The development of the flat panel display industry in Taiwan: The ‘Triple-alliance’ perspective

Jen-wei (John) Liu

This thesis is presented for the degree of Master of Philosophy at The Australian School of Business (School of Organisation and Management) The University of New South Wales

November 2009 Abstract

This study examines what role three institutions in Taiwan – state bureaucracy, local business groups and multinational corporations, played in the development of the flat panel display (TFT-LCD) technology. The thesis adopts the ‘Triple-alliance’ conceptual framework to understand how these three institutions cooperated to develop TFT-LCD; it argues that it is vital to understand the interactions among these three institutions in order to map the development of a high technology industry in a late industrializing country.

Both quantitative and qualitative approaches are adopted in order to validate the utility of the ‘Triple-alliance’ conceptual framework. The qualitative part of the study employed five semi-structured interviews to gain first hand understanding of the development of the TFT-LCD industry. The interviews were supplemented with secondary sources from annual reports, newspapers, and trade journals. In the quantitative part of the study, data was collected from 100 information technology related firms listed on the Taiwan Stock Exchange. Hierarchal multiple regression analysis investigated the relationships between institutional factors and firm performance.

The findings reveal that there exists a strong relationship between institutional alliances and firm performance. The positive relationship highlights the significance of the ‘Triple-alliance’ framework in helping explain the institutional mechanism often adopted by late developing nations and firms to help develop indigenous technologies.

ii Contents

Abstract ii List of tables vi List of figures vii List of appendices vii Originality statement viii Copyright statement ix Authenticity statement ix Abbreviations used in thesis x Acknowledgements xi

Chapter One: Introduction 1 1.1 Introduction 1 1.2. Background of the TFT-LCD industry in Taiwan 2 1.3 Structure of the thesis 4 1.4 Synopsis of the thesis 5

Chapter Two: Theoretical underpinnings 9 2.1 Introduction 9 2.2 The ‘neoclassical economics’ perspective 10 2.3 The ‘developmental state’ perspective 11 2.4 The ‘modernization’ or neo-technology perspective 13 2.5 The ‘public-private nexus’ perspective 16 2.6 Conclusion 21

Chapter Three: Conceptual framework 23 3.1 Introduction 23 3.2 Analytical framework 23 3.3 Characteristics of the state 25 3.4 Characteristics of East Asian business groups 27

iii 3.5 Characteristics of multinational firms 29 3.6 The mechanism of foreign-owned technology transfers in Taiwan 30 3.7 Conclusion 32

Chapter Four: The development of the microelectronic industry in Taiwan 34 4.1 Introduction 34 4.2 The development of the microelectronic industry (1960s−1980s) 35 4.3 Motivation for the creation of and indigenous TFT-LCD industry 39 4.3.1 Trade deficit with Japan 40 4.3.2 Emergence of Korean electronic firms 41 4.3.3 Rising demand for Information Technology products 42 4.4 Contribution of the state 44 4.4.1 Industry-specific policy 45 4.4.2 The creation of TFT-LCD specific industrial park 46 4.4.3 Provision of human resources 49 4.5 Conclusion 50

Chapter Five: Firm-specific case studies 52 5.1 Introduction 52 5.2 Case study methodology 52 5.3 First round of consortia (1990−1998) 54 5.3.1 Background of first consortia 54 5.3.2 Formosa Plastic Group (FPG) 55 5.3.3 Discussion 58 5.4 Second round of consortia (1999−2005) 59 5.4.1 Background of second consortia 59 5.4.2 Acer group 62 5.4.3 Chi Mei group 64 5.4.4 Contribution of multinational electronic corporations 66 5.5 Conclusion 68

iv Chapter Six: Statistical analysis 70 6.1 Introduction 70 6.2 Proposed hypotheses 70 6.2.1 The role of the multinational corporations in ‘Triple- alliance’ 71 6.2.2 The role of the state in ‘Triple-alliance’ 71 6.2.3 The role of the local business groups in ‘Triple-alliance’ 72 6.3 Research methodology 74 6.3.1 Characteristics of selected firms 74 6.3.2 Construction of variables 75 6.4 Analytical approach 79 6.5 Regression results 81 6.6 Summary 83

Chapter Seven: Conclusion 85 7.1 Introduction 85 7.2 Summary of major findings 85 7.3 Contributions of the study 87 7.4 Limitations of the study 88 7.5 Future directions 89

References 90

Appendices 104

v List of tables

Table 1.1: Type of display production in Taiwan 4 Table 4.1: Taiwan’s electronic manufacturing industry 43 Table 4.2: Worldwide increase in Information Technology products 44 Table 4.3: Tainan Science Industrial Park investments overview 47 Table 4.4: Survey of TFT-LCD firms in Tainan Science Industrial Park in 1996 48 Table 5.1: First round of consortia 1990–1998 59 Table 5.2: Second round of consortia 1999−2005 67 Table 6.1: Distribution of companies by LCD activities 74 Table 6.2: Distribution of companies by respective industries 75 Table 6.3: Definition of dependent variables 76 Table 6.4: Definition of independent variables 77 Table 6.5: Qualitative variables and descriptive statistics 78 Table 6.6: Quantitative variables and descriptive statistics 78 Table 6.7: Pearson correlation coefficients 80 Table 6.8: Collinearity statistic table 80 Table 6.9: Regression results table 81

vi List of figures

Figure 3.1: The ‘Triple-alliance’ model 25

List of appendices

Appendix 1: List of interviewees 104 Appendix 2: Participant information statement and consent form 105 Appendix 3: Alphabetical list of one hundred firms 109

vii Originality statement

I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project’s design and conception or in style, presentation and linguistic expression is acknowledged.

Signed ……………………………………………...... Date ……………………………………………......

viii Copyright statement

I hereby grant the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all proprietary rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation. I have either used no substantial portions of copyright material in my thesis or I have obtained permission to use copyright material; where permission has not been granted I have applied/will apply for a partial restriction of the digital copy of my thesis or dissertation.

Signed ……………………………………………...... Date ……………………………………………......

Authenticity statement

I certify that the Library deposit digital copy is a direct equivalent of the final officially approved version of my thesis. No emendation of content has occurred and if there are any minor variations in formatting, they are the result of the conversion to digital format.

Signed ……………………………………………...... Date ……………………………………………......

ix Abbreviations used in thesis

BAIR Bureaucratic-Authoritarian Industrializing Regimes CEPD Council for Economic Planning and Development EPC Economic Planning Council ELD Electro-luminescence Display ERSO Electronics Research and Service Organisation FED Field Emission Display HSIP Hsinchu Science-based Industrial Park ITRI Industrial Technology Research Institutions IT Information Technology IC Integrated Circuit LCD Liquid Crystal Display LED Light Emitting Diodes MNCs Multinational corporations NICs Newly industrialized nations NIEs Newly industrialized economics OBM Original Business Manufacturing ODM Original Design Manufacturing OEM Original Equipment Manufacturer PC Personal Computer PDP Plasma Display Panel R&D Research and Development TFT-LCD Thin Film Transition-Liquid Crystal Display

x Acknowledgements

This is more than a thesis. It is a milestone of my journey throughout life. There were difficult days during the journey, but because of assistance from my fellow colleagues and my supervisor, the journey was much more satisfying.

To my supervisor, Dr. Pradeep Ray, thank you so much for your invaluable comments and support. Your advice, encouragement and much-needed comments throughout the writing of this thesis made the task easier. One paragraph is simply not enough to express my gratitude to you.

I would also like to thank my colleague, Mr. Zahid Riaz, for his assistance along the way, especially in the quantitative part of the study. Without Zahid’s assistance, I would still have been hopelessly lost in the dataset.

The infrastructure support and job opportunities of Graduate Research School and School of Organization & Management allowed me to complete the thesis without unnecessary trouble. I am grateful for the support of staff from both organizations.

No acknowledgement is complete without the mention of one’s family. I am grateful for the support I received from my family members. The thesis is a testament of your commitment to education. Thank you all from the bottom of my heart.

xi Chapter One: Introduction

1.1 Introduction

This thesis examines the underlying factors behind the establishment of a global ‘Thin Film Transition-Liquid Crystal Display’ (TFT-LCD) industry in Taiwan, despite its latecomer industrializer status. Various types of Thin Film Transition (TFT) display technologies have been developed over the years: Electro-luminescence Display (ELD), Field Emission Display (FED), Liquid Crystal Display (LCD), Light Emitting Diodes (LED), and Plasma Display Panel (PDP). The focus on TFT-LCD technology is attributed to the fact that it is the most widely adopted type of display technology. TFT-LCD technology accounted for approximately 85 per cent of the global display technology market, with the value reaching close to US$10 billion (OSL, 2000). The technology has broad applications from small to medium-size notebook PCs to large TV monitors.

How was Taiwan − a late industrializing country in the 1950s, dislocated from international sources of technology, backward in engineering, with a poor industrial infrastructure – able to transform itself into a high technology powerhouse in a technology intensive TFT-LCD industry in a mere three decades? This is the broader empirical setting within which the thesis is cast. The thesis adopts the ‘Triple- alliance’ framework and maps the development of the TFT-LCD industry in Taiwan from the 1990s onwards through intensive case study analysis. The triple alliance framework was first propounded by Evans (1977), who argued that development in late industrializing countries is often contingent upon the conjoint involvement of three institutions: the nation-state, multinational corporations (MNCs) and local business groups. Gold (1986) and Kim (1997) later adopted this analytical framework to explain the success of East Asian nations. While the framework is largely confined to the field of sociology, the study suggests that the framework is useful in understanding the development of high-technology industries in latecomer industrializers. The development of ‘TFT-LCD’ industry demonstrates the utility of the ‘Triple-alliance’ model.

1 Whereas some studies focused mainly on econometric modeling of the impact of foreign technologies on latecomer firms’ performance, the literature is rather sparse on detailed industry case studies, a few exceptions notwithstanding (see Matthews, 1999). Extant industry case studies present aggregated analyses using broad industry nomenclature, viz. electronics and semiconductors. This is a limitation that this study seeks to address. It presents a longitudinal study of the birth and growth of a specific (TFT-LCD) industry, and thus imparts a keen insight into the sub-processes and drivers that punctuate growth – unlike cross-sectional pooled studies.

Furthermore, much of the empirical literature has dwelled on partial theories of how state intervention vis-à-vis unfettered markets assisted in development of latecomer firms in East Asia, not stepping beyond the simple state versus market dichotomy. Indeed, a major limitation in the theoretical literature has been the absence of studies on the dual and conjoint impact of both state and markets, not as adversaries, but as shapers of institutions enabling industrial development. The triple alliance model addresses this challenge, and undertakes a bold attempt to integrate the neoclassical, statist and neo-technology perspectives. Finally, a major limitation in the literature is the lack of modeling of industry case studies into a more generalized framework to test the empirical validity of propositions. This is also undertaken in this study and renders this study unique in its present form.

The chapter is divided into three major parts. Section 1.2 discusses the development trend of TFT-LCD industry in Taiwan. Section 1.3 discusses the structure of this thesis and offers a broad outline of each chapter. Section 1.4 provides the synopsis of each subsequent chapter.

1.2. Background of the TFT-LCD industry in Taiwan

The microelectronic industry in Taiwan has been an important contributor to its economic growth. Strong growth in various electronic and information technology products has spawned related and supporting industries for parts and components through the spillovers it generated. Of different types of parts and components, the flat panel display is most critical, since it serves as an integral intermediate input in

2 virtually every display related electronic device. Flat panel display consists of wide range of technologies, including TFT-LCD, LED, STN-LCD, and PDP. These technologies were initially developed in US for military uses, and later commercialized by Japanese electronic firms. By the early 1990s, Japanese electronic firms became the global leader in flat panel display technology. These firms continued to dominate the production of flat panel display throughout the 1980s and early 1990s until Korean electronics firms began to enter the market in the mid-1990s (Matthews, 2005).

