urging the president to change the official name of the island-state from Republic of China to Taiwan. A month later, two hundred thousand people took to the streets of Kaoshiung, Taiwan’s second largest city, in support of President Chen Shui-bian’s appeal for a new constitution. Chen had repeatedly said that a new constitution would be completed by 2006 and implemented by 2008, with its contents being decided by referendum. The opposition parties opposed Chen’s plans. KMT chair and presidential candidate Lien Chan repeatedly called them “silly” and “nonsense.” Opinion polls conducted in early November by the China Post, Taiwan’s leading daily newspaper, other media outlets, and private pollsters placed President Chen in the lead. Chen’s new constitution and referendum law seemed to be more popular among the electorate than the blue camp had assumed. Consequently, the blue camp made a policy U-turn. On 15 November Lien Chan came out with his constitutional plans, and a few days later he even supported a referendum law. At the end of November, the blue camp—having a majority in parliament—pushed through their version of the law. Voting in the LY resembled a circus performance of blue-camp legislators. They were running around yelling and holding selfmade posters saying that President Chen had once more cheated the people. Lien Chan’s bowing to public opinion polls raises the question whether Lien Chan has not changed from a bureaucrat to a populist too. In the past, Lien Chan lacked charisma, leadership, and political marketing skills. The 2004 election campaign made him a clone of Song—a person he admires for his talent of inducing the masses. Lien Chan emerged as the protector of the Taiwan people, the true leader who would wipe out all evil things. Under his leadership, the KMT joined Song’s populist movement. Anyhow, the DPP was quite at a loss and had to accept that their issues were taken away by the blue camp. Damage control was necessary and resulted in Chen Shui-bian’s plan of holding two national referenda on national security issues coinciding with the presidential election.37

ADOLF HITLER AND CHEN SHUI-BIAN: THE CLIMAX OF POPULIST RHETORIC In February, Chen Shui-bian and former President Lee Teng-hui came up with the idea of forming a human chain from the very north of Taiwan to the very south of the island. The human chain should symbolize resistance to China’s military threat and be in remembrance of

37

Chen’s referenda plan led to several discussions in local and foreign academic circles. The majority of scholars— most notable here are Flemming Christiansen of Leeds University and Huang Kuang-Kuo of the Department of Psychology, National Taiwan University—branded Chen Shui-bian a populist causing tensions between the Taiwan Straits and instigating racial unrest in Taiwan.37 Chen Shui-bian may have used the referenda to boost his popularity, but the content and meaning of the referenda do not in any way constitute a contradiction to his beliefs and ultimate goals. Thus, Chen may be called a populist for applying populist methods, but he certainly is not a demagogue. Moreover, Chen’s populism is by far less radical and less institutionalized than Song Chuyu’s populist movement. It is astonishing that neither local nor foreign scholars have looked into the question whether Song Chu-yu is a populist, demagogue or even a cult leader.

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Political Parties and Electoral Politics in Taiwan

21

the 2-28 Incident.38 Two million people took part in the rally, which surprised the blue camp. Lien Chan and Song Chu-yu, both born in China, accused Chen Shui-bian of creating ethnic division and conflict in Taiwan.39 As to counter the success of the green camp, the KMT/PFP planned to stage a rally on 13 March attracting even more participants. It was at that time that the electoral campaign entered its second stage: the personal attack on Chen Shui-bian. In the run-up to the planned rally, the blue camp started a media war against President Chen. More than a dozen different ads were placed in Taiwan’s leading newspapers and aired by major television stations, most of which were entitled ‘Change the President, Save Taiwan,’ and contained the message that incumbent President Chen is the scum of the nation. The tone and language used in the opposition’s campaign leaflets and ads were without doubt the worst ever found in any election campaign on the island. Newspaper ads even compared President Chen Shui-bian with Adolf Hitler and asked the electorate to end Chen’s dictatorship by voting for Lien Chan and Song Chu-yu. A photo of Hitler was added to make the message better understood.40 In central Taiwan, the KMT campaign headquarters distributed posters showing terrorist Bin Laden expressing his admiration of Taiwan’s ‘dictator’ Chen.41 In another ad, former Iraqi president Saddam Hussein can be seen. Apart from other claims, the ad stated that the referenda were illegal since they were to be held concurrently with the presidential election. The KMT referred to Article 17 of the referendum law, which according to the party clearly forbids the holding of referenda on the same day a national election takes place. Mysteriously, Article 17 of the law does not mention such a regulation. To put it differently, the KMT deliberately misled the public into believing that the referenda were illegal and that President Chen is above the law: a dictator like Saddam Hussein.42 The 3-13 rally attracted four million people around the island, which marked a new record in Taiwan’s election history. The rally itself resembled more a revolution than an electoral campaign activity, and climaxed when the Lien Chan made a dramatic gesture in Taipei and prostrated himself, kissing the ground alongside his wife and KMT Secretary-General Lin Fong-cheng in front of the Presidential Office. Shortly before Lien's surprise prostration, Song Chu-yu, leading the march in central Taiwan, also knelt on the ground with his wife and kissed the ground. Song and Lien said that the move was meant to demonstrate their love for Taiwan.43 One day prior to the election, incumbent President Chen Shui-bian and Vice President Lu Hsiu-lien were both shot while campaigning in southern city of Tainan, in an apparent political assassination attempt. The injuries were not life threatening, and both Chen and Lu were released from hospital on the same day. Nevertheless, the attack provoked shock and unease among the population. Subsequently, both presidential candidates agreed to cancel all campaign activities. The election, however, had to take place as scheduled on the following day since the election law only allows for suspension of election upon the death of a candidate. Supporters of Lien Chan and Song Chu-yu doubted the authenticity of the attack and worried that it would influence the outcome of the election. In an attempt to win back sympathy votes, supporters of Lien Chan and Song Chu-yu spread malicious rumors accusing 38

On 28 February 1947, protests against the KMT government were brutally suppressed and several thousand Taiwanese killed. 39 Basically, the population can be divided into two groups: those people of Chinese origin who arrived in Taiwan before the end of Second World War and those afterwards. The latter group is referred to as the ‘mainlanders.’ 40 The ad can be viewed at http://www.eastasia.at/vol3_1/ad1.htm 41 The ad can be viewed at http://www.eastasia.at/vol3_1/binl.htm 42 The ad can be viewed at http://www.eastasia.at/vol3_1/saddam.htm 43 A photo of the event can be viewed at: http://www.taipeitimes.com/News/front/photo/2004/03/14/2003113135

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

22

Christian Schafferer

President Chen of having planned the assassination attempt. Legislator Chen Wen-chien, for instance, claimed in her talk show broadcast live on one of Taiwan’s most popular TV channels that she would have proof that Chen faked the whole attack. She claimed that a nurse working at the hospital Chen and Lu were treated after the attack had called her. During their conversation, the nurse allegedly told the legislator that Chen’s wound would not stem from a gunshot and that security police had already arrived in the morning to prepare for the arrival of Chen and Lu. Apart from making such malicious claims, she instructed the viewers to boycott the ‘illegal’ referenda by yelling at polling station staff: “I refuse to take the ballots for the referenda.” During the TV show, other prominent figures, such as author Li Ao, described President Chen as a ‘slick trickster.’ The claims and insults let to an uproar, especially because of the fact that Chen Wen-chien failed to offer any evidence to substantiate her claims. The next day, incumbent President Chen Shui-bian won the election by a slight margin of 29,518 votes. Voter turnout averaged 80.28 percent, two percentage points lower than in 2000. Compared with the previous election, Chen Shui-bian and Lu Hsiu-lien could garner ten percentage points more votes. In the counties of Nantou and Taichung, the DPP presidential hopeful could increase his share of votes by fifteen percentage points. Lien Chan and his running mate Song Chu-yu lost support in every of the twenty-four counties and cities. Even in the capital, the alliance’s stronghold, the blue camp lost five percentage points.

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

THE AFTERMATH: THE LIEN-SONG REBELLION After the election defeat, Lien Chan and Song Chu-yu made public their intention to file lawsuits nullifying both the election result and the election itself. The first lawsuit based on Article 104 of the presidential election law and aimed at Chen Shui-bian and his running mate Lu Hsiu-lien. Lien and Song accused Chen and Lu of having (i) staged the assassination attempt, (ii) prevented some 300,000 soldiers and police officers from exercising their rights to vote, (ii) holding the two referenda together with the presidential vote as to influence the outcome of the election. The second lawsuit based on Article 102 of the presidential election law and accused the Central Election Commission of vote rigging and of illegally holding referenda together with a national election. The lawsuits seemed to be an appropriate way of dealing with the blue camp’s suspicions. The motives behind the lawsuits are, however, questionable. When Song Chu-yu and Lien Chan were informed about their defeat, they were also made understood that there was no chance of changing the result within the legal and moral boundaries. Song and Lien, nevertheless, decided to continue their media war against Chen Shui-bian. What started as a protest movement during the election campaign soon developed into a rebellion, or even a coup attempt. There was no other option more feasible to Lien Chan and Song Chu-yu than the highly questionable approach of discrediting the president by any possible means. The aim of the rebellion was not to win the lawsuits in court (for it was already clear that the blue camp would lose both suits) but to discredit—if not even topple—the government headed by Chen Shui-bian. It should have been a coup d'état without tanks and soldiers on the streets—a silent coup from within the system against the system.

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Political Parties and Electoral Politics in Taiwan

23

Table 5. Legislative Yuan Election Results (2004 and 2001)

Affiliation Votes DPP 3,447,740 KMT 2,949,371 PFP 1,917,836 TSU 801,560 others 1,211,348 Total 10,327,855

2001 2004 % Seats % Votes % Seats 33.4 87 38.7 3,471,429 35.7 89 28.6 68 30.2 3,190,081 32.8 79 18.6 46 20.4 1,350,613 13.9 34 7.8 13 5.8 756,712 7.8 12 11.7 11 4.9 948,524 9.8 11 100.0 225 100.0 9,717,359 100.0 225

Change % % Votes Seats 39.6 7.0 2.3 35.1 15.0 16.2 15.1 -25.2 -26.1 5.3 0.3 -7.7 4.9 -16.8 0.0 100.0 -

Compiled by the author based on data provided by the Central Election Commission

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

THE 2004 LEGISLATIVE YUAN ELECTION: SIGNS OF CHANGE On 11 December 2004, voters in Taiwan again went to the polls to elect 225 members of the Legislative Yuan. The election result was a victory for the KMT, and a defeat for Song’s PFP and the green camp. Compared with 2001, the KMT succeeded in increasing its share of votes and seats by about 15 percent (see Table 7.5). The DPP and the TSU could not get a majority in parliament. There are two major reasons for this predictable result. Firstly, both parties were too optimistic and nominated far too many candidates. To make things worse, there was no effective cooperation between the two parties. Votes were thus not shared among the hopefuls of the two parties contesting in the same electoral constituency. Secondly, President Chen Shui-bian ran a highly focused campaign predominately addressing national identity issues, which was criticized by the pro-China media and the opposition parties. Stories about a possible war with China if Chen Shui-bian obtained a parliamentary majority were common and seemed to work in favor of the People’s Republic of China and the pro-China forces in Taiwan. Song Chu-yu and his PFP, however, were the biggest losers. The party lost one quarter of the popular support and parliamentary seats it obtained in 2001. Song’s populist rhetoric and the violent post-election protests instigated by his party colleagues seemed to have seriously hurt the image of the party.44 Although the KMT leadership was part of the rebellion against Chen Shui-bian, Lien Chan, other senior party members and legislators were in rare cases involved in the violent protests following the disputed March election, thus giving the electorate the impression of a more reasonable party. Song’s defeat was predictable when the KMT first agreed but later postponed a merger between the KMT and the PFP until after the election. The election result, however, was too impressive and the KMT put merger plan on ice. After this disgraceful experience, Song Chu-yu was looking for a new partner. At the end of December 2004, there were several indications for a possible cooperation between the DPP

44

The Advocates, a local think tank, released an opinion poll in mid-June revealing that 67.8 percent of the electorate strongly opposed the protests of the blue camp. Polls taken in March showed that about 52.9 percent opposed the protests. (18 June 2004).