The entry of Taiwanese firms into flat panel production was largely due to the need on the part of Japanese manufacturers to secure a steady source of supply of flat panels for the reliable production of electronic and information technology products (Murtha, 2001; Shih, 1996). High cost pressures at home following wage increases, and yen appreciation, drove Japanese electronic firms to seek out and migrate its production to high-skill and low-cost locations. Interestingly, even as Japanese firms began to team up with local Taiwanese business groups via joint ventures and technology licensing, technocrats in Taiwan began to seek technology upgrading, and it chose high-value, critical-component TFT-LCD, to be its strategic industry (MOEA, 1998). This chance event and the selective targeting of the TFT-LCD industry began under the auspices of the state bureaucracy in Taiwan. Following the lead of the state, local business groups began to invest in TFT-LCD, however, these investments were constrained by lack of proprietary production knowledge (Linden & Murtha, 1998).

It is instructive to note that Taiwan’s earlier foray to establish an indigenous flat panel industry had failed due to lack of production know-how. This situation was reversed only after the involvement of multinational electronic firms from Japan and the US. Initially, Taiwanese firms could produce only the more primitive flat panel, and only gradually did it move toward the production of advanced flat panel and focus on TFT-LCD technology (MOEA, 2005). Table 1.1 shows the type of display technologies produced by Taiwanese firms from 1985 to 2006. The significance of TFT-LCD to Taiwanese firms can be understood with the huge increase in production value. It is estimated that the industry grew from a mere US$133 million industry in 1989 to become US$ 8.8 billion in total production value by 2003 (MIC, 2003).

3 Table 1.1: Type of display production in Taiwan

Period 1985 - 1989 1990 - 1995 1996 – 2000 2001-2006 Inches 3 7 10.4 13.5 Pixel 89,000 307,000 389,000 420,000 Technology PMLCD, esp PMLCD, esp AMLCD, esp AMLCD, esp Type STN-LCD STN-LCD TFT-LCD TFT-LCD Application Calculators & Computer Notebook PC & Mobile Computer Monitor & Mobile Communication Monitor Notebook PC Communication & LCD TV Source: Market Intelligence Center, report in Taiwan Economic Journal, Jan. 1996.

Taiwanese firms did not enter into the TFT-LCD industry until quite late. By then, Japanese and South Korean electronic firms had already established a major presence in the industry. It was not until the 1997 Asian financial crisis, when many of TFT-LCD manufacturers in Japan decided to exit the industry and transferred production and product technology to Taiwanese firms, that the TFT-LCD industry in Taiwan began to enjoy rapid growth. By 2005, three countries monopolized close to 90 per cent of global TFT-LCD production. These are South Korea with 40 per cent, Taiwan with 30 per cent and Japan with 20 per cent (Chang, 2008).

1.3 Structure of the thesis

The research question this thesis seeks to answer is why Taiwan was able to succeed in the global TFT-LCD industry despite its later-comer status, the presence of high entry barriers and lack of indigenous technologies. In answering this question, the thesis adopts the ‘Triple-alliance’ model to analyze the interaction among three participating institutions: the state, local business groups and multinational electronic firms. The thesis argues that the alliance and cooperation among these three institutions was critical to the success of Taiwan’s TFT-LCD industry.

The thesis is organized as follows. Chapter Two provides a comprehensive review of the relevant literatures. Chapter Three introduces the analytical framework. The chapter describes the mechanism of the ‘Triple-alliance’ model, combined with a brief description of relevant organizations. Chapter Four presents the developmental background of the microelectronic industry in Taiwan, and the contribution of the

4 state to the development and upgrading of the microelectronic industry in Taiwan. Chapter Five focuses on the evolution of the TFT-LCD industry from the 1990s to 2005. Case studies of three business groups are provided to better illustrate the interaction among business, state and multinational electronic firms. Chapter Six focuses on the statistical analysis to test the validity of the ‘Triple-alliance’. Three hypotheses are raised in answering and testing the effectiveness of ‘Triple-alliance’ argument in this chapter. The chapter also offers the empirical testing of 100 public- listed electronic firms in Taiwan to test whether firms with an alliance to the state and multinational electronic firms enjoy better performance than firms that do not have such an alliance. Chapter Seven presents the conclusion of the thesis. It begins with the summary of the study, and then points to the limitations of the study and directions for further research.

1.4 Synopsis of the thesis

Chapter Two provides the theoretical underpinnings of the study from four major standpoints. The first − the ‘neoclassical economics’ perspective − sees industrialization as a market-led development phenomenon; its proponents argue that the success of East Asian economy came as the result of bold initiatives of private enterprise, taking on risks, and responding to incentives offered by trade liberalization. The second – the ‘developmental state’ perspective − sprang from the field of political science. Its proponents envision industrialization as being guided by the state, and the success of East Asian economies as the result of bold initiatives of state bureaucrats that directed resources to both public and private sectors for rapid industrialization. The third is the ‘modernization perspective’ or neo-technology hypothesis, which sees industrialization as the process of technology transfer from advanced countries to lesser-developed nations. According to the ‘modernization perspective’, the success of the East Asian economy came as the result of global diffusion of advanced technology by multinational corporations. The fourth − the ‘public-private nexus’ perspective − focuses on integrating the three former perspectives. Based on the ‘public-private nexus’ perspective, it sees the success of the East Asian economy as the result of alliances among various public and private institutions to successfully capture the

5 technology. In conclusion, the ‘neoclassical’, ‘developmental state’, ‘modernization’ and the ‘public-private nexus’ perspectives have each been indispensable in explaining the economic development of East Asia. The four approaches outlined in Chapter Two demonstrate the necessity of considering the interaction between public and private sector that is so critical in developing a better understanding of why the East Asian economies achieved unprecedented success in their industrialization.

Chapter Three presents the ‘Triple-alliance’ conceptual framework to analyze the development of the TFT-LCD industry in Taiwan. This conceptual framework highlights the mutual interdependence and dynamics of three major socio-economic institutions: the state, local capital and multinational firms. The evolution of Taiwan’s microelectronic industry underscores the importance of these three institutions in driving the nation’s continuous industrial upgrading and technology deepening. In this alliance, each institution took on specific roles: the state formulated vertical industrial policies via public institutions in a way that engendered comprehensive development of microelectronic industries in Taiwan; local business groups took on risks and mobilized resources to foster vital domestic capabilities in developing advanced technologies; and the linkage with MNCs allowed effective coordination of demand and supply between local firms in Taiwan and overseas markets. The major advantage of the ‘Triple-alliance’ model is that it allows detailed examination of the process of foreign technology transfer and domestic technological capability development at the institutional level. In summary, Chapter Three demonstrates that the successful creation of the new industry was achieved within the context of the dynamic interactions among institutions.

Chapter Four traces the historical development of the microelectronic industry in Taiwan. This industry began in Taiwan with Western MNCs taking advantage of incentives offered by Taiwanese states to begin the production of consumer electronics in the early 1960s. The involvement of MNCs became critical in the later stages as Taiwan sought to develop its indigenous TFT-LCD industry. By the late 1980s, Taiwanese manufacturers became one of the largest subcontractors for large multinational electronic firms. The alliance with MNCs allowed local firms to emerge as one of the major institutions in the alliance. The alliances formed among state, MNCs and local business groups became critical in the development of the indigenous

6 TFT-LCD industry, since the production of TFT-LCD requires advanced processing capability in manufacturing along with a significant level of capital investment. The production of TFT-LCD requires a high degree of technology competence in chemical and electronic components. MNCs became critical suppliers to local firms in critical components, such as color filters, glass substrate, IC driver and liquid crystals. The state provided various forms of assistance to relieve the private sector from significant capital requirements that are needed for investments in the production of TFT-LCD.

Chapter Five provides firm specific case studies to better understand the motivation for Taiwan to enter into TFT-LCD industry, and how it developed absorptive capacity. The decision to produce TFT-LCD domestically was the result of changing external circumstances. Three factors in particular have led both the state and electronic firms in Taiwan to make investment in TFT-LCD industry: the increasing trade deficits with Japan, rising competition from Korean electronics firms, and constant shortages of flat panels that led to inordinate production delays. TFT- LCD was deemed critical to both the public and private sector in Taiwan, since it was seen as a critical component that can be used as the base for the growth of the whole IT industry in Taiwan. Both the state and the private sector envisioned the future in which Taiwan-made panels are installed in Taiwan-designed consumer electronics, personal computers and other electronic products (Shih, 1996). In short, the state performed the role in leading development through a mix of policy designs and resource allocations, while the local business groups with the technological assistance of multinational electronic firms mobilized the available resources in producing the products.

Chapter Six uses econometric modeling to examine the impact of institutional factors in the creation of display industry for latecomer firms in Taiwan. Specifically, it examines whether the existence of ‘Triple-alliance’ leads to better performance of individual firm sales. Three related hypotheses are proposed to test the effect of the ‘Triple-alliance’. Using a sample of 102 firms drawn from publicly listed companies on the Taiwan Stock Exchange (TSEC) in 2006, hierarchical regressions are performed to test the empirical validity of the proposed framework and the hypotheses harbored therein. The empirical results show that individual firms that possessed institutional linkage to the state and multinational firms did achieve superior

7 performance, thus confirming the plausibility of the argument that the institutional alliance of the state, local business group and MNCs is essential to facilitate the creation and expansion of TFT-LCD industry.

Chapter Seven, the concluding chapter, summarizes the major finding of the thesis and discusses the limitations and possible directions for future research.

The major finding of this thesis suggests that for latecomer firms to develop competitive advantage in the TFT-LCD industry, it is critical to form a vertically coordinated system that includes both the state and MNCs. It is further demonstrated that firms attempting to enter the flat panel display industry must also have the backing of business groups in order to develop competitive advantage. Through validating the ‘triple-alliance’ model, the thesis confirms the utility of the ‘institutional-based view’ as a useful perspective in the field of International Business. Institutions should not be viewed in abstract, but rather as distinct social-economic organizations that are capable of interacting with each other effectively to formulate suitable industrial policies, as well as fostering the creation of technology-intensive products in a practical setting. The thesis proposes several directions for future studies, including understanding the actual process of technology transfer and expanding the empirical design of the study to private owned firms in the electronic industry.

8 Chapter Two: Theoretical underpinnings

2.1 Introduction

Since the end of World War II, a group of East Asian countries have posted stellar improvements in their economic status within the world economy. Within four decades, Japan, South Korea, and Taiwan attained high per capita growth rates, with relatively low-income inequality, successfully exporting many technology intensive products. A voluminous literature has been generated on the topic of East Asian development to explain the factors that have brought about economic success. Different schools of thought, mainly from the politico-economic persuasion, have utilized varied approaches in explaining the reasons behind the miraculous growth that has taken place.

The objective of this chapter is to first summarize theoretical mainsprings that explain how East Asian states were able to achieve rapid industrialization. The chapter is divided into two main parts. The first part focuses on a review of three major strands of literature on economic development, with a view to explaining the process of industrialization in East Asia. The second part of the chapter begins with the focus on integrating these three different lines of literature by comparing and contrasting each perspective. It summarizes the literature into three major themes: the role of the state, the effectiveness of the state intervention, and the relationship between the public and private sectors. It then discusses three related studies that explain the success of East Asian industrialization as a result of institutional alliance between public and private sectors. The chapter harbors the hypothesis that institutional underpinnings of a ‘public-private alliance’ is perhaps the most plausible explanation of East Asia’s industrialization success.

9 2.2 The ‘neoclassical economics’ perspective

The ‘neoclassical economics’ perspective stresses the importance of liberalized markets and open economies. Based on the simple Heckscher-Ohlin (H-O) factor proportions model, this perspective argues that openness to trade and technology flows optimize global resource allocation and lead to industrial efficiency (Lall & Latsch, 1998). It highlights the centrality of the price mechanism that signals the incentives to make the right resources allocation (Stiglitz, 1996) in capital, labor, and technology (Lall & Latsch, 1998). Product markets give the correct signals for investment in new activities and factor markets respond to these signals (Lall, 1994). To proponents of the ‘neoclassical economics’ perspective, the success of the East Asian economy came as the result of bold initiatives of private enterprises taking on risks and responding to incentives offered by trade liberalization1.

Hence in the East Asian debate, the neoclassical economists highlight the role that liberal international trading and domestic labor markets played in its development (Little, 1979, 1982). To the proponents of the neoclassical perspective, the success of the East Asian economy came as the result of private enterprises responding to incentives offered by trade liberalization, based on the comparative advantage and abundance of certain factors of production such as low labor costs. According to this view, the government’s role was limited to providing only public goods and services such as education, law enforcement, and physical infrastructure that were difficult to arrange through the market (Little, 1979). Conservative government budgets were a sign of minimal state interference, hence the market experienced organic growth with its rational individualistic premise, rather than growth initiated by state interference (Little, 1982; see also Wade, 1995 for a critique). Trade is the main channel of technology flows, and any restrictions on trade would deprive ‘follower’ countries of the benefits of technology diffusion. Full and rapid liberalization in developing countries therefore is the best policy.

1 This view is well elaborated by Balassa, Bela (1982) Development Strategies in Semi-industrial Economies. Johns Hopkins University Press; Little, Ian (1979) Industry and Trade in Some Developing Countries: A Comparative Study. Development Centre of the Organization for Economic Co-operation and Development, Oxford University Press; Little, Ian (1982) Economic Development: Theory, Policy, and International Relations. Basic Books Press.

10 Balassa (1982, 1988) attributes industrialization success in Taiwan largely to the function of the free market, and credits the state only for maintaining conservative budget and taming inflation, which in turn encouraged and induced private investments. More importantly, Balassa (1988) emphasizes the importance of private investments in East Asia to take full advantage of the nation’s comparative advantages in low costs and educated labor forces. The World Bank’s study titled ‘East Asian Miracle’ (World Bank, 1993) also noted that East Asian states possess the qualities that are conductive for trade, including near-free-trade regimes, an export orientation, and neutral incentives between foreign and local markets, which permitted the realization of static comparative advantage.

Although the neoclassical perspective acknowledges that East Asian governments played a role during the period of economic development, it argues that the role was mostly confined to providing macroeconomic stability and allowing the basic market mechanism to function without interference, thus allowing resources to be allocated in the most efficient manner. In general, it harbors the view that government interventions have high costs (bureaucratic waste and widespread infractions of individual liberties), which are likely to outweigh any possible benefits. In summary, the ‘neoclassical economic’ perspective views the rapid economic growth as the result of trade liberalization and adoption of market-oriented incentives.

2.3 The ‘developmental state’ perspective

The ‘developmental state’ perspective originated from the recognition that the market is not always efficient in allocating resources and, in some cases, markets fail to exist at all (Lall, 1995). Such a view led to the need for the direct intervention of the state that could ensure allocation of resources for rapid productivity rise in strategic industries (Chang & Rowthorn, 1995). It also views East Asia’s rapid industrialization as a direct result of an assertive and autonomous state in which the state decided which industries should be developed and chose who should enter these industries (Dunning, 1997; Haggard, 1990; Johnson, 1982, 1995; Wade, 1990).

11 Proponents of the ‘developmental state’ perspective criticize neoclassical economists for ignoring domestic politics as an important factor (Haggard, 1990) and argue that the success of East Asia follows a different course to that described by neoclassical economics. It views the successful industrialization of East Asia as a direct result of an autonomous state that asserts itself toward industrial development for national security purposes2.

The logic of the state-led development paradigm can be explained as follows. For any change to transpire, a major transformation in the economic structure is required, and with it simultaneous changes in many sectors of the economy. Thus those who control individual components may be unable to achieve the transformation, due to lack of systemic vision or strategic uncertainty about the behaviour of other agents. Hence the state can drive the private sector agents into concerted action without making them spend excessive resources in information gathering, processing, and political bargaining (Chang & Rowthorn, 1995). Dunning (1997) postulated that the role of the government within the economic and social milieu in 1990s is such that it is not only a creator and sustainers of the institutional infrastructure, but also a fashioner of value systems and ideologies – a focal point around which resource allocation decisions are co-coordinated.

Johnson (1982, 1995), through his studies of Japan’s Ministry of International Trade and Industry that governed Japan’s industrialization, concluded that it was the recruitment of bureaucrats from elite universities based on merit that had a very large impact on the governance of the country. This view became widely popular as case after case emerged to confirm the influence of state bureaucrats. Haggard (1990) argues that government influence was highly evident in South Korea and Taiwan, and that a major reason why East Asian states succeeded in rapid industrialization was that the authoritarian regime both provided political stability and mobilized necessary resources for the development.

2 This view is well elaborated by: Haggard, Stephan (1990) Pathways from the Periphery: The Politics of Growth in the Newly Industrializing Countries. Cornell University Press; Johnson, Chalmers A. (1982) MITI and the Japanese Miracle: The Growth of Industrial Policy, 1925-1975. Stanford University Press; Wade, Robert (1990) Governing the market: economic theory and the role of government in East Asian industrialization Princeton University Press.

12 In the case of Taiwan, Wade (1990) argues the government’s leadership and assistance to private business was the critical reason for its success. Wade viewed the state as a monolithic professional bureaucracy, capable of formulating coherent and consistent industrial policies, and mobilizing national resources to implement such sector-specific policy to effect structural changes (Wade, 1990). The state’s willingness to support capital-intensive projects was largely grounded in its security needs, and the general lack of capability in private enterprises to shoulder these projects. Wade demonstrated that in Taiwan the state held control over various institutions, including public sector enterprises, banks, and research institutions as public institutions. These institutions, in turn, allowed the state to implement its industrial policies to develop certain strategic industries, and facilitate industry upgrading through R&D (Wade, 1990). In short, the state’s actions to prop up various industries reinforced its capacity to enforce its vision onto the society by will, thus creating a perpetual state of development.

The ‘developmental state’ perspective thus categorically rejects the ‘neoclassical economics’ argument that governments in East Asia acted in a limited capacity. They present empirical evidence from the East Asian experiment to demonstrate the state’s heavy involvement in many aspects of industrialization, where competent and rational bureaucrats in charge of various aspects of industrialization retained critical decision- making in resource allocations.

2.4 The ‘modernization’ or neo-technology perspective

The ‘modernization’ perspective is a shorthand term that encapsulates various neo-technology theories. It assumes that trade leads to the diffusion of knowledge, allowing technology gaps to be closed over time (in product cycle models). The technology-receiving country absorbs and uses technology efficiently through the metaphor of the ‘Flying Geese’ formation, in which development in follower countries is a function of transfer and subsequent diffusion of technological know- how from developed countries (Bernard & Ravenhill, 1995; Kojima, 2003). The ‘modernization’ perspective envisions a regional hierarchy production network, with each nation taking advantage of its own given comparative advantage. The ‘Flying

13 Geese’ formation envisions tiers of division of labor between developed nations and newly industrialized countries (NIEss), where the former transfer mature technologies and industries to the latter. This particular view of economic development was used as the basis of model on the regional development of East Asia, since it emphasized how the technology is transferred and dispersed from developed nations to less developed nations3.

Akamatsu first presented the ‘Flying Geese’ (FG) perspective in the late 1930s. The framework was based on the experience of Japan’s industrialization which indicates countries can be categorized into advanced leader, fast follower, and newly rising groups based on the degree of industrialization of respective country. Akamatsu (1932) explains how certain industrial sectors migrate from advanced nations to lesser-developed nations as the life of the product ages. Based on Akamatsu’s ‘Flying Geese’, Vernon (1971) examined the life cycle of individual products to gain better understanding of the competitiveness of firms in the manufacturing sector. The international product cycle (IPC) hypothesis envisioned four phases: first, innovation of product and growth of sales in the developed nation’s market; second, saturation of the domestic market in the developed nations and export of products to the foreign market; third, manufacturing of products in the foreign market in the form of foreign direct investments; and fourth, reverse import of products from foreign markets back to the home market. Hence both the FG and the IPC framework emphasize dynamic changes in less developed economies as a result of technology spillovers from advanced countries and a gradual catching-up phenomenon in the former group (see Matthews, 2005).

In a similar vein, Deyo (1987) summarized the ‘modernization’ perspective into a logical sequence of a developmental path where NIEs move through series of path- dependant steps that began with the initial importation, then to import substitution and eventually re-export to less developed nations. Accordingly, NIEs initially imported intermediate and capital goods from Japan for processing and exported the finished

3 The proponents of modernization perspective have gone through several revisions since its original exposition of how textile industry migrating from Japan to other Asian nations. See: Deyo, Frederic C. (1987) The Political economy of the new Asian industrialism. Cornell University Press, 1987; Hobday, Michael (1995) East Asian latecomer firms: Learning the technology of electronics. World Development, 23(7): 1171 – 1193.

14 goods to US markets, and only gradually moved up the hierarchy of export-oriented production once Japan conceded market share of some of mature technology sectors to NIEs to free resources and concentrate on other sectors (Deyo, 1987).

The microelectronic industry is frequently cited as a sector that shows the pattern of technology transfer from advanced developed nations to less developed nations as the product ages. Bernard and Ravenhill (1995) presented a critique of the flying geese theory and evolution of the product life cycle since its first inception. The basic premise of the flying geese model sees supply architecture that build around Japan’s innovation of components, machinery, and materials (Bernard & Ravenhill, 1995; Hart-Landsberg and Burkett, 1998). For the newly industrializing economies Korea and Taiwan, industrialization has been dependant on imported technology from Japan to facilitate their export-oriented policies (Bernard & Ravenhill, 1995). Hobday (1995) argues that the redeployment of resources from the more developed state to a less developed state was smutually beneficial, since it formed an alliance network that sees private factories and plants to import capital goods and key components from Japan and export to American market (Hobday, 1995). Hobday observes that newly industrialized countries, such as Taiwan and Korea, can advance to a more technology-intensive part of the value chain through rapid absorption of know-how spilled from developed nations.

According to Tsai (2006), East Asia’s success in global microelectronics is to be attributed to the change of industrial structure from vertical to horizontal integration across the region. Cross-border expansion created opportunities for those nations that had low costs, but were currently without sophisticated technology to participate in the lower end of manufacturing. The popularity and the standardization of the personal computer industry (PC) illustrates that firms that were able to deliver the low-cost PC in the fastest time became firmly established in this field, having accumulated manufacturing and engineering experience from previous development in light consumer electronics, and possessing low costs that attracted investments from foreign multinationals to invest its production lines (Tsai, 2006).

15 2.5 The ‘public-private nexus’ perspective

Scholarly debate on East Asia’s developmental experience pitches the ‘neoclassical economics’ and ‘developmental state’ perspectives as the two opposite ends of a spectrum. But what exactly is the role of the state vis-à-vis market? The neoclassical rationalists assign the state to a minimalist role, and hence come under fierce attacks in the light of obvious state intervention in industrializing East Asia nations. The more moderate proponents of ‘neoclassical economics’ adopt a more conciliatory approach, admitting the importance of having a stable macroeconomic environment and well-developed human capital infrastructure towards developing absorptive capacity for successful industrialization. In the seminal publication titled “East Asia Miracle” (World Bank, 1993), neoclassical economists conceded that state interventions might have brought a few benefits, but emphasized state intervention was effective only insofar as it corrected generic market failures (Lall, 1994) and was confined only to good macroeconomic management. The evolved market-friendly view acknowledged the involvement of the state, but concluded the economic success stemmed from the application of a set of common, market-friendly economic policies that emphasize macroeconomic stability and investment in education, leading to both higher accumulation and better allocation of resources (Lall, 1994; World Bank, 1993). However, no one denies the importance of good macroeconomic policies, and the need for sound education policies is hardly in doubt (Lall, 1994). Hence the emphasis of the World Bank on the latter two is not really where the two perspectives diverged. What was in dispute, however, was whether East Asia deployed the sector- specific selective interventions in favor of strategic industries with high income elasticity and a rapid growth of productivity (Fransman, 1993; Lall, 1995).