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

24

Christian Schafferer

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

and PFP.45 Will this cooperation lead to a depolarization of Taiwan’s current polarized party system?

45

Kao Tian-sheng, “Song chu yu ke wang chong hui ‘zhu liu pai’?” [Song Chu-yu back to the mainstream again?] New Taiwan Weekly, 15-21 January 2005, 11-16.

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

In: Taiwan: Economic, Political and Social Issues Editor: L. N. Roberts and C. R. Lewis

ISBN 978-1-60456-808-0 ©2008 Nova Science Publishers, Inc.

Chapter 2

CAN THE DEVIATION BETWEEN OWNERSHIP AND CONTROL EXPLAIN BANK PERFORMANCE IN TAIWAN? Chung-Hua Shen1, Chien-An Wang2 and Meng-Wen Wu1 1

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Department of Money and Banking National Chengchi University 2 Department of Banking and Finance National Chi-Nan University

ABSTRACT This chapter uses ownership data of Taiwan banks to investigate whether the deviation between ownership and controlling power is informative enough at an earlier stage to allow us to predict bank failure and bank performance. The gap between ownership and controlling power is proxied by voting rights (=voting rights/cash flow rights), as suggested by Claessens et al. (2000). This ratio is also referred to as moral hazard I (MH I here). For comparison and to reflect the unique local banking environment, we develop and test a seat-control rights ratio (=seat-control rights/cash flow rights), which we refer to here as moral hazard II (MH II). Both moral hazard indices are useful [precursors] in terms of predicting returns (return on assets), but MH II outperforms MH I in predicting risk (nonperforming loans). In other words, in the case of Taiwan, MH II provides more useful information than MH I.

Keywords: voting rights, seat-control rights, bank governance, bank performance, moral hazard indices

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

26

Chung-Hua Shen, Chien-An Wang and Meng-Wen Wu

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

1. INTRODUCTION Whether or not the soundness of corporate governance of a firm has an effect on its performance has recently been the subject of considerable debate. 1 While a wealth of studies has investigated corporate governance and corporate performance, research into the effect of bank governance on its performance has been relatively scant. But, it is important to note that bank governance is distinctly different from corporate governance. On this point, Adams and Mehran (2003), for instance, find that bank holding companies (BHCs) and manufacturing firms differ in terms of board size, the percentage of outside directors, ownership and the ratio of stock options to salary plus bonus compensation.2 Macey and O’Hara (2003) make a distinction between governance in banking industry and that in non-banking firms, and argue bank governance is more closely aligned to the Franco-German model than the AngloAmerican model. The former puts an emphasis on the interests of long-term stakeholders, including shareholders, bank employees and depositors; by contrast, the latter focuses exclusively on shareholders. This means that in the bank governance model, bank directors and officers are should be highly committed to ensuring the safety and soundness of banks. They must also explicitly and systemically take solvency risk into account. There is every reason to believe that bank governance is fundamental to the risk and safety of banks, and as such, the relationship between bank governance and bank performance is an interesting though a relatively unexplored issue. The purpose of this paper is to determine whether or not bank governance is typically an important factor when explaining bank default. In the non-banking industry, corporate governance focuses more on “agency problems” which refer to conflicts among owners and managers (Demsetz, Saidenberg and Strahan, 1999).3 In this regard, Claessens, Djankov and Lang (2000) and La Porta, Lopez-de-Silanes, and Vishny (2002, henceforth LLSV) suggest calculating the ratios of “cash-flow rights” to “voting rights” as the proxy for “one-share, onevote”. In Claessens et al’s study of non-banking firms in nine Asian countries, voting rights frequently exceed cash-flow rights via pyramidal control structures and cross-holdings. The tendency for firms in Asia to have concentrated control ownership coupled with dispersed cash-flow distribution rights often creates agency problems, in that the interests of controlling shareholders and minority shareholders are not perfectly aligned. In the banking industry, governance has been more closely linked to the “moral hazard” problem because capital structure in banking firms is more heavily financially leveraged compared to that in non-financial firms. That is, when banks lack net worth, managers have little to lose from a bad outcome but very much to gain from a good outcome.4 Accordingly, 1

While Klapper and Love (2002) report that corporate governance is correlated with operating performance, Coles, McWilliams and Sen (2001) do not find such a strong relationship. And, exploring a slightly different issue, Gompers, Ishii and Metrick (2003) contend that governance is positively related to stock returns. 2 They find that board size and the percentage of outside directors are both significantly larger in BHCs than in manufacturing firms. 3 The agency problem arises because the role and functioning of the modern firm is based on the assumption that it has a widely-dispersed ownership structure. Recently, researchers have begun to question this assumption and argue that even among the largest American firms, there is at least modest concentration. See Claessens, Djankov and Lang (2000) and Holderness (2003) for the case of the U.S. and Yeh, Lee and Woidtke (2005) for the case of Taiwan. 4 Banks also suffer from the problem of mismatch in the term structure and liquidity of their assets and liabilities, i.e., bank assets are usually illiquid, while their liabilities are liquid.

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Can the Deviation between Ownership and Control Explain Bank Performance…

27

while it is often argued that bank failure or bad performance can be attributed to business practices within banks, such as connected lending, overcapacity, unsound accounting and auditing practices, insufficient internal controls and poor management, among others,5 the most oft-cited cases are the consequence of the severe moral hazard problem.6 Two questions arise. First, does distressed bank have a more severe moral hazard problem than nondistressed bank? Second, does the moral hazard of a bank have a negative effect on its NPL (Non-performing Loan) ratio but a positive effect on its ROA(Return of Asset) ratio? Although the moral hazard problem in the banking industry is well reported throughout the literature, systematic studies on exactly how the problem affects the performance of banks have far too long been overlooked, and this is particularly true in emerging countries. This is in large part because it is not easy to construct the “cash-flow rights” and “voting rights” of banks in emerging markets on account of the unavailability of appropriate data. In light of the difficulty in obtaining the necessary data, we target our analysis on one country, Taiwan, which is an ideal arena for the scope of this empirical research as some banks were in severe distress immediately after the 1997 Asian financial crisis whereas others were still in good shape. This provides us with an arena in which to investigate and compare the relationship between bank governance and bank performance. In this paper, the cash-flow rights and voting rights of 33 banks listed on the Taiwan Stock Exchange Council (TSEC) are calculated. To answer the first question, i.e., whether a distressed bank has a more severe moral hazard problem, the performance of banks is measured using CAMEL7 commonly to assess a bank’s overall financial status and its degree of compliance with safety and soundness covenants.8 Because the values of the five components are mainly taken from financial statements, they are typically evaluations of the performance of banks. In this study, the ratio of voting rights to cash-flow rights is employed to account for the effect of the moral hazard on bank performance, as reflected in the CAMEL indicators. Because this ratio is widely accepted by academia the world over, it is referred to as the Moral Hazard Index I (hereafter MH I) in this paper. Besides MH I, the seat-control rights ratio is also calculated on the basis of Taiwanspecific features. In Taiwan, like in many other emerging countries, bank boards are mainly

5

Literature that uses micro data abounds. See, for example, Thomson (1991), Barker and Holdsworth (1993), Berger, Davies and Flannery (2000), Cole and Gunther (1998), DeYoung, Flannery, Lang and Sorescu (1998), Hirtle and Lopez (1999), Berger and Davies (1998), DeYoung, Hughes and Moon (2001) as well as Gilbert, Meyer and Vaughan (1999, 2002), among others. 6 Demsetz, Saidenberg and Strahan (1999) point that in non-financial corporations, limited liability gives bank shareholders an incentive to expropriate wealth from bondholders by increasing risk. Since the government protects bondholders (particularly depositors) from the consequences of bank risk taking, their incentive to monitor and constrain risk taking is weak. The absence of such discipline is known as “moral hazard” or the “moral hazard problem associated with deposit insurance” in the banking literature. 7 On July 9, 1996, the Federal Financial Institutions Examination Council (FFIEC) proposed revisions to the agency's Uniform Financial Institutions Rating System, commonly referred to as the "CAMEL" rating system. The CAMEL rating system is an internal supervisory tool for evaluating the soundness of financial institutions on a uniform basis and for identifying those institutions requiring special supervisory attention or concern. (C) means capital, (A) means asset quality, (M) means management or administration, (E) means earnings, and (L) means liquidity. 8 Literature attesting to the effectiveness of using CAMEL abounds. See, for example, Thomson (1991), Barker and Holdsworth (1993), Berger, Davies and Flannery (2000), Cole and Gunther (1998), DeYoung, Flannery, Lang and Sorescu (1998), Flannery (1998), Hirtle and Lopez (1999), Berger and Davies (1998), DeYoung, Hughes and Moon (2001) and Gilbert, Meyer and Vaughan (1999, 2002), among others.