The indeterminacy in and lack of closure on the debate led to the ‘public-private nexus’ perspective – an integrated framework that draws on the three perspectives introduced in the previous sections. These three perspectives have led the focus to institutional foundations, drawing the connection between social institutions and economic performance in the context of East Asia (Chu & Hu, 1999). Weiss (1995) has proposed that three interlinked themes have emerged from the debate in East Asia’s development: the role of the government; why government interventions have worked relatively well; and the relationships between the state and private enterprise.

16 Acknowledgement of the state’s intervention led to the following question: ‘Why did government’s intervention work relatively well in East Asia?’ Two diverging arguments have been posed. The positivist camp (Cumings, 1987; Noble, 1998; Wu, 1992) saw that competent and autonomous bureaucrats played a leading role throughout the East Asian development phases. Cumings (1987) showed that the cause of East Asian’s economic growth is the cooperation between state and private businesses to carefully coordinate and mobilize its limited resources in favor of selected industries for high growth.

Cumings (1987) referred to such phenomena as ‘Bureaucratic-Authoritarian Industrializing Regimes’ (BAIR). The BAIR model allowed enterprising bureaucrats to formulate a policy of state action in areas unattractive to private capital in order to create conditions that would encourage and sustain private investments (Cumings, 1987). This model called for an alliance between the bureaucrats and a selective group of private industrialists to cooperate to advance the state’s economic interests, especially in favor of strategic industries, prone to various incidences of market failure. In this arrangement, the bureaucrats would provide capital and offer legal protection, thereby providing risk insurance for industries prone to increasing returns to scale and dynamic comparative advantage, while the selective group of private industrial group would supply management and technology know-how.

By and large, the leading NIEs established proper economic structures and resource allocation mechanisms conducive to the development of high valued-added, technology-intensive industries, or in other words they picked ‘winners’. Wu (1992) argues that the distinctive relationships that fostered cooperation between government and private enterprise led to an enhancement of technology absorption, adaptation, and diffusion – that allowed rapid industrialization. Indeed, the success of East Asian states in producing highly sophisticated integrated circuits in the domestic electronic industry proves the effectiveness of such intervention (Wu, 1992). Noble (1998) argues that the industrialization strategy adopted by East Asian states emphasizes developing an absorptive capacity to utilize foreign technology for the purpose of its own technological upgrading and industrial deepening of key industrial sectors.

On the other hand, the negativist camp led by Krugman (1994) saw that economic miracle as an unsustainable mirage. He suggested that the rapid

17 industrialization was the direct result of an astonishing mobilization of resources, instead of an improvement in efficiency and productivity (Krugman, 1994). Following this line of reasoning, the East Asia economic miracle can be explained by massive growth in investment demand and only modest growth in efficiency, with most growth driven by inputs that will ultimately lead to a slowdown in the economy (Krugman, 1994).

Weiss (2003) considered positivist and negativist arguments and observed that rational and competent bureaucrats played an autonomous and leading role throughout the development phase of East Asia states. The highly capable elite bureaucrats resided within the economic agency of the state and exacted high economic performance by targeting certain industries they deemed strategic through offering various forms of incentives and subsidies to induce private investments (Weiss, 2003). The seemingly conflicting paradigms differed in their emphases – while one gave more attention to rational and impartial bureaucrats who did not influence resource allocation per se, the other gave more weight to mobilization of state apparatus towards creating winners. The two perspectives nevertheless shared the same premise in focusing on the relationship between the state and private enterprises – being positive and cooperative rather than conflicting and interfering. This relationship is characterized as an ‘embedded autonomy’ in which the state was embedded in the society it governs, while through certain kind of institutional arrangement it retained its autonomy to implement its industrial policies for economic development (Evans, 1982). The central argument of ‘embedded autonomy’ is that it is vital to recognize that the state has played a strategic role in the process of economic development in both approaches; indeed, the government has formulated a long-term national investment policy, and has harnessed domestic and international market forces to this end (Onis, 1990).

Scholars from various disciplines also participated in the debate on the relationships between the state and private enterprises. Amongst these are the theory of ‘Dependency Development’ (Evans, 1977; Gold, 1986), studies of public-private interdependency (Li, 1995; Weiss, 1994, 1995) and foreign-domestic alliance networks (Biggart & Hamilton, 1988; Chu, 1999; Kim, 1997). The first group of scholars focusing on the rising Asian economies during the 1980s used the concept of

18 ‘Dependency Development’ as its framework before making theoretical adjustments to explain the East Asian nations’ development. The new hypothesis recognizes that national industrialization through cooperation among state, domestic businesses, and multinational companies exists in East Asia as well as in South America. However, in the case of East Asian nations, such alliance does not ultimately depend upon foreign capital and expertise. The development of indigenous industries in East Asia nations ultimately depends on the interaction between the state and the local business.

The theory of ‘Dependency Development’ is one of the major paradigms that models hierarchical interactions between advanced and less developed nations. The theory concurs with the insight of the ‘modernization’ perspective in which it sees successful industrialization as the result of technology and knowledge diffusion from advanced nations to developing nations. This insight gave rise to a whole new set of paradigms that sees the process of industrialization as a wave of cycles in which mature technology migrated from developed countries to developing countries, which were endowed with competitive advantages of low labor costs and other factors of production. Developing nations were able to manage the absorption and diffusion of technological know-how through cooperation between the state and the private sector. The proponents of this paradigm often highlight the intrinsic structural arrangement embedded during the development process, and relegate late-coming nations in East Asia to a dependency status (Evans, 1977; Gold, 1986).

In his original study of how South American nations and firms develop their indigenous industries, Evans (1977) observed that institutions become dependant on the capital and technology from Western nations. Evans pointed out Brazil’s industrialization in the technology-intensive petrochemical industry was attained through the cooperation among three major institutions: state, domestic businesses, and multinational companies. The development of the Brazilian petrochemical industry is significant, since it is not an enclave industry which simply exports the natural resources, nor is it the sort of final assembly, light-consumer goods industry that typifies the early stage of import substitution, thus multinational companies must be recruited to assist the development (Evans, 1977).

The cooperation of these three different institutions was captured in what Evans termed the ‘Triple Alliance’ model. Evans (1977) indicated that while the triple

19 alliance among state, domestic, and international business successfully transformed the petrochemical industry in Brazil by redressing the intrinsic structural disadvantage of each institution, this alliance was ultimately dependant upon foreign technology, management expertise and capital for successful development.

Gold (1986) demonstrated how the state and the local business groups formed an alliance for economic development in Taiwan that is intentionally designed to bar the penetration of foreign firms into the Taiwanese market. The paradigm demonstrate that the state and private sector of Taiwan do not merely serve foreign interests; instead, it adopts a more sophisticated relationship, in which dependence on foreign technologies and markets were initially allowed for Taiwan’s economic development, but progressively reduced when absorptive capacity is developed (Gold, 1986).

Studies of ‘public-private interdependency’ proposed by Weiss (1994, 1995) posit that an interdependent relationship exists between the public and the private sector − highlighted by coordination and cooperation (see also Porter & Sakakibara, 2004; Tezuka, 1997). Weiss (1995) defined such relationship as one of ‘Governed- interdependence’ in which the systematic coordination and mobilization of resources is based on the cooperation between government and private industry, as also among industry participants in the latter group. Weiss (1994) noted that active governments pursuing strategic trade and industrial policies, constituted an important component of East Asia’s high-performance economies. The availability of policy instruments allowed bureaucrats to mobilize financial and human resources and effectively monitor their uses (Amsden, 1991), which in turn garnered more social support for the state. The reciprocal interaction between the state and society led to the creation of a triple alliance that forms the foundation of a good institutional-fit by compensating for the intrinsic weaknesses of the state and business through combining both disciplined support and risk sharing schemes (Weiss, 1995).

In defense of the public-private interdependency’, Li, the former Minister of Economic Affairs of Taiwan, explained that Taiwan’s success in establishing technology-intensive industries toward becoming the global leader was due to joint efforts between government agencies and the private sector (Li, 1995). During the initial development phase of Taiwan’s microelectronic industry, the government not only initiated several tax incentives to encourage direct investment from the private

20 sector, but also provided direct funding for the establishment of microelectronic firms (Li, 1995).

Studies of ‘foreign-domestic alliance networks’ argue that the high growth rate achieved by East Asian nations would not have been possible without specific institutional arrangements (Biggart & Hamilton, 1988). Successful industrialization can be explained by how social institutions interact with each other, which resulted in shaping the economic conditions of the society. The development of indigenous industries in East Asia nations ultimately depended on the interaction between these institutions. Chu (1999) argues the development strategy adopted by the government of Taiwan since the 1980s emphasized technological upgrading and industrial deepening; while the governing regime was less intrusive compared to its East Asian neighbors, the state was not beholden to external interests. Kim (1997) argues that in South Korea the government’s leadership and financial assistance provided the direction for private business to invest. Successful industrialization was fundamentally rooted in the policy aiming to direct resources into selected industries so as to give producers in those industries a competitive advantage (Kim, 1997).

In summary, the ‘public-private nexus’ perspective argues that the major reason why East Asian nations were able to successfully capture the diffusion of technology is through a web of alliances. To recapitulate Wade’s (1984) observation, the government sought to accelerate the development of technology-intensive industries through both direct and indirect methods; it recruited private enterprise from both domestic and international settings and guided them towards certain industries and product-markets (Wade, 1990).

2.6 Conclusion

This chapter has reviewed major theories that sought to explain the successful industrialization of East Asian states. Each theory diverged in its emphasis in explaining why East Asia was successful in its pursuit of industrialization. There is a general consensus in the four perspectives on the institutional characteristics of East Asian nations as ‘export-oriented, state-led’ – albeit the lack of determinacy about the

21 precise nature of involvement and role of nation-states in fostering industrial development. It is ‘export-oriented’ since late-coming East Asian nations had to utilize their comparative advantage to concentrate on exporting. The move toward structural transformation, i.e., developing knowledge- and technology-intensive industries, came as East Asian states re-oriented themselves away from import- substitution toward export-orientation. It is ‘state-led’ since only the state has resources in the initial stages to provide the directional thrust to steer private capital for industrial upgrading.

The ‘neoclassical’, ‘developmental state’, ‘modernization’, and ‘public-private nexus’ perspectives have each been indispensable in explaining the economic development of East Asia. The four approaches outlined in this chapter have demonstrated the necessity to take into consideration the interaction between the public and private sectors that is central in developing a better understanding of why the East Asian economies achieved unparallel success in its industrialization.

In essence, the economic growth and industrialization in East Asia can largely be attributed to three institutions: a state-bureaucracy which oriented itself toward development, a market-friendly environment that allowed private enterprises to participate in economic development, and a high degree of dependency on technology transfer from multinational corporations. Over time, these three institutions created an alliance where the state set goals and recruited local and multinational firms in their effort to move up the value chain. The institutional foundations of ‘public-private alliance’, the evolution of such relationships and how it can be an effective framework to understand the process of technology development will provide the theoretical underpinnings for analysis in the next chapter.