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

28

Chung-Hua Shen, Chien-An Wang and Meng-Wen Wu

comprised of family-controlling shareholders, or representatives of a legal entity,9 and the daily operations and management of banks are customarily controlled by “standing directors” who are re-elected from members of the board and who are typically fewer than five persons in number. It is true that controlling shareholders could own fewer shares, but equally possibly control more than half of the seats -- for example, three members of the standing directors. These seat-control rights by way of these “standing directors” may be based on a strategic alliance between two groups, an exchange of mutual interests, mutual benefits or political reasons, and so on. Meanwhile, standing directors might own just a few shares but still control an entire bank, perhaps therefore giving them the incentive to undertake riskier projects and related lending. In light of these possibilities, it is clearly apparent that there is a need for a second moral hazard index: seat-control rights over cash flow rights, which we refer to here as the “seatcontrol rights ratio” or Moral Hazard Index II (hereafter, MH II). The need for this is further strengthened by the fact there have been numerous scandals involving standing directors of financial institutions who hold only a small percentage of the shares of a bank but are still able to decide who obtains funds. In these cases, the principal of arm’s length lending was regularly being ignored. To obtain our MH II, the seat-control rights ratio is calculated using seat control rights over cash flow rights, and the numerator is the geometric average of the following three percentages: the seats of the controlling general directors/ the seats of all general directors, the seats of the controlling standing directors/ the seats of standing directors and the seats of all controlling directors/the seats of all directors. The two moral hazard indices, MH I and MH II, are then used to predict bank performance which is mostly proxied by the components of CAMEL.10 Valuable insight is gained from our results. As for the answer to our first question, it appears that the relationship between the moral hazard and bank performance is asymmetric from the results of the basic statistics. That is, distressed banks evidently belong to the group with the highest moral hazard indices, regardless of their moral hazard measures. In response to our second question, it appears that both indices are effective predictors of future returns but that MH II outperforms MH I in predicting non-performing loans from the results of our regressions.11 The remainder of this paper is organized as follows. Section 2 presents an overview of the corporate and banking governance environment of Taiwan. Section 3 provides the sources of the data. Section 4 investigates the international and local moral hazard indices and the basic statistics. Section 5 highlights the results of the empirical analysis and analyzes how well banking failures can accurately be predicted using the two moral hazard indices. Section 6 concludes.

9

Corporate governance in the banking industry in Taiwan is discussed in detail in Section 2. We employ (C) means capital, proxied by capital adequacy ratio (CAR), (A) means asset quality, proxied by non-performing loan ratio (NPL ratio) and standard deviation of return of asset ratio(σROA ratio), (M) means management or administration, proxied by (In)efficiency ratio and Burden ratio, (E) means earnings, proxied by return of asset ratio(ROA ratio) and return of equity(ROE ratio), and (L) means liquidity, proxied by liquidity ratio. 11 For example, Morck et al., (1988) found a significant non-linear relation between the percentage of shares owned by managers and corporate value, which is positive, negative and positive for three different stages of managerial ownership, reflecting the alignment effect and the entrenchment effect. Chen and Steiner (2000) and Davies and Hillier and McColgan (2005) find similar results. 10

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Can the Deviation between Ownership and Control Explain Bank Performance…

29

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

2. BANK BOARDS IN TAIWAN Based on their study of the ownership of banks around the world, LLSV (2002) find that government ownership of banks is widespread as recently as the 1990s largely because the financial industry has historically been heavily regulated by the authority. A similar phenomenon prevailed in the majority of Taiwan’s local banks prior to the passage of Taiwan’s Amendment to the Banking Law in 1989. To illustrate, a little more than half of twenty-four banks were state-owned or family-controlled at the end of 1989.12 In the 1990s, the Ministry of Finance (MOF) deregulated the restrictions on bank charters, thereby allowing for the establishment of new commercial banks. The minimum amount of paid-in capital for the establishment of a new commercial bank was set at NT$10 billion (U.S. $330 million). To prevent new commercial banks from being controlled by interest conglomerates whose main purpose could be to pursue personal interests, three layers of restrictions on shareholders are stipulated in the Banking Law. First, no shareholder, either a natural or legal entity, can hold more than 5% of all outstanding shares. Second, joint inter-related shareholders13 cannot hold more than 15% of total shares, and third, at least 20% of total shares must be issued to the general public. With further deregulation in 1992, sixteen new commercial banks were granted the right to establish in compliance with these regulations. These regulations, however, stipulate very little about pyramidal control structures and cross-holdings. In Taiwan, the board structure of corporations and banks are, in a broad sense, modeled after their German counterparts. Companies are comprised of two separate organizations -- a board of directors and a board of supervisors -- where directors manage the company and supervisors monitor the directors. What this means is that crucial managerial decisions are decided by directors or by the votes of directors during board meetings and that supervisors mostly participate in decision-making or the voting process. The minimum number of directors required is three for publicly-issued but not listed companies and five for listed companies. All directors may be elected as standing directors,14 and boards of directors and 12

The term ‘government-owned banks under direct control’ means the government can prohibit them from being part of any concentration of private economic power. At the end of June 2000, there were 31 commercial banks, and among these, three had been established during Japanese occupation (The First Commercial Bank, The Hua-Nan Commercial Bank and The Chang-Hwa Commercial Bank), two had relocated to Taiwan from mainland China (The International Commercial Bank of China and The Shanghai Commercial and Savings Bank), two had been established before 1990 in Taiwan (The United World Chinese Commercial Bank and The Overseas Chinese Commercial Bank), 6 are small and medium enterprise banks, 17 were established after 1991, and one was reorganized from an investment and trust company in 1992 (Chinatrust Commercial Bank). After 1997, five credit cooperatives were re-organized into five commercial banks. Our sample contains all of the above banks, which are owned by the private sector and are listed on the TSEC or the Over-TaiwanCounter Exchange (OTC). This will be discussed in the Data section. 13 According to Rule 25 of the Amendment to the Banking Law, joint inter-related shareholders includes shareholders with more than 15% of total shares, spouse, first and second generation relatives, and any other corporation operated by the shareholder or his/her spouse. 14 In a general case, authorities of board of directors by the Taiwan Corporate Law involves the following: (1) determining the agendas of the member conventions; (2) examining membership qualifications and imposing membership suspensions; (3) electing or recalling “standing directors”, vice-chairman and chairman of board of directors; (4) investigating and resolving the resignation of directors, standing directors, vice-chairman and chairman of board of directors; (5) examining the establishment of various committees and task forces; (6) employing and dismissing work personnel; (7) making annual plans, reports, budgets and final accounts; and (8) overseeing other operations. Then, five standing directors are elected from the directors, one chairman and one vice-chairman of the board are elected out of the five standing directors by the directors. See the Corporate Law of Taiwan.

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

30

Chung-Hua Shen, Chien-An Wang and Meng-Wen Wu

supervisors in Taiwan are parallel organizations, which means that decisions made by directors do not require the a priori approval of supervisors. Also, Taiwan’s Corporate Law stipulates that shareholders shall elect both supervisors and directors and that only current shareholders qualify as candidates. Since supervisors are also shareholders and, more importantly, are often from subsidiaries of the same interest group, they often ignore their designated job of monitoring the directors. Thus, it is not unusual to find that family members of controlling shareholders serve as supervisors.15 Several Taiwan bank scandals have been related to connected lending and have been intertwined with board structure. In line with Sections 23A and 32B of the Federal Reserve Act in the U. S., Taiwan’s Amendment to the Banking Law sets stringent limits on interrelated lending. For example, a bank can loan to the relatives of board members up to 5% of its total lending, but it must have the equivalent collateral as back-up. In practice, such restrictions on inter-related lending have loopholes, such as the lack of a clear definition of what is meant by ‘related-party’. For example, lending to a subsidiary is restricted, but lending to a grandson’s subsidiary is not. In other words, these regulations do not take complex pyramidal control structures and cross-holdings into account, making arm’s length lending superficial.

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

3. SOURCES OF THE DATA Our sample comprises 33 publicly listed, commercial banks in Taiwan, and sample period covers the 1998-2000 periods. We use three sources of data. For a preliminary assessment of board composition, we first compile the names of all shareholders of each listed bank from company prospectuses. We then investigate the parent company and its subsidiaries using Business Groups in Taiwan, an annual publication by the China Credit Information Services LTD. This information enables us to identify the controller of each bank. If the controlling shareholders are the institutional shareholders, we further identify the owner of the institution. The data source is the Taiwan Economic Journal (a private data vendor in Taiwan, hereafter TEJ). We determine each institution’s pyramidal control structures and cross-shareholding practices and calculate control rights, seat-control rights and cash-flow rights. Finally, we take the year-end financial ratios that are applicable to CAMEL from each bank’s balance sheets and income statements, both of which are available from the TEJ.

4. MORAL HAZARD INDICES AND BASIC STATISTICS 4.1. Controlling Shareholders First, we define who the controlling shareholders are because our Moral Hazard Indices are based on the concept of ‘one-share, one-vote’ for each controlling shareholder. Following the study of Claessens et al. (2000), ownership by a family group, rather than a single 15

With regard to the recent heated discussion as to whether a board should include independent directors and supervisors, Taiwan is also enacting a similar law. The drafting of the law, however, takes time and is not finished at the time of writing this paper.

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Can the Deviation between Ownership and Control Explain Bank Performance…

31

individual, serves as the unit of analysis, where ‘family group’ is defined as a group of people that are related through blood or marital ties. Following Claessens et al. (2000), we calculate ownership on the basis of cash-flow rights and controlling power on the basis of voting rights. We trace the chain of ownership and use pyramidal control structures and cross-holdings to distinguish between voting rights and cash-flow rights (see Appendix for details). We then define the controlling shareholders or family group that are found to have the largest controlling power based on the sum of direct and indirect voting rights. In the study of La Porta et al. (1999), when a firm is owned by both a family and a widely-held corporation, each with 15% of the voting rights, it is referred to as a ‘widely-held firm’. In their study of firms in nine Asian nations, Claessens et al. (2000) suggest that 20% be used when determining final ownership, but Yeh and Lee (2000) contend that 15% may be more appropriate for Taiwan corporations. Because the largest shareholders usually are the controlling shareholders in Taiwan’s banks, we can easily identify the controlling shareholders.

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

4.2. Cash-Flow Rights, Seat-Control Rights and Voting Rights Here we explain the calculation of voting rights, cash-flow rights, and seat-control rights. According to La Porta et al. (1999) and Claessens et al. (2000), ownership can be defined simply as cash-flow rights, whereas controlling power can be defined as voting rights. In the event that voting rights are highly leveraged via pyramid schemes or cross-shareholding, it leads to an ever-increasing difference between cash-flow rights and voting rights. This study follows the concept of ultimate owners, as proposed by Claessens et al. (2000), and it meticulously traces chains of ownership. Calculating this involves the following steps. First, we identify the ultimate owners, referred to as the controlling shareholders, who, as the term suggests, have the greatest controlling power. We do this by summing the “direct” and “indirect” voting rights of each shareholder. We count “direct” voting rights on the basis of the number of shares registered in the name of the ultimate owners (including family members), while we count “indirect” voting rights on the basis of the number of shares held by entities that are, in turn, controlled by the ultimate owners. We provide an example of our calculation in the Appendix. In the second step, we calculate the two moral hazard indices using the data of the complete chain of ownership. MH I takes voting rights as numerator, as defined by Claessens et al. (2002), and the other, and MH II, takes seat-control rights as numerator, as formulated in the present paper. More specifically, Moral Hazard I = voting rights /cash-flow rights Moral Hazard II = seat-control rights/cash-flow rights where voting rights = direct holdings + indirect holdings; indirect holdings are equal to Σ(min (chains of each controlling share)); seat-controlling rights = geometric average (the seats of the controlling general directors/all general directors; the seats of the controlling standing directors/all standing directors; the seats of all controlling directors/all directors ); and cash-

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

32

Chung-Hua Shen, Chien-An Wang and Meng-Wen Wu

flow rights = direct holdings + indirect holdings, where indirect holdings are equal to Σ(multiplications of chains of controlling shares).

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

4.3. Basic Statistics of the Total Sample Table 1 presents the procedures for the calculation of MH I and MH II for the 33 TSElisted banks during the 1998-2000 period, which comprise 5 distressed and 28 non-distressed banks. The name of each bank is reported below the table due to space constraints. The first column identifies the type of controlling shareholder, where 19 of the 33 are family-owned, 10 are government-owned and the remaining four are separated. Accordingly, the lion’s share of Taiwan banks during the sample period is family-owned. As shown in Table 1, distressed banks are not restricted to any specific type of bank but are evenly dispersed among familyowned (two banks), government-owned (one bank) and others (two banks). Columns (1) to (3) give direct- and indirect- voting rights and their totals, and columns (4) to (6) give direct and indirect cash-flow rights and their totals. Column (7) presents the results of Moral Hazard I, and columns (8a) to (8c) present the seat-control rights, where we calculate the number of seats of the controlling shareholders held by the three types of directors. Column (9) presents the results of Moral Hazard II, and it is abundantly clear that it has a much wider dispersion than MH I. We conjecture that this is probably because pyramidal control structures and cross-holdings are less severe in the Taiwan’s banking industry, making most MH I close to unity. As shown in Table 1, the means of MH I and MH II are 1.076 and 1.734, respectively. Table 2 compares the performance and the moral hazard indices of the two groups of banks, i.e., the distressed (Group A) and non-distressed banks (Group B). The average values of MH I for Group A and Group B are respectively 1.635 and 1.254, and because the t-test is equal to 31.90, the difference is significant. Thus, distressed banks have a somewhat higher MH I. With the average value of MH II for Group A and Group B respectively, 7.283 and 2.638, the difference is also significant. The gap between the two groups is even larger in the case of MH II. It is clear that the number of non-performing loans ratio is substantially higher for distressed banks (15.154) than for non-distressed banks (4.020). As regards the return on assets ratios (ROA ratios), though the gap between Group A (–2.030) and Group B (0.385) is large, it is not significant. Table 3 divides MH I in three panels by the cut-off points of 1.0 and 1.345, and MH II into three panels by the cut-off points of 1.851 and 3.979, which is nine panels totally. The cut-offs are designated to be at the 33 and 67 percentile, respectively, to equally divide the sample into three segments. To test for sensitivity, different cut-offs are used, but the results do not change qualitatively. Thus, the nine panels in Table 3 have MH I as the horizontal axis and MH II as the vertical axis. The banks in Panel 1, where MH I is lower when voting rights are roughly equal to cashflow rights and MH II is also lower, have the least severe agency problem. This stands in stark contrast to those banks in Panel 9, where MH I is higher when bank voting rights and cash-flow rights are the farthest apart and MH II is also higher. Undoubtedly, these banks have the most severe agency problem. In Panel 9, where banks that have the most severe agency problem, four of the seven are distressed banks, (i.e, (A1)

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook Central,

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Table 1. Descriptive Statistics of the Two Moral Hazard Indices for All Banks Bank List1,2

Types of controlling shareholder3

A1 A2 A3 A4 A5 A6 A7 A8 A9 A10 A11 A12 A13 A14 A15 A16 A17 A18 A19 A20 A21 A22 A23 A24 A25 A26

Other Type Other Type Family Family Government Government Government Government Government Family Family Family Other Type Family Government Government Government Family Family Government Family Government Family Family Family Family

(1) Direct voting rights 0.00 3.21 1.17 3.38 0 25.65 37.25 38.15 44.75 0.71 6.13 1.73 53.01 0.67 49.96 48.04 13.70 2.95 0.54 58.72 3.03 38.33 0.00 5.60 6.97 0.28

(2) Indirect voting rights 15.46 9.71 2.76 4.35 10.26 12.00 9.87 8.85 0.00 9.77 1.42 3.99 0.77 9.50 0.00 3.05 15.98 18.18 8.64 0.00 43.28 54.12 44.26 17.61 49.38 35.45

(3) Total voting rights= (1)+(2) 15.46 12.92 3.93 7.73 10.26 37.65 47.12 47.00 44.75 10.49 7.56 5.72 53.79 10.17 49.96 51.10 29.68 21.14 9.18 58.72 46.30 92.45 44.26 23.21 56.35 35.73

(4) Direct cash-flow rights

(5) Indirect cash- flow rights

0.00 3.21 1.17 3.38 0 25.65 37.25 38.15 44.75 0.71 6.13 1.73 53.01 0.67 49.96 48.04 13.70 2.95 0.54 58.72 3.03 38.33 0.00 5.60 6.97 0.28

10.68 4.58 2.76 0.96 4.656 2.56 2.54 1.77 0.00 9.77 0.95 3.99 0.05 9.50 0.00 0.51 7.33 12.84 6.89 0.00 31.58 9.16 17.61 7.18 49.38 26.29

(6) Total cash-flow rights= (4)+(5) 10.68 7.78 3.93 4.34 4.656 28.21 39.78 39.91 44.75 10.49 7.08 5.72 53.06 10.17 49.96 48.56 21.02 15.79 7.43 58.72 34.61 47.48 17.61 12.78 56.35 26.57

(7) Moral I= (3)/(6)

(8a) Director board 4

(8b) Director board 4

(8c) Director board 4

(9) Moral II = 5

1.45* 1.67* 1.00 1.86* 2.20* 1.33* 1.18 1.18 1.00 1.00 1.07 1.00 1.01 1.00 1.00 1.06 1.41* 1.38* 1.24 1.00 1.34* 1.95* 2.51* 1.83* 1.00 1.34*

0.29 0.84 0.33 0.35 0.13 0.50 0.72 0.70 0.78 0.23 0.26 0.14 0.90 0.17 0.29 0.57 0.53 0.67 0.27 1.00 0.46 0.93 0.44 0.79 0.75 0.47

0.67 1.00 0.39 0.17 0.37 0.53 0.76 0.64 0.93 0.32 0.47 0.27 0.89 0.67 0.93 0.84 0.31 0.83 0.33 1.00 0.80 1.00 0.50 0.65 0.78 0.67

0.41 0.88 0.35 0.30 0.19 0.51 0.73 0.68 0.84 0.26 0.33 0.19 0.90 0.31 0.58 0.70 0.46 0.72 0.29 1.00 0.59 0.95 0.46 0.74 0.74 0.54

4.03* 11.62* 9.05* 6.02* 4.49* 1.82 1.85 1.69 1.89 2.55* 4.84* 3.37* 1.69 3.23* 1.08 1.43 2.01 4.67* 3.98* 1.7 1.74 2.02 2.65* 5.67* 1.34 2.09

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest Ebook Central,

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Table 1. (Continued) Bank List1,2

Types of controlling shareholder3

A27 A28 A29 A30 A31 A32 A33

Family Family Separated Family Family Family Family

(1) Direct voting rights 1.48 11.26 3.16 2.42 3.44 1.69 12.22

(2) Indirect voting rights 5.02 29.24 0.00 21.67 35.18 7.60 23.99

(3) Total voting rights= (1)+(2) 6.50 40.50 3.16 24.09 38.62 9.29 36.21

(4) Direct cash-flow rights

(5) Indirect cash- flow rights

1.48 11.26 3.16 2.42 3.44 1.69 12.22

2.90 22.67 0.00 21.63 21.00 7.60 23.99

(6) Total cash-flow rights= (4)+(5) 4.38 33.93 3.16 24.05 24.44 9.29 36.21

(7) Moral I= (3)/(6)

(8a) Director board 4

(8b) Director board 4

(8c) Director board 4

(9) Moral II = 5

1.50* 1.19 1.00 1.00 1.58 1.00 1.00

0.16 0.63 0.00 0.47 0.54 0.34 0.46

0.31 0.56 0.00 1.00 1.00 0.48 0.78

0.21 0.62 0.00 0.60 0.59 0.37 0.51

4.99* 1.78 0 2.73 2.79 4.22* 1.57

Notes: 1. There are five distressed banks in the sample: (A1) Kao-Hsiung Small and Medium Enterprise Bank; (A2) Tai-Dong Small and Medium Enterprise Bank; (A3) Tai-Chung Bank; (A4) Chung-Hsin Bank; and (A5) Oversee Chinese Bank.2. 2. There are 28 regular banks in the sample: (A6) Chang-Hwa Bank; (A7) First Bank; (A8) Hua-Nan Bank; (A9) ICBC Bank; (A10) Hsin-Chu Bank; (A11) Taipei Commercial Bank; (A12) Tainan Commercial Bank; (A13) Kao-Hsiung Bank; (A14) Chinatrust Commercial Bank; (A15) Farmers Bank; (A16) Chiao-Tung Bank; (A17) United World Chinese Commercial Bank; (A18) Grand Commercial Bank; (A19) DaAn, Bank; (A20) Taipei City Bank; (A21) The Chinese Bank; (A22) Taiwan Cooperative Bank; (A23) Cathay Bank; (A24) Cosmos Bank; (A25) Union Bank; (A26) Sino-Pac Bank; (A27) E-Sun Bank; (A28) Fubon Bank; (A29) Pan-Asia Bank; (A30) Taishin Bank; (A31) FarEastern Bank; (A32)Ta-Chong Bank; and (A33) En-Tie Bank. 3. Types of controlling shareholder: family, government, separated ownership and no significant controlling shareholder and other (for investment or political party). 4. Types of board of directors: (8a) the seats of the controlling general directors/all general directors; (8b) the seats of the standing controlling standing directors/all standing directors; and (8c) the seats of all controlling directors/ all directors. 5. In (7) and (9), “*” indicates that voting rights are larger than cash-flow rights. The only difference between MH II and MH I is the numerator which is the geometric averaging of (8a), (8b) and (8c) for MH II.

Can the Deviation Between Ownership and Control Explain Bank Performance…

35

Kao-Hsiung Small and Medium Enterprise Bank; (A2) Tai-Dong Small and Medium Enterprise Bank; (A4) Chung-Hsin Bank; and (A5) Oversee Chinese Bank.). The fifth distressed bank (A3) Tai-Chung Bank falls in Panel 7. Table 2. Basic Statistics of the Moral Hazard Indices of the Two Groups

Moral Hazard (voting rights/cash flow rights)

Moral Hazard (seat-control rights/cash flow rights)

Ratio of Non-Performing Loans (NPL ratio)

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

ROA ratio

Group A: Distressed Banks1

Group B: Non-distressed Banks

t-test 2

Average

1.635

1.254

31.900***

Std

0.451

0.359

Max

2.204

2.511

Min

1.000

1.000

Average

7.283

2.638

Std

3.071

1.409

Max

11.676

5.752

Min

4.319

0.000

Average

15.154

4.020

Std

5.286

2.212

Max

23.550

9.813

Min

10.137

0.960

Average

-2.030

0.385

Std

1.766

0.421

Max

-0.410

1.100

Min

-4.750

-0.767

13.665***

9.943***

0.144

*, ** and *** indicate significance at the 10%, 5% and 1% level, respectively.