22 Chapter Three: Conceptual framework

3.1 Introduction

This chapter proposes a conceptual framework titled ‘Triple-alliance’ to map the salience of and mutual dependence between three institutions: the state, the local capital and the multinational firms. It delineates the role of each participating institution in the context of development of the microelectronic industry in Taiwan, and discusses the interaction between the public and private sector to demonstrate how the ‘Triple-alliance’ facilitates technology transfer and absorption. The chapter concludes that understanding the interactions of these three institutions is essential for understanding how technology-intensive industries are developed in East Asian NIEs, and in particular Taiwan. Indeed, the process of technology accumulation is characterized by the combination of both direct-support and technology-sharing schemes.

3.2 Analytical framework

The ‘Triple-alliance’ model postulates that a nation’s industrialization is fundamentally underpinned by its institutional arrangements. In other words, different institutional arrangements in different countries would result in different patterns of industrializations. Three institutions are crucial in the creation of technology-intensive industries: the state, the local business groups and the multinational businesses. The concept of institutional alliance facilitates the understanding of how organizations within distinct institutions interact with each other for the purpose of knowledge- creation, and how they drive the specialization of industries in a given nation (Left, 1978). Indeed it is the network of institutions in the public and private sectors whose activities and interactions initiate, import, modify and diffuse new technologies (Biggart & Hamilton, 1988: 59-61).

23 The general consensus is that East Asian states initiated industrial growth, especially in technologically intensive industries, by interacting closely with the private sector and coordinating its development (Breznitz, 2005). Accounts on the development of microelectronic industry in East Asian NIEs highlighted this nature of interaction, demonstrating the coordinating role of the state in connecting the public and private sectors (Amsden & Chu, 2003; Cho & Matthews, 2000; Fuller, 2002; Hong 1997). In each account, the state essentially functions as the conduit that facilitates the technology transfer from advanced economics, via the process of absorbing, modifying and spinning-off various forms of knowledge to the domestic private sector. The state also performs the role of coordinator in the industry to ensure the spin-off technologies are shared equally among firms, avoiding monopolization of the technology by any one firm.

According to Lall (1996), the rapid growth of East Asian economies can be partially attributed to the strategy adopted by the state to amplify the interaction of foreign direct investment and local technological activity in building export competitiveness. Lall notes that the interaction among the state, local capital and multinational companies within the specific institutional set-up can invariably affect the strategy and structure of participating firms, and ultimately the economic performance of the state. Governments of successful industrialized East Asian nations performed more than good macroeconomic policies management; they deliberately shaped industrial upgrading (Lall, 1994). Simon (1992) noted that East Asian NIEs attempted to make effective use of foreign participation by viewing it as a means to overcome domestic limitations in advanced technologies. In addition to resource allocation and infrastructure improvements, there is a ‘temporal coherence’ that sees the industrial development as essentially a sequential one, where policies and projects adopted in the short term must provide maximum flexibility for adaptation and refocusing over the longer term. Such development strategy aimed to facilitate rather than preclude entry into higher value-added segments of the industries in the future (Simon, 1992).

Figure 3.1 summarizes the framework employed in this thesis for the analysis of technology creation. The ‘Triple-alliance’ model suggests that in order to understand the pattern of technology development and innovation of the nation, it is necessary to

24 understand how each major actor functions on its own, and interacts with its counterparts. In the case of East Asian NIE, the ‘Triple-alliance’ model exhibits two characteristics: the public sector, which sets the goals for industrialization, and the private sector, which includes both local and multinational business groups drafted by the state to carry out these goals.

Figure 3.1: The ‘Triple-alliance’ model

Role: Technology absorption Local Business and product development. Groups

Technology- intensive Industry

State Multinational Corporations Role: Policy design and Role: Supplier of timely implementation. market information and proprietary technologies.

3.3 Characteristics of the state

A singular feature of East Asia’s ‘Triple-alliance’ model is the autonomy of policymakers and the involvement of the state in the pursuit of industrialization. ‘Late developing countries’, like Japan, often adopted the strategy of allocating limited resources to promote certain strategic industries and national champions in order to effectively compete with more advanced economics allocating limited resources for the long-term goal of national industrialization (Johnson, 1982). Evans (1985) notes that the ability of the state to enforce such policies rests on two distinct qualities: autonomy and capacity. State autonomy refers to the relative autonomy of state

25 officials from dominant economic classes and social groups, which allow those officials to implement any policies they see fit. State capacity refers to the existence of a state apparatus that has control over sufficient resources, so it can offer incentives and facilitate the implementation of state policies (Evans, 1985).

In applying the framework to East Asian NIEs, Evans (1995) notes that, while multinational and local capital are indispensable to its industrial development, it is the state that is the dominant partner, since it is the only institution that actively pursues its agenda independent of any external interference. The state acts as the initiator of industrialization, and also possesses the wherewithal to catalyze particular outcomes it envisions, with the availability of policy instruments, financial and human resources (Evans, 1995).

Amsden (1989, 2001) argues that firms based in ‘late developing countries’ sought assistance from the state since it lacked proprietary technologies available to firms in advanced economics. The state helped the private sector absorb the advanced technology by providing subsidies that aimed to improve the productivity, R&D capability and ultimately values of its export goods (Amsden, 1989). According to Wade (1990), industrialization of Taiwan after the war was largely carried out by the state, whereby it did not merely implement functional macroeconomic policies as observed by neoclassical economics, but, more importantly, it intentionally targeted the development of certain specific industries, including the microelectronic industry (see also Lall, 1995). Wade (1990) argues that since the late 1950s, the Taiwanese government did two important things. One is the promotion of exports, and the other is selective promotion of certain industry specific products. The government has sought to change the composition of exports, and shift upwards into the new generation range of high technology engineering products, as well as create a solid base for information technology industries in Taiwan.

In summary, the ‘Triple-alliance’ model argues that the success of East Asian NIEs results from the state being entrepreneurial in its pursuit of industrialization. Had the state not been entrepreneurial by effectively involving itself to push for local industrialization effort, there would be no effective sponsor for industrialization in these late developing countries. For example, in the effort to establish indigenous semiconductor industries in Japan, South Korea and Taiwan, the state provided

26 massive injections of capital and monetary support to local corporations involved in the production of integrated circuits, as also actively soliciting and forming partnerships with multinational corporations to gain access to much needed production technology and product knowledge (Matthews, 1995, 1997; Simon, 1992; Weiss, 1995).

3.4 Characteristics of East Asian business groups

The second pillar in the East Asian ‘Triple-alliance’ model is the participation of local business groups. A business group is defined as set of legally separate firms bound together in persistent formal and informal ways (Granovetter, 1985). The question of how business group have arisen has been widely debated, and several views have emerged to explain its existence depending on the field of study (Chandler, 1977, 1990; Chung, 2001, 2005; Left, 1978). Chandler’s seminal work argues that firms that have grown into large complex business groups focused on making investments in areas of manufacturing, distribution and management hierarchy simultaneously, and with the development of these three segments, ‘business groups’ are formed along with commensurate ‘organizational capabilities’. Left (1978) focuses at the meso-institutional level, and argues that business groups evolve to cope with the underdevelopment of markets in which they reside. Chung (2001, 2005) shows that the emergence of business groups in Taiwan was the result of economic underdevelopment that forced firms to internalize various aspects of value added aspects of production, which eventually resulted in having operations in a wide range of industries (Chung, 2005). The two distinct views of how firms grow differ in their emphasis. Chandler (1990) focuses on the proactive nature of the firm, whereas others point out the responsive nature of the growth as a result of market failure between various value adding functions of the industry value chain. Though the focus is different, it is generally agreed that a firm grows into more complex vertical and horizontal structures when it makes full use of its existing resources and move into new product-markets.

In the case of East Asian NICs, business groups exist within in the political- economic context of the particular state in which it resides (Hamilton & Kao, 1987;

27 Numazaki, 1988, 1991; Orrus et al., 1997; Whitley, 2000). Hamilton and Kao (1987) see the success of East Asian industrialization as the result of the state fostering the growth of business groups which it then compels to take on specific industrial projects the state requires. Whitley (2000) proposes that the social and political context sets the parameter in which the business group can grow and operate through comparative studies on business groups across East Asian nations, and argues that business groups are intertwined with the political economic systems in the countries where they operate (Whitley, 2000). Numazaki (1988, 1991) sees the emergence of business groups in Taiwan as the byproduct of the complex network of firms that are based upon the geographical and historical ties of founding families. The collective studies define business groups in NIEs as a loosely coupled network of firms owned by the same individual or related persons who join together in multiple enterprises (Orrus et al., 1997).

In recent times, business groups have been recognized as an institution that is critical for the nation’s industrialization efforts; many authors have documented the role of large complex business groups in different countries (Amsden, 1989, 2001; Chandler, 2005; Kim, 1997; Morikawa, 1992; Schive, 1990). Amsden (1989) argues that conglomerates are often chosen to carry out the nation’s industrialization since it is the only institution that possesses capabilities in production, project execution and innovation. Amsden also notes that large business groups build up their organizational capabilities in incremental gradual steps, through the process of ‘learning’ at the factory floor level, which allows them to absorb ‘tacit’ elements of the knowledge, instead of giant leaps (see also Kim, 2003; Lall, 1995). Both Morikawa (1992) and Kim (1997) demonstrate through detailed firm-specific case studies that large complex business groups played an indispensable role during the initial industrialization and subsequent technological upgrading in Japan and South Korea, respectively. Indeed, large business groups like chaebols succeeded precisely because they were able to absorb complex technologies and achieve economies of scale through their vertically and horizontally integrated structures – something well- beyond the order of smaller firms (Lall, 1995).

History shows that successful firms in high-technology industries were those that were able to combine and coordinate technological knowledge with that of

28 product development, manufacturing, marketing and distribution into an integrated learning base (Chandler, 2005). The participation of local capital is necessary, since the state does not possesses any direct capacity in the production and manufacturing of goods. In the case of Taiwan, local business groups that invested in the microelectronic industry often consisted of multidivisional companies with a large number of product lines such as household appliances, telecommunications equipments, electronics products and computers (Schive, 1990). These firms, as part of multidivisional groups were able to develop their technological competencies through sophisticated ‘product-modification strategy’ that involved ‘scaling down’ of product features and designs in accordance to market conditions in less-sophisticated consumer markets. Thus, instead of emphasizing superior technologies as would multinationals from advanced economics do, firms from NIEs focus on technology adaptations and product modifications to produce less sophisticated products with much lower costs and prices (Schive, 1990).

3.5 Characteristics of multinational firms

The third entity, namely the multinational company (MNC), is the source for advanced technologies and customers of finished products from NIEs. The trade and investment flows catalyzed by multinational firms allowed firms in NIEs to quickly and effectively gain access to advanced technologies; moreover, the involvement of MNCs provides timely market information about foreign markets.

In the case of the computer manufacturing industry, many of the multinational electronic companies initially established assembly plants in East Asian NIEs, such as Taiwan, to take advantage of the relatively low cost environments and highly educated labor force. By 1984, electronics firms from Japan and US together accounted for 70 per cent of foreign capital in Taiwan (Chou & Kirby, 1998). The finished products were then exported mostly to the US and, to a lesser degree, Japan. In the case of East Asian NIEs, MNCs performed another important function as the linkage between NIEs and foreign markets. Ernst (2000) argues that in the case of Taiwan’s computer industry, the subcontracting relationships between MNCs in electronics and many of local computers assembling firms allows local firms to

29 market their products directly through these MNCs, which essentially function as the customer for electronic products produced by firms in NIEs (Ernst, 2000). Poon (1998) argues the global microelectronic industry has evolved to become a ‘horizontal network of manufacturers and suppliers’, and MNCs have largely contracted out the manufacturing activities to firms based in late developing countries. In addition, MNCs use consumer feedback in their home markets that provide timely market information, which firms based in late-developing countries utilize to their advantage to serve these markets.