Table 3. Banks in Different Panels

1.851 > Moral Ⅱ

MoralⅠ＝1

1.345≧MoralⅠ＞1.000

MoralⅠ＞1.345

Panel 1 (A25)Union Bank (Family) (A29)Pan-Asia Bank (Separated) (A33)En-Tie Bank (Family)

Panel 2 (A6)Chang-Hwa Bank (Government) (A8)Hua-Nan Bank (Government) (A13) Kao-Hsiung Bank (Other) (A15)Farmers Bank (Government) (A28) Fubon Bank (Family) (A20)Taipei City Bank (Government) (A21) The Chinese Bank (Family)

Panel 3 (A6)Chang-Hwa Bank (Family) (A16)Chiao-Tung Bank (Government)

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

36

Chung-Hua Shen, Chien-An Wang and Meng-Wen Wu Table 3. (Continued)

3.979≧MoralⅡ≧1.851

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

MoralⅡ＞3.979

MoralⅠ＝1

1.345≧MoralⅠ＞1.000

Panel 4 (A9)ICBC Bank (Government) (A10)Hsin-Chu Bank (Family) (A12)Tainan Commercial Bank (Family) (A14)Chinatrust Commercial Bank (Family) Panel 7 (A3) Tai-Chung Bank (Family) (A19)Da-An, Bank (Family) (A32)Ta-Chong Bank (Family)

Panel 5 (A7)First Bank (Government) (A23) Cathy Bank (Family) (A26)Sino-Pac Bank (Family) (A30)Taishin Bank (Family)

Panel 8 (A11)Taipei Commercial Bank (Family)

MoralⅠ＞1.345 Panel 6 (A17) United World Chinese Commercial Bank (Government) (A22)Taiwan Cooperative Bank (Government) (A31)Far-Eastern Bank (Family)

Panel 9 (A1) Kao-Hsiung Small and Medium Enterprise Bank (A2)Tai-Dong Bank (Other) (A4)Chung-Hsin Bank (Family) (A5)Oversee Chinese Bank (Government) (A18)Grand Commercial Bank (Government) (A24)Cosmos Bank (Family) (A27)E-Sun Bank (Family)

Note: The cut-off points are based on actual data.

It should be noted that all non-distressed banks do not necessarily fall in Panel 9. In similar manner, three banks that are non-distressed banks also fall in Panel 9 with the most severe agency problem (i.e., A18 (Grand Commercial Bank, A24 (Cosmos Bank), and A27 (E. Sun Bank)) Accordingly, and this is interesting, there seems that banks with the highest moral hazard indices may not be in distress but that distressed banks mostly likely have the highest moral hazard indices. Table 4 again divided into 9 panels by the same criteria of MH I and MH II in Table 3, and presents the average CAMEL in each panel. There are several interesting findings. First, the capital adequacy ratio (CAR), the proxy for the (C) component of CAMEL, is almost the same across the nine panels, likely because the minimum ratio of 8% is required for all banks. Next, as concerns the other components, the highest NPL ratio of 10.636 (proxy for asset quality (A)), the highest (In)efficiency of 1.543 and the highest Burden Ratio of 0.023 (proxy for management or administration (M)), the lowest ROE ratio of -12.233 and ROA ratio of 1.233 (proxy for earnings (E)) all fall in the panel with the most severe agency problem (Panel 9). Third, the lowest NPL ratio (1.610 in Panel 5), the lowest (In)efficiency ratio (In)efficiency (0.690 in Panel 8), the lowest Burden Ratio (0.006 in Panel 8) and the highest

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Can the Deviation Between Ownership and Control Explain Bank Performance… 37 ROE and ROA ratios (8.787 in Panel 8 and 0.739 in Panel 5, respectively) are not in Panel 1 with the least severe agency problem. Table 4. Moral Hazard Indices and CAMEL MoralⅠ MoralⅡ

MoralⅠ＝1 C A

1.851＞

M

MoralⅡ

E L

C A 3.979≧MoralⅡ

M

≧1.851

E

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

L

C A MoralⅡ ＞3.979

M E L

Panel 1 CAR 10.513 NPL ratio 4.406 σROA ratio -0.216 (In)efficiency ratio 0.813 Burden ratio 0.009 ROE ratio 5.140 ROA ratio 0.245 Liquidity Ratio 4.918 Panel 4 CAR 11.407 NPL ratio 5.191 σROA ratio -0.085 (In)efficiency ratio 0.938 Burden ratio 0.013 ROE ratio 2.059 ROA ratio 0.405 Liquidity Ratio 13.551 Panel 7 CAR 10.008 NPL ratio 8.194 σROA ratio -0.254 (In)efficiency ratio 1.048 Burden ratio 0.016 ROE ratio -1.588 ROA ratio -0.100 Liquidity Ratio11.580

1.34≧MoralⅠ＞1 C A M E L

C A M E L

C A M E L

Panel 2 CAR 11.452 NPL ratio 4.009 σROA ratio -0.005 (In)efficiency ratio 0.824 Burden ratio 0.008 ROE ratio 0.732 ROA ratio 0.456 Liquidity Ratio 9.284 Panel 5 CAR 11.31 NPL ratio 1.610 σROA ratio 0.094 (In)efficiency ratio 0.691 Burden ratio 0.008 ROE ratio 8.157 ROA ratio 0.739 Liquidity Ratio 15.163 Panel 8 CAR 11.043 NPL ratio 2.905 σROA ratio -0.175 (In)efficiency ratio 0.690 Burden ratio 0.006 ROE ratio 8.787 ROA ratio 0.358 Liquidity Ratio 15.095

MoralⅠ＞1.34 Panel 3

C A M E L

C A M E L

Panel 6 CAR 11.763 NPL ratio 4.203 σROA ratio -0.266 (In)efficiency ratio 0.734 Burden ratio 0.007 ROE ratio 6.527 ROA ratio 0.269 Liquidity Ratio 17.717 Panel 9 CAR 12.312 NPL ratio 10.636 σROA ratio -0.713 (In)efficiency ratio 1.543 Burden ratio 0.023 ROE ratio -12.233 ROA ratio -1.233 Liquidity Ratio 14.644

CAR is CAR (the capital adequacy ratio); NPL ratio is the non-performing loan ratio; (In)efficiency ratio// is the ratio of non-interest expense to net interest income plus non-interest income; Burden ratio is the ratio of non-interest expense minus non-interest income to average asset; ROE ratio is the ratio of net income after tax to equity; ROA ratio the ratio of net income after tax to total asset; Liquidity ratio is the ratio of liquid asset to deposit.

Thus, again, banks with the lowest moral hazard indices may not perform better than banks with higher moral hazard indices, but equally important, banks with the highest moral hazard indices typically perform the worst with regard to the CAMEL elements of asset quality, management skills and earnings. Our two moral hazard indices provide the “necessary conditions” to predict the performance of individual banks. Thus, the answer to our first question is that most distressed banks do indeed have a more serious moral hazard problem and worse performance than non-distressed banks. The basic results presented here

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

38

Chung-Hua Shen, Chien-An Wang and Meng-Wen Wu

may be affected by other factors, such as policy changes and economic conditions, among others. The regression analysis in the next section eliminates this weakness. We also employ time dummies to examine robustness.

5. MORAL HAZARD INDICES AND BANK PERFORMANCE We employ the regression analysis to investigate whether the two moral hazard indices are informative enough to able to account for variations in the ROA ratio and the NPL ratio. We expect that an increase in both moral hazard indices raises the NPL ratio but decreases the ROA ratio. y it = α 0 + α 1 Loanit + α 2 Non − Interest it + α 3 Statei + α 4 Moral _ Hazard i

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

+ α 5 Asset + α 6 GDPt + ε it ,

i = 1, 2..., N ,

t = 1, 2..., T

(1)

where the dependent variables are the non-performing loan (NPL) ratio) and the return of asset (ROA) ratio which serve as proxy for the risk and profit of a bank, respectively. The explanatory variables are Loan which is attempted using five different types of loan one-by one, i.e., inter-related loans, collateral loans, credit loans, total loans and the growth rate of loans. Non-Operation Revenues which is the non-interest income ratio with non-interest income as the numerator and operating income as the denominator; and State which is a dummy variable if the bank is owned by the government and not otherwise. Moral Hazard is the moral hazard indices defined above; Asset is the total bank asset after logarithmic transformation; and Macro is proxied by the real GDP growth rate. See Table 5 for the detailed definitions of the variables. Tables 6 and 7 present the estimated results using the NPL ratio as the dependent variable. There are six specifications in each table depending on the types of loan that are used. In Table 6, the coefficients of the moral hazard index, proxied by MH I, are all insignificantly different from zero, causing us to reject our initial expectation that an increase in moral hazard tends to be associated with an increase in the NPL ratio. One plausible reason, as we have suggested above, may be that there is not too much pyramidal control structures and cross-holdings in Taiwanese banks, and therefore, most observed voting right ratios are around unity (see Table 1). Thus, the slight variations in MH I make it difficult to account for the large variations in the NPL ratio. The explanatory variables show the expected signs. For example, the coefficients for inter-related loan, collateral loan, credit loan and growth rate of loan are significantly positive, but the coefficients of State are found to be insignificant. The coefficients of assets are significantly negative, which seemingly means that large banks have low NPL ratio. The coefficients of GDP are negative but insignificant. Table 7 uses the same specifications as in Table 6 except that MH II, not MH I, is taken as the proxy for the moral hazard. It is surprising that, in contrast to the insignificant coefficients for MH I in Table 6, the coefficients for MH II are overwhelmingly significantly positive, which confirms our earlier expectation that higher moral hazard index increases the

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Can the Deviation Between Ownership and Control Explain Bank Performance… 39 Table 5. Description of the Variables Proxy Dependent Variables

Independent Variables

Calculation =Non-Performing Loan / Total Loan

Risk

NPL Ratio

Return Loan

Non-Operation Revenues

ROA Ratio Related Loans Collateral Loans Credit Loans Total Loans Growth Rate of Loans Non-Interest income ratio

Moral Hazard Indices

MH

=Voting Rights /Cash Flow Rights =Seat-Control Rights/Cash Flow Rights

State

MH Dummy Variable

Asset

Log TA

=EAIT/TA Related Loans Collateral Loans Credit Loans = Related Loans+ Collateral Loans + Credit Loans = ( Total Loans t － Total Loans t - 1 ) / Total Loans t - 1 , where t is year =Non-interest Income/ Operating Income

D＝1 if state-owned bank D＝0 if private bank Log of Total Assets

Macro GDP Growth Rate ( G D P t － G D P t - 1 ) / G D P t - 1 , w h e r e t i s y e a r The source of variables is the Taiwan Economic Journal (TEJ), (a private data vendor in Taiwan).

Table 6. Dependent Variable: NPL Ratio; Moral Hazard is MH I NPL ratio Constant Related Loans

67.606 (3.168)*** 2.459 (1.809)*

Collateral Loans

66.856 (3.146)***

29.037 (2.030)**

0.020 (1.775)*

Credit Loans Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

MH I 65.277 44.217 (3.096)*** (2.699)***

149.443 (4.709)*** -4.043 (-0.442) 0.057 (0.515)

0.034 (1.701)*

Total Loans

0.062 (0.242) 0.003 (0.685)

Growth Rate of Loans

0.028 (4.380)*** 8.261 (3.358)***

10.731 (4.206)***

Non-Operation Revenues

0.004 (0.018)

0.023 (0.106)

-0.113 (-0.414)

0.297 (1.459)

0.088 (0.495

-0.237 (-0.222)

State

2.357 (1.414)

2.284 (1.381)

2.205 (1.334)

-0.155 (-0.092)

1.467 (1.124)

0.179 (0.115)

Moral Hazard Index

1.011 (0.763)

1.072 (0.812)

1.064 (0.803)

1.450 (1.097)

1.449 (1.159)

0.315 (0.263)

Asset

-2.951 (-2.793) ***

-2.899 (-2.770) ***

-2.824 (-2.716)***

-1.792 (-2.189)**

-0.929 (-1.361)

-7.454 -(4.465)***

GDP

-1.348 (-1.406)

-1.404 (-1.462 )

-1.370 (-1.425)

-1.266 (-1.296)

-1.256 (-1.364)

-0.986 (-1.134

R*2 0.072 0.071 0.068 0.043 0.144 0.280 Notes: (1.) Loan1: Related loans ($); Loan2: Growth rate of loans (%); Loan3: Collateral loans ($); Loan4: Credit loans ($); Loan5: Total loans ($); Non_R: Non-Operation Revenues ($); G_Own: the Percentage of State Ownership (%); MH: Moral Hazard Index; G_GDP: the Growth Rate of GDP in (t-1) (%). (2.) Numbers in parentheses are t values. ***, ** and * indicate significance at the 1%, 5% and 10% level, respectively.

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

40

Chung-Hua Shen, Chien-An Wang and Meng-Wen Wu

risk of operating a bank. Important to note, it also confirms that using seat-control rights, instead of voting rights, as numerator of the moral hazard indices may be more useful in predicting bank performance. This is probably because controlling shareholders in Taiwan could control a relatively large number of seats, irregardless of whether they holds a large percentage of shares; this would make the variation in seat-control rights greater than that in voting rights. Simply put, it could very well be that seat-control rights may best capture the true moral hazard of controlling shareholders. Table 7. Dependent Variable: NPL ratio; Moral Hazard is MH II NPL Ratio Constant Related Loans

51.848 (2.527)***

51.521 (2.518)***

MH II 50.105 37.523 (2.475)*** (2.439)***

1.667 (1.269)

Collateral Loans

0.014 (1.254)

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Asset GDP R*2

0.052 (0.484) 0.054 (0.220)

0.023 (1.184)

Total Loans

Moral Hazard Index

125.038 (3.954)*** -4.211 (-0.473)

Credit Loans

Growth Rate of Loans Non-Operation Revenues State

24.012 (1.786)*

0.003 (0.744)

-0.007 (-0.031) 2.806 (1.784)* 0.711 (3.092)***

0.005 (0.027) 2.773 (1.772)* 0.715 (3.119)***

-0.084 (-0.323) 2.705 (1.732)* 0.719 (3.132)***

0.208 (1.079) 0.975 (0.601) 0.782 (3.486)***

7.538 (3.210)*** 0.021 (0.126) 2.447 (1.931)* 0.718 (3.352)***

-2.175 (-2.119)** -1.438 (-1.572) 0.155

-2.147 (-2.111)** -1.476 (-1.612) 0.154

-2.079 (-2.062)** -1.453 (-1.586) 0.153

-1.476 (-1.902)* -1.367 (-1.481) 0.145

-0.692 (-1.066) -1.352 (-1.545) 0.227

0.025 (3.901)*** 10.257 (4.118)*** -0.215 (-0.208) 0.663 (0.436) 0.479 (2.238)*** -6.196 (-3.704)*** -1.094 (-1.294) 0.319

Notes: (1). Loan1: Related loans ($); Loan2: Growth Rate of Loans (%); Loan3: Collateral Loans ($); Loan4: Credit Loans ($); Loan5: Total Loans ($); Non_R: Non-Operation Revenues ($); G_Own: the Percentage of State Ownership (%); MH: Moral Hazard Index; G_GDP: the Growth Rate of GDP in (t-1) (%). (2). Numbers in parentheses are t values. ***, ** and * indicate significance at the 1%, 5% and 10% level, respectively.

Tables 8 and 9 present the estimated results using the ROA ratio as the dependent variable. Unlike the results for MH I, reported in Table 6, the coefficients of MH I in Table 8 are overwhelmingly significantly negative. Thus, an increase in moral hazard has a negative effect on the ROA ratio. Given the results in Tables 6 and 8, MH I has no effect on risk (NPL

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Can the Deviation Between Ownership and Control Explain Bank Performance… 41 ratio) but has a great impact on return (ROA ratio). The fast growth of loans, however, decreases the ROA ratio probably because the sample period was just after the Asian crisis. Table 8. Dependent Variable: ROA Ratio; Moral Hazard is MH I ROA ratio

MHⅠ

Constant Related Loans

-27.685 (-6.344)*** -1.258 (-4.528)***

Collateral Loans

-27.281 (-6.254)***

-26.824 (-6.187)***

-12.617 (-3.457)***

-0.010 (-4.421)***

Credit Loans

-0.018 (-4.339)***

Total Loans

0.0003 (0.389)

Growth rate of Loans NonOperation Revenues State Moral Hazard Index Asset GDP R*2 Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

-10.912 (-3.435)***

0.013 (0.296)

0.004 (0.080)

0.078 (1.384)

-0.098 (-2.157)**

-1.854 (-3.393)*** -0.072 (-1.820)*

-1.001 (-2.938)*** -0.506 (-1.868)* 1.385 (6.408)*** 0.339 (1.727)* 0.384

-0.963 (-2.836)*** -0.537 (-1.983)** 1.357 (6.319)*** 0.367 (1.864)* 0.378

-0.945 (-2.778)*** -0.528 (-1.938)* 1.337 (6.252)*** 0.350 (1.773)* 0.374

-0.158 (-0.420) -0.735 (-2.497)*** 0.605 (3.319)*** 0.332 (1.527) 0.246

-0.279 (-0.961) -0.732 (-2.638)*** 0.514 (3.392)*** 0.310 (1.518) 0.329

-52.253 (-8.496)*** 0.441 (0.249) -0.015 (-0.716) -0.018 (-0.364) -0.007 (-5.954)*** -1.960 (-3.963)*** 0.084 (0.406) -0.454 (-1.504) -0.295 (-1.270) 2.729 (8.437)*** 0.254 (1.509) 0.571

Notes: (1) Loan1: Related Loans ($); Loan2: Growth Rate of Loans (%); Loan3: Collateral Loans ($); Loan4: Credit Loans ($); Loan5: Total Loans ($); Non_R: Non-Operation Revenues ($); G_Own: the Percentage of State Ownership (%); MH: Moral Hazard Index; G_GDP: the Growth Rate of GDP in (t-1) (%). (2) Numbers in parentheses are t values. ***** and* indicate significance at the 1%, 5% and 10% level, respectively.

Table 9. Dependent Variable: ROA ratio; Moral Hazard is MH II ROA ratio Constant Related Loans Collateral Loans Credit Loans Total Loans

MHⅡ -23.678 (-5.972)*** -1.046 (-4.122)***

-23.370 -22.962 -11.021 -9.762 (-5.888)*** (-5.840)*** (-3.425)*** (-3.485)***

-0.009 (-4.021)*** -0.015 -3.958)***

-45.552 (-7.842)*** 0.473 (0.289) -0.012 (-0.604) -0.019 (-0.425)

0.0003 (0.399)

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

-0.006 (-5.532)***

42

Chung-Hua Shen, Chien-An Wang and Meng-Wen Wu Table 9. (Continued)

ROA ratio Growth Rate of Loans NonOperation Revenues

MHⅡ

0.016 (0.394)

0.008 (0.194)

0.070 (1.377)

-0.073 (-1.807)*

-1.600 (-3.271)*** -0.053 (-1.507)

-1.798 (-3.929)*** 0.092 (0.484)

-0.556 (-1.640)

-0.638 (-2.417)***

-0.640 (-2.290)**

State

-1.179 (-3.880)***

-1.153 -1.138 (-3.799)*** (-3.751)***

Moral Hazard Index

-0.223 (-5.015)***

-0.226 (-5.080)***

-0.226 -0.267 -0.254 -5.079)*** (-5.694)*** (-5.687)***

-0.163 (-4.148)***

Asset

1.179 (5.943)***

1.156 (5.860)***

1.138 (5.812)***

0.519 (3.198)***

0.450 (3.327)***

2.377 (7.735)***

GDP

0.362 (2.045)**

0.385 (2.168)**

0.371 (2.087)**

0.359 (1.860)*

0.338 (1.852)*

0.281 (1.807)*

R*2

0.498

0.495

0.492

0.406

0.467

0.635

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Notes: (1) Loan1: Related loans ($); Loan2: Growth Rate of Loans (%); Loan3: Collateral Loans ($); Loan4: Credit Loans ($); Loan5: Total Loans ($); Non_R: Non-Operation Revenues ($); G_Own: the Percentage of State Ownership (%); MH: Moral Hazard Index; G_GDP: the Growth Rate of GDP in (t-1) (%). (2) Numbers in parentheses are t values. ***** and* indicate significance at the 1%, 5% and 10% level, respectively.

As shown in Table 9, the coefficients of MH II are all significantly negative. This evidence together with the results in Table 7 demonstrates that MH II is useful in predicting both the NPL ratio and ROA ratio, whereas MH I is only useful in predicting a useful predictor for ROA ratio. Accordingly, MH II is more informative than MH I in our sample. Other explanatory variables mostly show the expected signs.