3.6 The mechanism of foreign-owned technology transfers in Taiwan

Technology transfers from MNCs in electronics to domestic firms in NIEs exemplify the utility of ‘Triple-alliance’ model. Two characteristics stand out in the transfer mechanisms the government supervised: the establishment of industrial parks for firms that satisfy the state’s technological initiatives, and the collaboration on research and development projects between public and private sectors.

In Taiwan, the state’s commitment to industrial development and technology deepening is evident in the establishment of organizations and policies that allow the transfer of foreign-owned technologies. The mechanism of foreign-technology transfer combined both direct-support and technology-sharing schemes between public and private sectors. The state offered MNCs in electronics direct financial support in the form of tax holidays and import-export duty exemption in exchange for foreign investments. These incentives, in turn, attracted many MNCs in electronics to choose Taiwan as their investment destination. Of all the sources of investments, Japan has provided the largest proportion of foreign investments, due to its geographical proximity to Taiwan (Chen & Sewell, 1996). In order to effectively take advantage of foreign investments in Taiwan, the state began to establish ‘science- based industrial parks’. The first industrial park was established in Hsinchu in the early 1980s. Hsinchu, located at the northern region of Taiwan, was chosen due to its easy access to the international airport and its proximity to many research-intensive institutions, including National Chio Tung University, National Tsing Hua University

30 and the Industrial Technology Research Institute. When the first ‘science-based industrial park’ was established, it recruited mainly firms that focused on the production of microelectronic products, including: electronic peripherals, intergraded circuits and notebook computers. It has since widened its industry recruitment to include firms that focus on biotechnology, communication equipment, opto- electronics industries and precision machinery industries (HSPA Annual Report, 2000). Admission to the park is largely dependent upon the potential of a firm to contribute to technological development in Taiwan. The park provides generous financial incentives and logistical assistance to multinational firms inside the park, and in return, the firm must spend a certain proportion of its revenue on R&D and employ certain percentage of local scientists and engineers in the firm (Simon, 1992). The ‘science-based industrial parks’ have become a huge success since their establishment. They have attracted MNCs in electronics to invest in varieties of technology-intensive products, including computer peripherals, consumer electronics and integrated circuits (Tung, 2001).

In addition to establishing ‘science-based industrial parks’ to attract MNCs in electronics, the state also established institutions that encourage the collaboration of research and development between public and private sectors. The state agency ‘Industrial Development Bureau’ (IDB) and the state-funded organization ‘Industrial Technology Research Institution’ (ITRI) exemplify the government’s effort to foster domestic-multinational cooperation. The mandate of IDB is to assist the systematic upgrading of technological and organizational capabilities of the private sector in Taiwan (Matthews, 1995). IDB tracks the progress of the industry and monitors the performance of firms that reside in ‘science-based industrial parks’. These data- collection efforts allow the state to identify any shortcoming in its development strategies, and discard ineffective policies as quickly as possible.

While IDB largely focused on the effectiveness of its developing strategies, ITRI promoted the upgrading of industries through its sponsorship in vital R&D projects, then passing the results to private sectors. ITRI was founded as a non-profit research institution in the early 1970s by the government. Initially, ITRI was charged with the mission to provide technical support to state-owned companies. It has since evolved into a research institute with more than 5,000 researchers conducting R&D in

31 various cutting-edge technologies, including electronics and opto-electronics projects (Cheng, 2007). ITRI collaborate with IDB to search for advanced technologies around the world that it deemed necessary. Once IDB identified the technologies that fitted into the overall industrial development in Taiwan, it funded ITRI for the acquisition of the technology. Often ITRI used its own researchers to modify the advanced technologies it acquired from abroad, and modified the technology to suit the domestic needs before passing these technologies to the local firms for mass production (Tung, 2001).

The transfer of foreign technologies to domestic firms in Taiwan is largely considered to be the prime reason Taiwan was able to become on of the largest producer of microelectronic products in the world. By the early 1990s, total foreign direct investment in the electronic sector reached US$1.2 billion compared to a mere US$370 million when the state first initiated the projects to upgrade Taiwan’s technologies bases (Tung, 2001). The success can be largely attributed to the various organizations discussed in this section, which function primarily as intermediaries between the local and multinational firms and provide research support and financial incentives to the respective parties. These intermediaries form the linkage between local and multinational firms which solidify the structure of ‘Triple-alliance’.

3.7 Conclusion

The ‘Triple-alliance’ model demonstrates that industrial development in East Asia was catalyzed by the interaction of three major socio-economic institutions: the state, the indigenous business groups and multinational firms. The evolution of Taiwan’s microelectronic industry underscores the importance of these three institutions in driving the nation’s continuous industrial upgrading and technology deepening. The state formulated industrial policies via public institutions in ways that were constructive to the comprehensive development of microelectronic industries in Taiwan; local business groups fostered domestic capabilities in developing advanced technologies; and the linkage with MNCs allowed effective coordination of demand and supply between firms in NIEs and overseas markets.

32 In conclusion, the ‘Triple-alliance’ model introduced in this chapter highlights a particular division of labour among these three institutions. The utility of this approach is that it allows detailed examination of the process of foreign technology transfer and domestic industry development at the institutional level. It demonstrates that the successful creation of the new industry is achieved within the context of the dynamic interactions among both market and non-market institutions. The next chapter will trace the historical development of these institutions with the application of the ‘Triple-alliance’ model.

33 Chapter Four: The development of the microelectronic industry in Taiwan

4.1 Introduction

This chapter presents the development of the microelectronic industry in Taiwan through the lens of the ‘Triple-alliance’ framework. Taiwan’s emergence as one the largest global electronic producer has been well documented4. By 2006, Taiwanese- affiliated manufacturers produced 80 per cent of desktop and notebook computers worldwide (Einhorn, 2006). The huge production volumes created the need for local firms to have steady supplies of flat-panel displays. Thus, flat panel display ‘TFT- LCD’ devices became a critical component for many of the computer and electronic manufacturers, since it was the major display device in notebook computer accounting for 30−40 per cent of production costs (Optolink Report: Jan, 1994).

The chapter begins with a brief review of the development of microelectronic industry in Taiwan from 1960 to 1990. The development of electronic hardware started with the development of consumer electronics during 1961 and progressed to form various linkages with multinational electronic firms. These linkages proved to be critical in later stages as Taiwan sought to develop its indigenous TFT-LCD industry. By the late 1980s, Taiwanese manufacturers became some of the largest subcontractors for large multinational electronic firms. The decision to produce TFT- LCD domestically was the result of changing external circumstances. Three factors led both the state and electronic firms in Taiwan to invest in the TFT-LCD industry: the increasing trade deficits with Japan, rising competition from Korean electronic firms and constant shortages of flat panels that led to production delays.

4 For a comprehensive treatment on the development of the electronic industry in Taiwan refer to following: Amsden & Chu (2003) Beyond Late Development. Cambridge: MIT Press USA; Chung, Dieter Ernst (2000) Inter-organizational knowledge outsourcing: What permits small Taiwanese firms to compete in the computer industry? Asia Pacific Journal of Management Vol. 17, Issue 2; Jason Dedrick & Kenneth L. Kraemer (1998) Asia's Computer Challenge: Threat or Opportunity for the United States & the World Oxford University Press.

34 The chapter utilizes the ‘Triple-alliance’ model as the analytical framework for understanding the path of development in Taiwan’s microelectronic industry. Based on the ‘Triple-alliance’ model, the chapter provides a prelude to the hypothesis in Chapter 5 that the initial unsuccessful attempt to create the TFT-LCD industry was largely due to the absence of multinational electronic firms, and its subsequent success was due to the alliance of and interactions between three institutions – the state, MNCs and business groups.

4.2 The development of the microelectronic industry (1960s−1980s)

Taiwan’s first foray into the electronic sector was in the creation of low-cost assembly sites with the establishment of ‘Export-processing Zone’ by the Ministry of Economic Affairs. The state bureaucracy saw the opportunity to create an industrial zone by utilizing low labor cost and tax holidays to attract multinational firms to set up their manufacturing facilities. This arrangement oversaw the state providing necessary infrastructure with tax holidays to attract investments from multinational firms (Li, 1995).

In 1965, the Executive Yuan authorized the creation of such zones by passing the act ‘Law on the Establishment and Management of Export-processing Zone’. (Gold, 1986). The condition of entry was based on meeting certain criteria, such as hiring and training local staff for assembly and sourcing components locally. This created an opportunity for technology diffusion to local firms, thus beginning Taiwan’s movement towards knowledge-intensive industries. The original scheme of ‘Export-processing Zone’ attracted many Japanese firms who, as a response to rising wage pressures at home, were on the lookout for geographically proximate low-cost locations with skilled labor force.

The early development of Taiwan’s electronic industry revolved around a few loosely coupled technical agreements between Japanese radio and television manufacturers and Taiwanese firms (Gold, 1986). Local firms began by supplying commodity goods, such as television cabinets, taking advantage of local content

35 requirements, and only gradually they learning to make components like transistors and picture tubes on their own (Gold, 1986). As these domestic enterprises cooperated successfully with multinationals from Japan and the US to assemble light consumer electronic goods, local business groups, such as Tatung Electronics, began to take more active interest in the electronic industry. Joint-venture agreements became a preferred method of business between local and multinational electronic firms, with many local business groups signing joint-venture agreements with leading multinational electronic firms in the US and Japan.

The developmental strategy based on utilizing low labor costs for exports, however, soon faced problems of rising labor costs and volatile international markets. The first global oil crisis in 1972 and 1973 prompted the creation of Economic Planning Council (EPC) for the task of comprehensive economic planning. EPC announced comprehensive infrastructure upgrading, widely known as ‘Ten Major Projects’ (Appraisal for Ten Major Development Projects, 1979) The project included the construction of logistic infrastructures, a heavy chemical industries complex and nuclear power plants to ensure the infrastructure capacity in Taiwan met the demand for a booming economy at the time (Appraisal for Ten Major Development Projects, 1979). The objective of this project was to both develop physical infrastructure and upgrade the overall industry structure by shifting the focus to technology-intensive and capital-intensive industries (Appraisal for Ten Major Development Projects, 1979).

As the popularity of joint-venture agreements between multinationals and local firms spread, EPC decided to take a more comprehensive approach in managing technology diffusion among local firms effectively by formulating its very first technology policy. In 1976, the ‘First Five Year Science & Technology Policy’ was officially announced. The initial ‘First Five Year S&T Policy’ placed major attention to and resources on cultivating manpower and promoted basic scientific research through national universities (Li, 1995).

The government further strategically intervened to move Taiwan into the microelectronic industry by merging EPC with the ‘Finance and Economic Committee’ to form the newly created ‘Council for Economic Planning and Development’ (CEPD). CEPD not only faced the task of economic management, but

36 it was also given the financial resources to implement any industrial project it deemed necessary. In light of its new mandate, the newly formed CEPD consulted many international organizations in regard to the industry upgrading in Taiwan. Various leading international research institutions consulted CEPD in regards to which industries should be the focus of upgrading. The choice of Arthur D. Little, the American consulting firm, is the result of the firm representing various academic institutions, including Stanford University, Massachusetts Institute of Technology and Xerox Plat Alto Lab. In 1979, Arthur D. Little Inc. advised CEPD to upgrade the industrial technologies through investments in six sectors: consumer electronics, desktop computers, heavy machinery, locomotives, semiconductor and petrochemicals (Gold, 1986). These industries were chosen because they had huge linkage effects on other industries with large market potential, high-technology intensity with high value added, and low energy consumption with low pollution effect (Appraisal for Ten Major Development Projects, 1979). In 1979, consumer electronics and desktop computer industries were targeted for development by CEPD as ‘strategic industries’. Both industries were seen as generators of ‘high value-added’ goods for the future, offering great potential for product linkages with other peripherals that included display monitors, integrated circuit, software, and a host of other peripheral equipment (William, 1997).