CONCLUSION Employing Taiwan data, we investigate two moral hazard indices of 31 listed banks in Taiwan to determine whether these indices could explain its performance. The first moral hazard index is based on Claessens et al’s voting rights/cash flow rights (which is referred to as MH I), while the second is based on seat-control rights/cash flow rights (which is developed and referred to here as MH II). More specifically, we compare MH I with MH II to determine which is more informative in terms of signaling a bank’s non-performing loans and return on assets. Our results are highlighted in the following. First, variations are larger in MH II than in MH I, reflecting the fact that the severity of the agency problems may be underestimated when cash flow rights is used to study bank performance. We conjecture that this is probably a consequence of the fact that pyramidal control structures and cross-holding are less severe in Taiwan banking than in non-financial industries, compared to Claessens et al’s (2000) study. Future studies should examine the validity of seat-control rights ratios in other countries.

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Can the Deviation Between Ownership and Control Explain Bank Performance… 43 Secondly, when the basic statistics are used, the moral hazard indices are necessary, but not sufficient conditions to serve as precursors to overall bank performance. That is, simply judging from the results of the basic statistics, banks with high moral hazard indices or worse performance are not necessarily in bad shape, but banks that are in bad shape do have more severe moral hazard problem and worse performance than normal banks. Finally, both of the moral hazard ratios are useful in predicting return (ROA ratio), but MH II outperforms MH I in predicting risk (NPL ratio). That is, MH I is useful in predicting the return of banks, but it reveals little information about bank risk. Against this, MH II is informative about both return and risk.

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

APPENDIX: EXAMPLE OF HOW TO CALCULATE VOTING RIGHTS AND CASH-FLOW RIGHTS Figure 1 depicts the direct and indirect control chains of the United World Chinese Commercial Bank. The controlling shareholder is the government. The voting rights are calculated as follows. Voting rights. Direct voting rights is to sum all ownerships directly by the government. The number is 11.95%. Indirect voting rights, 16.56%, is the sum of minimum shares of eight indirect chains of the controlling shareholders. These indirect chains are as follows: 6.5% is the controlling power through International Bank of China (=min(44.47%, 6.5%)), 1.95% is the controlling power through China Farmers Bank (=min(45.12%,1.95%)), 1.89% is the controlling power through Taipei Commercial Bank (=min(85.71%,1.89%)), 1.08% is the controlling power through Chiao-tung Bank (=min(40.99%,1.08%)), 1.95% is the controlling power through Taiwan Cooperative Bank (= min(35.02%, 0.95%)), 0.95% is the controlling power through Taiwan Business Bank, 1.25% is the controlling power through Huanan Bank (=min(38.15%, 1.25%)),and 5% is the controlling power through Taiwan Business Bank and Huanan Bank (=min(38.15%, 5%, 0.95%)). Accordingly, total voting rights is equal to 28.51%, which is the sum of the direct (11.95%) and indirect voting rights (16.56%). Cash-Flow Rights. The calculation of direct cash-flow rights involves the same procedures as direct voting rights, and yields 11.95%. Indirect cash-flow rights, 8.43%, is the sum of the multiplications of the eight indirect chains of the controlling shareholders. That is, 2.89% is the ownership through International Bank of China (= 44.47%*6.5%), 0.87% is the ownership through China Farmers Bank (=45.12%*1.95%), 1.61% is the ownership through Taipei Commercial Bank (=85.71%*1.89%), 0.44% is the ownership through Chiao-tung Bank (=40.99%*1.08%), 0.34% is the ownership through Taiwan Cooperative Bank (=35.02%*0.95%), 0.47% is the ownership through Taiwan Business Bank, and 1.81% is the ownership through Hua-nan Bank and Taiwan Business Bank (=38.15%*5%*0.95%).

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

44

Chung-Hua Shen, Chien-An Wang and Meng-Wen Wu Land Bank

Chunghwa Post Corp., Ltd

KOWA Corp. (Construction)

Central Trust of China

1.63%

0.06%

1.19%

1.75%

Shanghai Bank

Bank of Taiwan

0.42% 0.18%

United World Chinese Bank Taichung Commercial Bank

0.96%

1.95%

6.5% International Bank of China

The Farmers Bank of China

44.47%

1.89%

1.08%

Chiaotung Bank

Taipei Commercial Bank

45.12%

85.71%

11.95%

0.95%

1.95%

40.99%

Taiwan Cooper -ative Bank

100%

1.25%

5%

Huanan Bank

Taiwan Business Bank

35.02%

38.15%

Government

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Figure 1. International Moral Hazard Index: An Example of United World Chinese Commercial Bank, and the controlling shareholder is the government.

Total cash-flow rights is therefore 20.38%, which is the sum of direct cash-flow rights (11.95%) and indirect cash-flow rights (8.62%). Moral Hazard I. Based on the above calculations, MH I is equal to voting rights /cash flow rights, which is equal to 28.51%/20.57% = 1.386. Because this ratio is higher than unity, we conclude that there is an agency problem in the United World Chinese Commercial Bank. Moral Hazard II. In 2000, the percentages of three different directors over cash-flow rights are 2.5146(the numerator is the seats of the controlling general directors/ the seats of all general directors), 1.5926(the numerator is the seats of the controlling standing directors/ the seats of standing directors), 2.2183(the numerator is the seats of all controlling directors/the seats of all directors). MH II is the ratio that the geometric average ratio of the percentage of three different directors over cash-flow rights. The number is 2.0711. Compared with MH I = 1.386, MH II is usually much higher.

REFERENCES Anderson R.C. and D.R. Fraser, (2000), “Corporate control, bank risk taking, and the health of the banking industry.,” Journal of Banking and Finance, Vol.24, No.8, pp.1383-1398. Barker, D. and D. Holdsworth, (1993), “The causes of bank failures in the 1980s.,” Research Paper 9325, Federal Reserve Bank of New York. Baumol, W. J., 1(959), Business Behavior, Value and Growth, New York, Macmillan. Berger, A.N., S.M. Davies and M.J. Flannery, (1998), “Comparing market and supervisory assessments of bank performance: who knows what when?,” Finance and Economics Discussion Series 1998-32, Board of Governors of the Federal Reserve System (U.S.). Berger, A.N., and S.M. Davies, (1998), “The Information Content of Bank Examinations.,” Journal of Financial Services Research, Vol.14, pp.117–144.

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Can the Deviation Between Ownership and Control Explain Bank Performance… 45 Berger, A.N., S.M. Davies and M.J. Flannery, (2000), “Comparing Market and Supervisory Assessments of Bank Performance: Who Knows What When?,” Journal of Money, Credit, and Banking, Vol.32, pp.641–667. Berle, A. and G.. Means, (1932), Modern Corporation and Private Property, New York: William Hein and Co. Chen, C.R., and T.L. Steiner, (1999), “Managerial ownership and agency conflicts: A nonlinear Simultaneous equation analysis of managerial ownership, risk taking, debt policy, and dividend policy.,” The Financial Review, Vol.34, pp.119-136. Chen, C.R. and T.L. Steiner, (2000), “Tobin’s Q, Managerial Ownership and Analyst Coverage: A Nonlinear Simultaneous Equations Model.,” Journal of Economic and Business, Vol.52, pp.365-382. Claessens, S., S. Djankov and L.H.P. Lang, (1999a), “Who control East Asian corporation,” Policy Research Working Paper 2054, The World Bank. Claessens, S., S. Djankov, J. Fan and H.P. Lang, (1999b), ”Expropriation on minority shareholders: Evidence from East Asia.,” Policy Research Working Paper 2088, The World Bank. Claessens, S., S. Djankov, J. Fan and L.H.P. Lang, (2000), “Corporate Diversification in East Asia: The Role of Ultimate Ownership and Group Affiliation.,” Policy Research Working Paper 2054, The World Bank. Cole, R.A. and J.W. Gunther, (1998), “Predicting Bank Failures: A Comparison of On- and Off-Site Monitoring Systems.,” Journal of Financial Services Research, Vol.13, pp. 103117. Coles, J.W., V.B. McWilliams and N. Sen, (2001), “An examination of the relationship of governance mechanisms to performance.,” Journal of Management, Vol.27, pp.23-50. Davies, J.R., D. Hillier and P. McColgan, (2005), “Ownership structure, managerial behavior and corporate value.,” Journal of Corporate Finance, Vol.11, No.4, pp.645-660. DeJuan, and Aristobulo, (1996), “The Roots of Banking Crisis: Microeconomic issues and Regulation and supervision.,”in Hausmann and Rojas-Suarez( Ed.,1996a), pp.83-102. Demsetz, R.S., M.R. Saidenberg and P.E. Strahan, (1999), “Agency problems and risk taking at banks.,” Federal Reserve Bank of New York in its series Research Paper Number 9709. DeYoung, R., M.J. Flannery, W.W. Lang and S.M. Sorescu, (1998), “Could publication of bank CAMEL ratings improve market discipline?,” Published in the Conference on Bank Structure and Competition. DeYoung, R.E., J.P. Hughes and C.G.. Moon, (2001), “Efficient risk-taking and regulatory covenant enforcement in a deregulated banking industry.,” Journal of Economics and Business, Vol.53, pp.255-282. Gilbert, R.A., A.P. Meyer and M.D. Vaughan, (1999), “The role of supervisory screens and econometric models in off-site surveillance.,” Federal Reserve Bank of St. Louis, pp.3156 Gompers, P.A., J.L. Ishii and A. Metrick, (2003), “Corporate Governance and. Equity Prices.,” Quarterly Journal of Economics, Vol.118, pp.107-155. Hirschey, M., (1999), “Managerial equity ownership and bank performance: entrenchment or size effect.,” Economics Letters, Vol.24, pp.209-213.

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

46

Chung-Hua Shen, Chien-An Wang and Meng-Wen Wu

Hirtle, B.J., and J.A. Lopez, (1999), “Supervisory information and the frequency of bank examinations.,” Federal Reserve Bank of New York in its journal Economic Policy Review, pp.1-20. Holderness, C., (2003), “A survey of block holders and corporate control.,” Economic Policy Review, Vol.9, pp.51–63. International Institute for Management Development (IMD), (1999), The World Competitiveness Yearbook. Swiss:Lausanne. Jensen, M.C. and W.H. Meckling, (1976), “Theory of the Firm: Managerial Behavior, Agency Cost and Ownership Structure.,” Journal of Financial Economics, Vol.3, pp.305360. Klapper and Love, (2002), “Corporate Governance, Investor Protection, and Performance in Emerging Markets.” World Bank Policy Research Working Paper 2818. La Porta, R., F. Lopez-de-Silanes and A. Shleifer, (1999), “Corporate ownership around the world.,” Journal of Finance, Vol.54, pp.471-517. Lee, T.S. and Y.H. Yeh, (2000), “Corporate Governance and Performance: The Case of Taiwan.,” 7th APFA Finance Conference, Shanghai. Morck, R., A. Shleifer, and R.W. Vishny, (1988), “Management Ownership and Market Valuation.,” Journal of Financial Economics, Vol.20, pp.293-315. Shleifer, A., and R. Vishny, (1986), “Large Shareholders and Corporate Control.,” Journal of Political Economy, Vol.94, pp.461-488. Shleifer, A. and R. Vishny, (1997), “A Survey of Corporate Governance.,” Journal of Finance, Vol.54, pp.737-783. Thomson, J., (1991), “Predicting Bank Failures in 1980s.,” Federal Reserve Bank of Cleveland Economic Review , first quarter, pp.9-20.

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Yeh, Y. H. and S. B. Chiu, (1996), “Ownership structure, capital structure, and firm value：a study of agency theory on Taiwan stock market.,” NTU Management Review, Vol.7, pp.57-90. Yeh, Y.H., T.S. Lee and T. Woidtke, (2005), “Family Control and Corporate Governance: Evidence from Taiwan.,” in Lang, L.H.P., Governance and Expropriation, Edward Elgar Publishing Ltd.

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

In: Taiwan: Economic, Political and Social Issues Editor: L. N. Roberts and C. R. Lewis

ISBN 978-1-60456-808-0 ©2008 Nova Science Publishers, Inc.

Chapter 3

PREDICTING BANK PERFORMANCE: A DISTRIBUTIONAL APPROACH Chiang Kaoa,* and Shiang-Tai Liub,** a

Department of Industrial and Information Management, National Cheng Kung University, Tainan, Taiwan 701, Republic of China b Graduate School of Business and Management, Vanung University, Chung-Li, Taiwan 320, Republic of China

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

ABSTRACT Data envelopment analysis (DEA) has been used as a tool for evaluating past accomplishment in the banking industry. However, due to a time lag, the results usually arrive too late for the evaluated banking institutions to react. This study makes advanced predictions of the performance of twenty-four commercial banks in Taiwan based on their financial forecasts. The forecasts based on uncertain financial data are represented in ranges and treated as stochastic data. A simulation technique is adopted to obtain the efficiency distribution of each commercial bank. Our finding is that all the efficiency scores calculated from the data published fall within the corresponding predicted ranges of the efficiency scores calculated from the financial forecasts. More importantly, the results show that the bad performance of two of the banks, which were taken over by the Financial Restructuring Fund of Taiwan, had been predicted in advance using the approach of this study.

Keywords: data envelopment analysis, imprecise data, performance, banking industry

*

E mail address: [email protected], Tel.: (+886) 6-2753396; fax: (+886) 6-2362162 (Corresponding author.) E mail-address: [email protected]

**

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

48

Chiang Kao and Shiang-Tai Liu

1. INTRODUCTION

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

An efficient financial system must have a sound regulatory system, not only to help financial institutions achieve expected development but also to prevent them from relying on risktaking. The most effective way to enforce financial rules and regulations in the financial supervisory system is to conduct financial examinations. In Taiwan, Republic of China, the principal government agencies responsible for the supervision of financial institutions are the Central Bank of China, the Ministry of Finance, and the Central Deposit Insurance Corporation. These three bank regulators use the CAMELS rating system, which consists of six categories, including Capital adequacy, Asset quality, Management, Earnings, Liquidity, and Sensitivity to market risk, to evaluate banks in Taiwan. This system relies on various financial ratios obtained from periodic reports of the entities under their jurisdiction. The ratios are also aggregated into performance indices based on various weighing or scoring schemes. The aggregation of the ratios can be a complicated process involving subjective judgment. The changes in economic conditions have made such aggregations even more difficult, increasing the need for a more reliable way to express a bank’s financial condition. Data envelopment analysis (DEA) for efficiency measurement has seen extensive applications in the study of commercial banks (Bauer et al. 1998, Berger and DeYoung 1997, Berger and Humphrey 1997, Bhattacharyya et al. 1997, Elyasiani and Mehdian 1990, Miller and Noulas 1996, Rezvanian and Mehdian 2002, Sherman and Ladino 1995, Yeh 1996, Yue 1992). Several authors have also proposed that DEA efficiency measures be used as the evaluative information for the management component of CAMELS (Barr et al. 1993, 1994, Brockett et al. 1997, Siems 1992, Siems and Barr 1998). Under the assumption of variable returns-to-scale, the DEA model for measuring the efficiency of the kth DMU, Ek, in a set of n DMUs can be formulated as follows (Charnes et al. 1978) t

Er = max . ∑ uk Yrk (1) k =1

s

s.t. ∑ v j X rj = 1, j =1

t

s

k =1

j =1

∑ ukYik − ∑ v j X ij ≤ 0, i = 1,..., n,

uk , v j ≥ ε > 0 , k=1,…,t, j=1,…,s, where X ij and Yik represent the jth input and kth output, respectively, of the ith DMU, and ε > 0 is a small non-Archimedean quantity (Charnes et al. 1979, Charnes and Cooper 1984).

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Predicting Bank Performance: A Distributional Approach

49

In most studies, the DEA approach has been used as a tool for evaluating accomplishments. The results highlight the status of the operational performance and are helpful for planning future activities to improve performance. However, this ex post facto evaluation might be a little late for an unsuccessful unit to find its weaknesses and make the appropriate adjustments. In this research, we predict the performance of commercial banks in Taiwan based on the forecasted financial data via DEA. The results are regarded as forwardlooking information, which can be used for planning management activities in advance to enhance operational performance. Since the input-output data are the financial forecasts of the banks, considerable uncertainty is involved. One approach for dealing with uncertain data in DEA is chance constrained programming. Due to the stochastic nature of the observations, the constraints that require the aggregated output to be smaller than the aggregated input are to be satisfied with specified probabilities. Different probabilities determine different efficiencies for the set of DMUs (Cooper et al. 2002, Cooper et al. 1996, Land et al. 1994, Olesen and Petersen 1995). In this approach, it is assumed that the efficiency of a DMU is stochastic, and the observation is just an occurrence of random phenomenon. Another approach is to treat the uncertain observation as interval data and calculate interval efficiency (Cooper et al. 1999, 2001, Despotis and Smirlis 2002, Kao 2006). Although interval efficiency provides more information regarding the range of the stochastic efficiency, the distribution or even the mean of efficiency is still not known. Apparently, if the intervals are narrow, then the resulting efficiency intervals will also be narrow, and they will be representative. If, on the contrary, the intervals are rather wide, then the resulting efficiency intervals will be wide as well. In this case, not much information is provided, and the base for making a decision is consequently weak. In this study, we treat the data as stochastic and find the distributions of the input/output data of each bank. By applying a simulation technique, the efficiency distribution of each bank is obtained. Efficiency distributions are obviously more informative than efficiency scores for drawing appropriate conclusions. This paper is organized as follows. Firstly, we use a simple example to explain the concept of distributional efficiency in stochastic DEA. A graphical expression of the production frontier illustrates the difference between using the average data and stochastic data to measure the relative efficiency of each DMU. Then we describe the stochastic nature of Taiwan’s commercial banks, and apply a simulation technique to find the efficiency distribution of each bank. Finally, the accuracy in approximating the distribution is discussed and some conclusions are drawn from the discussions.

2. GRAPHICAL ILLUSTRATION For simple problems, the efficiency distribution of each DMU can be derived analytically with the help of a diagram. In Figure 1, there are five DMUs, labeled as A, B, C, D, and E, using input X at levels 1, 2, 3, 4, and 5, respectively, to produce output Y of the amounts of 2, 5, 5, 6, and 5, respectively. Under the assumption of variable returns-to-scale, the production frontier constructed from these five DMUs is the piecewise line segments ABDD’, where A,

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

50

Chiang Kao and Shiang-Tai Liu

B, and D are on the frontier and are thus efficient. From the viewpoint of output, the efficiencies for A, B, C, D, and E are 1, 1, 5/5.5, 1, and 5/6, respectively. Y 7

B5

6

B4

5

D'

D ●E

C

B B3

4

B2

3 2

B1

A

1

1

2

3

4

5

X

Figure 1. Stochastic frontier of the example.

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

Now, suppose the output of B, YB, is a stochastic variable described by a uniform distribution over the range of [3, 7], instead of the constant 5. Denote f YB (y) and FYB (y) as the density function and distribution function of YB, respectively; that is,

⎧1 / 4, f YB (y)= ⎨ ⎩0,

3≤ y ≤7 otherwise

⎧1, ⎪ FYB (y)= ⎨( y − 3) / 4, ⎪⎩0,

7≤ y 3≤ y ≤7 y≤3

The production frontier constructed in this case becomes stochastic. There are four cases for the value of YB to discuss. Case 1. YB ∈ [3, 3.5] For YB in the range of [B1, B2]=[3, 3.5], the frontier constructed is the fixed piecewise line segments ACDD’, where B and E are inefficient and the others are efficient. The efficiency of B is EB = YB /3.5, depending on the value of YB, and of E is a constant, EE=5/6.

Taiwan: Economic, Political and Social Issues : Economic, Political and Social Issues, Nova Science Publishers, Incorporated, 2008. ProQuest

Predicting Bank Performance: A Distributional Approach

51

Case 2. YB ∈ [3.5, 4] For YB in the range of [B2, B3]=[3.5, 4], the frontier constructed is the stochastic piecewise line segments AYBCDD’. In this case, only E is inefficient, with an efficiency score of 5/6, the same as Case 1. Note that although YB is stochastic, EB is constant. Case 3. YB ∈ [4, 6] For YB in the range of [B3, B4]=[4, 6], the frontier is the stochastic line segments AYBDD’, where C and E are inefficient. Since the target output for C in this case is (YB+YD)/2 (note that YD=6), the efficiency of C is 10/(YB+6). The efficiency of E remains at 5/6. Case 4. YB ∈ [6, 7] Finally, for YB in the range of [B4, B5]=[6, 7], the frontier is the stochastic line segment AYB and the horizontal line thereafter. In this case, only A and B are efficient. The efficiency for C is 5/YB, for D is 6/YB, and for E is 5/YB. From the above discussion, we find that A is always efficient, no matter how YB varies in the range of [3, 7]. DMU B is efficient for YB ≥ 3.5, and the probability for YB ≥ 3.5 is 7/8. Accordingly, the probability of EB =1 is 7/8. For other values of EB, its cumulative density can be derived as:

FEB (z)=Pr{EB ≤ z}=Pr{YB /3.5 ≤ z}=Pr{YB ≤ 3.5z}=(3.5z−3)/4

Copyright © 2008. Nova Science Publishers, Incorporated. All rights reserved.

where the second equality is a consequence of EB=YB /3.5 as discussed in Case 1 and the last equality follows from the uniform [3, 7] distribution of YB. The smallest value of EB is 6/7, which occurs at YB=3, the lower bound of YB. To summarize, the complete distribution function of EB is:

⎧1, ⎪ FEB (z)= ⎨(3.5 z − 3) / 4, ⎪⎩0,

1≤ z 6/7 ≤ z