In 1982, the National Science Council unveiled the ‘Second Five Year S&T Policy’. In the policy, it stressed the importance of having suitable infrastructure for Taiwan’s industrial development. The previous ‘Export-processing zone’ model was adapted and modified to become the ‘Science-based industrial park’ (Li, 1995). There were two unique characteristics that distinguished the ‘Science-based industrial park’ from ‘Export-processing zones’: first, it focused exclusively on manufacturing consumer electronics and desktop computers; second, firms that were admitted to the ‘Science-based industrial park’ were based on systematic screening of their production characteristics and market potential (Chu & Hu, 1999). The first park, Hsinchu Science-based Industrial Park (HSIP), soon became a favorite destination of major semiconductor firms with its close link to major ports and proximity to ITRI. The congregation of academia and industry also created a community of entrepreneurs, researchers and venture capitalists that allowed for dynamic collaborations for industry upgrading (Hsu & Saxenian, 2001).

37 As the government designated the development of microelectronic sector as the strategically important industry in 1979, and offered financial and regulatory incentives to induce investments, local business groups began to take advantage of such incentives and invest in the microelectronic industry. Initially, production of desktop computers was hamstrung by the lack of manufacturing know-how and high degree of product unfamiliarity in the domestic market. This led many local firms to strike alliances with multinational electronic firms from Japan and the US to function as subcontractors (Chu, 1999).

The new role assumed by Taiwanese firms as the low-cost, manufacturing- subcontractor gave the rise to the new business model, known as the ‘Original Equipment Manufacturer’ (OEM). In OEM, Taiwanese subcontractors took on sophisticated manufacturing roles at the behest of Japanese and US based multinational electronic firms, and allowed these firms to concentrate their resources on brand building and product research (Miau, 1992). As Taiwanese firms took on the OEM activities, multinationals were willing to subcontract assembly to Taiwanese firms and focused resources solely on marketing. The main competitive advantage of Taiwanese firms rested on their capability to manufacture PCs at low cost and acceptable quality. This led to the emergence of clusters of manufacturers that catered to different product-markets, including monitors, notebook computer, printers and other peripheral equipment. Such OEM firms also began to attempt to design prototype products for manufacturing. Upon winning approval by the principal for such prototypes, they offered such products to consumers under multinational brand names (Ernst, 2000). Some OEM firms also went a step further to offering service in logistics and after-sales services for its vendors. Such division of labor allowed multinational electronic firms to focus solely on their firm-specific resources in product distribution and marketing. This new evolutionary relationship catapulted OEM firms to become ‘Original Design Manufacturing’ (ODM) firms and ‘Original Brand Manufacturing’ (OBM) firms (Ernst, 2000). Ernst notes that Taiwanese firms adopted two generic strategies in entering PC business. First, they acted as a subcontractor, where Taiwanese firms source components from outside and assemble them in their own factories. Second, they acted as an ‘Original Design Manufacture’, where Taiwanese firms may combine their own designs to produce equipment tailored to customers’ demands. The major difference between these two strategies is whether

38 the product bears the assembler’s own brand name or the name of its business customer (Ernst, 2000).

By 1989, OEM activities in Taiwan represented 40 per cent of total exports and propelled Taiwan to become one of the world’s largest microelectronic goods subcontractors. The accumulation of manufacturing capabilities by Taiwanese OEM firms created an interdependent relationship between multinationals and Taiwan- based OEM firms. By early 1990, 10 large Taiwan-based OEM firms accounted for close to 70 per cent of electronic hardware production globally. This interdependency allowed Taiwanese firms to utilize their scale for horizontal diversification into electronics-related products (Amsden & Chu, 2003). A group of OEM firms, including Acer, ADI, Compal, Chuntex Electronic, First International Computer, GVC, Invetec, Lite-on and Mitac, emerged with annual sales of more than US$500 million for each firm. (Einhorn, 2006). The case of Mitac Inc. confirms the significance of the OEM model. Mitac began as an OEM firm in personal computers in the late 1970s, and by using linkages with multinational electronic firms it diversified into manufacturing handheld electronic devices in the early 1990s. It paid off handsomely and Mitac became the world’s largest handheld electronic device producer in 2002 (Mitac Annual Report, 2003). Large Taiwanese OEM firms, such as Compal, Invectec and Quanta, imported critical components including display monitors and semiconductors directly from multinational electronic firms in Japan (Wang, 1995).

4.3 Motivation for the creation of and indigenous TFT-LCD industry

The creation of the indigenous flat panel display industry in Taiwan was the result of a number of external and internal circumstances prevalent at the time. The motivation to produce indigenous TFT-LCD in Taiwan was the result of three intertwined factors. First, the continual trade deficits with Japanese electronic firms compelled the state to localize the production of necessary components. Second, the emergence of Korean electronic firms at the expense of Taiwanese firms led local electronic firms to seek alliance with Japanese multinational firms in order to more

39 effectively respond to the increasing competition. Third, the rising demands for desktop and notebook computers created huge demands for flat panels that led to the investment in TFT-LCD industry (IEKC, 1997). The shift towards TFT-LCD industry can be divided into two distinct phases. The first phase was marked by the failure of the industry due to the lack of involvement of multinational firms. The second phase saw its successful establishment due to the alliance of three major institutions: the state, local business groups and multinational electronic firms (Developments and Prospect for the FPD Industry, 2003).

4.3.1 Trade deficit with Japan

Since the 1960s, many Japanese electronic firms began to relocate their production facilities to Taiwan due to the geographic proximity and low production costs. By relying on imported technologies from Japan, Taiwanese manufacturers were able to become among the largest global exporters in calculators and televisions. The reliance on Japanese technologies and components deepened further as Taiwanese firms moved to new product categories, such as personal computers and integrated circuits (Hung, 2006).

Much of this reliance resulted from the import of key components and equipment, which together generated US$10 billion of total trade deficit annually by 1990. The value of flat panel display alone accounted for US$2.1 billion (IEKC, 1995). The main cause for such huge trade deficits was the lack of technical know- how and production capacity of Taiwanese firms. Local TV and monitor producers, such as Tatung, still had to rely on Japan’s Sony and Hitachi for the supply CRT. Throughout the mid-1990s, the trade imbalance between the two nations was lessened but not totally erased. This bottleneck became the main motivation for both the state and the private sector in Taiwan to begin the effort in developing TFT-LCD locally (Tsai, 2006).

40 4.3.2 Emergence of Korean electronic firms

The entry of South Korean electronic firms into TFT-LCD industry became another major motivation for Taiwan to build its own TFT-LCD production base in order to compete effectively with South Korea-based electronic firms. The financial crisis in 1997−98 that swept through many Asian nations, including South Korea, was in many ways a blessing in disguise for Korean TFT-LCD manufacturers. Even as large-scale firms like Samsung, LG and Hyundai began to invest in LCD in the mid- 1990s, the subsequent financial crisis led to the collapse of the Korean Won that allowed these firms to export TFT-LCD devices at extremely competitive prices compared to Japanese and U.S-based competitors (Optolink Reports, 1999). This event coincided with the successful development of 11−12 inch AMLCD that could be applied to notebook PCs that allowed Korean participating firms to reap windfall profits. For large Korean electronic firms, the TFT-LCD industry was largely seen as a means of diversifying away from their core business in memory chips and integrated circuits, while at the same time applying their manufacturing competencies from this line of business. The entry into the TFT-LCD industry also allowed these firms to secure steady supply of TFT-LCD for their own electronic and information technology products (FEER, May 1998).

The Korean electronic firms’ entry into the TFT-LCD business since the mid- 1990s was due to the strategy Korea adopted to leapfrog Japanese competitors in terms of technology (Chang, 2008; Cho & Matthews, 2000). Korean electronic firms leveraged their resources and knowledge to enter into the production of memory chips, a crucial component in every electronic device, which allowed subsequent production of other electronic products (Cho & Matthews, 2000). Korean electronic firms also adopted the strategy of ‘parallel development’. This strategy called for the simultaneous planning, design and construction of production facilities (Chang, 2008) that allowed Korean firms to initiate the development for next-generation products while mass-producing the current generation of product simultaneously. The 1997 Asian financial crisis did not slow the pace of investments by Korean electronic firms. The continual investments by large electronic firms during 1997 attests to the strength and resources Korean electronic firms possessed. Korean electronic firms, such as Samsung Electronics, spent hundred of million of dollars investing in TFT-LCD

41 throughout the crisis. The case of Samsung Electronics demonstrates how local business groups have played an indispensable role in establishing Korea’s dominant position in TFT-LCD sector. Large domestic business groups made substantial investments into TFT-LCD industry. Following closely behind Samsung Electronic, Daewoo and LG both built their first pilot production line in 1994 and soon began production (Business Week, Feb 15, 2006).

Besides the involvement of the private sector, government agencies, such as the Ministry of Trade and Industry and the Ministry of Science and Technology, actively coordinated the push into the display industry and aggressively increased funding for research. The result of such active investment was evident in the high production volumes that Korean electronic firms achieved in a relatively short time. As Murtha (2001) showed, in 1996 only one Korea-based electronic firm had any significant production yield, yet by 1999, two Korean firms became the largest and second largest TFT-LCD producer globally. As a result of state and business cooperation, Korea emerged as the largest TFT-LCD producer globally in 2001.

4.3.3 Rising demand for Information Technology products

The need to secure a steady supply of TFT-LCD increased as Taiwanese firms became among the largest global suppliers of PC and a host of other electronic products (IEKC, 1999). The trade deficit with Japan and the emergence of Korean firms provided strong lessons for Taiwan to seek the localization in the production of TFT-LCD, since it was estimated that the display monitor accounted for roughly 30 per cent of production costs in each PC (IEKC, 1999). The previous boom in the electronic sector was now threatening to become constrained by the slightest disruption in the production of TFT-LCD by Japanese and Korean counterparts (TOSEA, 1999). In addition, due to huge demand and limited supplies, the continual price hike of TFT-LCD also stifled the growth of Taiwanese microelectronic industry as a whole (TOSEA, 1999).

Thus what ultimately pushed both the public and private sector in Taiwan to localize the production of TFT-LCD was the huge demand for flat display panel to be needed in all kinds of information technology products. Table 4.1 shows that the

42 productions of microelectronic products have come to represent 30−40 per cents of GDP in Taiwan by 1995. This coincided with the trend of booming global demand for information technology products.

Table 4.1: Taiwan’s electronic manufacturing industry

1990 1991 1992 1993 1994 1995 Value of $17.9 $20.3 $22.2 $19.1 $20.9 $24.9 Electronic Exports (US$ billion) Electronic 12.7 13.7 13.6 14.4 15.1 18.4 Mfg/GDP (%) Electronic 20.3 19.4 19.8 20.7 21.4 23.3 Exports/GDP (%) Electronic 14.7 14.8 16.0 16.4 17.1 18.4 Imports/GDP (%) Source: William R. Boulton (1997) WTEC Panel Report on Electronics Manufacturing in the Pacific Rim. World Technology Evaluation Center. May, 1997.

Table 4.2 demonstrates that global demand for information technology products had risen exponentially from 1995 to 2002 in every product category, and implies that investments into TFT-LCD were no longer a risky venture.

43 Table 4.2: Worldwide increase in Information Technology products (million units)

Product Category 1995 2002 Consumer Electronics 78.9 218.1 Desktop Computer 48.3 128.0 Digital Cameras 8.8 18.9 Handheld GPS Navigator 3.9 16.1 Mobile Handsets 1.8 9.6 Notebook PC 11.4 37.7 LCD TV 2.7 4.8 Totals 155.8 434.2 Source: IEKC, 2003.

The rising demand for TFT-LCD devices to supply all information technology products forced bureaucrats and executives to re-evaluate its previous stance on sourcing LCD directly from Japan or Korea. The ex-head of IDB, Mr. Yin, captures the new thinking of the state and business:

Every consumer of electronic products needs a display panel. Taiwan’s large original equipment manufacturing business means that firms require large supply of display panels and because of our subcontractor status, it means that product margins will always remain low. This is why we must control costs of upstream critical components to make profit. Building our own TFT-LCD site means that Taiwanese firms can enjoy higher profit margins down the road. (Tian-xie: March, 1997)

4.4 Contribution of the state

The contribution of the state in the establishment of TFT-LCD industry in Taiwan can be evaluated along two dimensions: industry-specific industrial policy, and the provision of infrastructure and human capital. The industry-specific policy refers to a series of state policies that evolved over time to sponsor the development of TFT-LCD industry in Taiwan. The state sponsored the establishment of science-based industrial parks across Taiwan and funded research projects that specifically targeted flat panel display technologies. These supply-side measures in turn induced

44 investments from the private sector. The establishment of science-based industrial park allowed firms to cluster around the major producer of flat panel to maximize efficiency and productivity (OSL, 2000). The availability of engineers and researchers from public institutions allowed the private firms to expand production without having to concern skill shortages thus attenuating coordination failures between product and factor markets (Lall, 1995). In turn, such developments enabled Taiwanese firms to focus their resources on building absorptive capacity for product development for commercial purposes, instead of reinventing the wheel to develop advanced technologies.

4.4.1 Industry-specific policy

Industrial policy is defined as a government initiative to ‘alter industrial structure to promote productivity-based growth’ (World Bank, 1993: 304). Noble (1998) argues that industry-specific policy induces an ability to absorb and utilize foreign technology as one of the key determinants to late-comer nations’ success in technology-intensive industrialization (Noble: 1998). Lall (1999, 2005) underlies the importance of the government in Taiwan by pointing out that Taiwan’s narrow industrial base due to its heavy dependence on small and medium sized enterprises (Lall, 1999) was mitigated by state bureaucrats who established an efficient system for technology transfer and absorption by linking foreign investors with local firms (Lall, 2005).

In 1996, the Ministry of Economic Affairs began to advocate the need to upgrade manufacturing industries in a series of its policy papers. One of the policy papers was titled ‘Industrial Technology Developing Strategy’, in which the TFT- LCD industry was picked as a strategic industry and placed under the special status for development, since it had deep linkages with three strategic industries: communications, information technology products and consumer electronics. It was also deemed to be highly relevant to the development of two adjacent industries: semiconductors and precision machinery and automation (Industrial Technology Developing Strategy, 1998). Taiwan’s bureaucrats and state-sponsored organizations, such as CEPD, ITRI and MOEA, played an instrumental role in leading the

45 development of TFT-LCD after examining the performance of the previous 1996 policy ‘Development of the Ten Emerging Industries.’ Bureaucrats from these organizations were instrumental in solving the basic infrastructure needs of LCD manufacturers even as ITRI transferred its in-house technology know-how to local firms at the incipient stage. In 2000, it announced the ‘Amendment of Statute for Industrial Upgrading’ that outlined five emerging industries the state aimed to develop5. By 2001, the state decided to target TFT-LCD industry as one of the key sectors in the overall development of the microelectronic industry in Taiwan. The decision was officially announced as the ‘Two Trillions and Twin Stars Industries Development Plan’– a plan that focused on reaching US$10 billion sales in the TFT- LCD industry by 2006.

The state-sponsored development of various LCD related materials to achieve the output plan in the TFT-LCD industry included development of glass substrates for the display panel, photolithographic materials for color and lighting and liquid crystals (Optolink Reports, 1999).

4.4.2 The creation of TFT-LCD specific industrial park

The state’s provision of both hard and soft infrastructure for the TFT-LCD cluster was one of the important features of the state’s leadership role in the alliance. Two science-based industrial parks – one in central, and the other in southern Taiwan − were created to specifically develop the TFT-LCD industry. The creation of TFT- LCD-specific industrial parks allowed the state to agglomerate all ancillary firms under one location. The aggregate demand for flat panel displays by various related industries, including desktop monitors, notebook computers and LCD TVs, created a readily available market for TFT-LCD producers and their upstream components suppliers. The creation of these science-based industrial parks became an indispensable conduit for the development of the TFT-LCD industry. Matthews (1997, 2003) has attributed the successful development of the microelectronics industry in

5 These five emerging industries are: Biotechnology, Communication Equipments, Display Technologies, Media Contents, and Semiconductor Manufacturing. TFT-LCD belongs to the category of Display Technologies. See: National Development Outline for High-valued Industries. Published by Ministry of Economic Affairs, Taiwan. in April, 2008.

46 Taiwan to the ‘industry cluster’ that is largely similar to the Silicon Valley. Such science-based industrial parks fostered firm-level cooperation and innovation (Matthews, 1997) and, in recent times, have evolved to a major mainspring for organizational learning (Matthews, 2003).

One of the major parks, ‘Tainan Science-Industrial Park’ based in Southern Taiwan, displayed the vigorous involvement of the state to foster the development of ‘emerging industries’ (Industrial Technology Developing Strategy, 1998). By 1999, 29 TFT-LCD related firms were setting up or waiting to be set up at all the science industrial parks. The aggregate investments committed by these firms totaled US$5.5 billion and amounted to 43.28 per cent of total investments in science industrial parks (Optolink Reports, 2000). Table 4.3 shows the total number of firms in each sector from 1995 to 2005. The majority of firms come from TFT-LCD related industries.

Table 4.3: Tainan Science Industrial Park investments overview

Industry Total Number of Total Number of Total Number of Firms in 1995 Firms in 2000 Firms in 2005

Biotechnology 3 44 74

Communication & 4 32 37 Consumer Electronics

Integrated Circuit 7 52 49

Precision 1 40 143 Machineries

Opto-electronics 5 67 140 Sources: Southern Taiwan Science Park Management Bureau, various years.

Table 4.4 shows the investment figures of Tainan Science Industrial Park in 1996. The table demonstrates that the establishment of the industrial park has created an industry cluster where local and multinational electronic firms have formed alliances to invest in the production of TFT-LCD together.

47

Table 4.4: Survey of TFT-LCD firms in Tainan Science Industrial Park in 1996

Name Capitalization in Product Technology Source US$

3M Taiwan Components & 780 million 3M, USA. Optronics Color Filters

Chi Mei Electronic Fujitsu, Japan. ITRI, 1.7 billion Components Display Taiwan.

Chi Mei TFT-LCD Fujitsu, Japan. ITRI, 2.4 billion Optorlectronics manufacturing Taiwan.

Corning Display TFT-LCD Glass 2.3 billion Corning, USA. Taiwan Substrate

Delta Electronics 450 million CCFL ITRI, Taiwan.

NEC Lighting CCFL and 792 million NEC, Japan. Taiwan Components

Panasocie Photo Components and 853 million Matsushita, Japan. Lighting Taiwan Machineries

Quanta Display 1.1 billion TFT-LCD Sharp, Japan. Inc.

Kenmos Tech 1.1 billion TFT-LCD Toshiba, Japan.

Taiwan Semiconductor Philips Electronic, 939 million Driver IC Manufacturing Netherlands. Corp

United Microelectronic 1.5 billion Driver IC REC, USA. Corporation Source: Report on inbound investments in 1996, Southern Taiwan Science Park Administration Office.

The rapid growth of the TFT-LCD industry underlies the significance of the state in the ‘Triple-alliances’ framework. The establishment and management of science-based industrial parks were justified by the logic that competitive industries could not exist without equally competitive infrastructures. The development of these

48 ‘science industrial parks’ had profound impact on both local and multinational firms. In 2000, the state unveiled the ‘Five Year Plan’ seeking to become the largest TFT- LCD producer in Taiwan. The plan called for the establishment of upstream LCD material firms to complement downstream LCD-manufacturing firms (Optolink Reports, 2001). This resulted in clusters of component suppliers and precision machinery producers that invested alongside major TFT-LCD producers inside the industrial park.

4.4.3 Provision of human resources

Clearly, the provision of human capital played a very important role in the creation of the microelectronic industry. It is estimated that from 1960 to 1990, Taiwan produced more than two million engineers. This was largely the result of the state policy that fostered industry-applied courses and encouraged cooperation between university and private industries (Hung, 2006). The quality of engineers produced by Taiwanese education is shown in the ‘Productivity and Pay’ comparison between Taiwan and other advanced economics. Based on the Global Competitiveness Report, Taiwan is in second place in the global ‘Productivity and Pay’, meaning that while productivity per engineer is high, the wage still maintain relative low comparing to other countries (Global Competitiveness Report, 2005).

The rapid development of TFT-LCD industry created high demand for engineers and R&D personnel. The state responded to the industry’s demand for suitable human resources in several ways. First, it allowed R&D researchers in public or quasi-public institution, such as ITRI, to move to the private sector without imposing any sorts of restrictions in technology transfer. The free flow of researchers and engineers offered several benefits. Second, it allowed public institutions, such as ITRI, to receive income in the form of royalty payments from the private sector when it hired engineers from them and used its technology for its product development. Third, the royalty income ITRI and other public institutions received were then ploughed back for more advanced research that would allow further upgrading of the industry (William, 1997).

49 In recent times, the state has taken up more basic R&D research and allowed the private sector to concentrate its limited resources on conducting manufacturing or processed-related research activities. A clear division of labor has emerged to distinguish between public and private sectors on their specific role in R&D. The public-funded organizations, such as ITRI, are increasingly focused on the research of most advanced technologies and products. In the case of display technologies, ITRI increasingly focuses on more advanced technologies, such as LED (Developments and Prospect for the FPD Industry, 2003), while individual firms are focusing on the development and commercialized end where emphasis is on redesigning of the manufacturing processes to allow low-defect production (Hung, 2006). Clearly without adequate supply of human resources, any plan to develop the TFT-LCD industry would not be effective. The state provided both engineers and researchers to ensure enough manpower for the development of TFT-LCD industry, which became one of its major contributions to the development of the industry (Taiwan’s Flat Panel Industry, 2008).

4.5 Conclusion

This chapter describes how the development of the microelectronic industry in Taiwan followed an evolutionary path. It began with low-cost and low-skilled assembly and gradually moved up the value chain ladder for skill-intensive design and manufacturing. This evolutionary path shows that in order to successfully move up to the technology frontier, domestic institutions must form a coalition with multinational companies, since one of the most common ways to compensate the shortcomings of domestic institutions is to form alliances and coalitions with multinational companies who are at the technology frontier.

The state thus played a very critical role in facilitating the development of technology-intensive products. As discussed in the previous two chapters, the state performed various functions to assist private sector in developing technologies. This chapter shows that the state performed three specific functions to assist the development of TFT-LCD industry: setting industrial policies, providing infrastructure and providing human capital.

50 This chapter uses the ‘Triple-alliance’ model as the analytical framework to show that the successful creation of the flat panel display industry is the result of complex web of linkages and alliances between the state, local business groups and multinational electronic firms. It demonstrates that mere cooperation between the public sector and domestic firms would not have been sufficient to produce technology-intensive products, since neither the state nor local firms possessed any proprietary technologies or production knowledge. The next chapter will focus on the role of the local business groups and multinational electronic firms.

51 Chapter Five: Firm-specific case studies

5.1 Introduction

In the previous chapter, the contribution of the state to the development of TFT- LCD industry was highlighted. This chapter presents mini case studies that pivot around the contribution of two other institutions towards the advancement of the TFT- LCD industry in Taiwan: local business groups and multinational electronic firms. In essence, while the state performed the role in leading development through a mix of policy-designs and resources-allocation, the local business groups, with the technological assistance from multinational electronic firms, mobilized the available resources in producing the products. The thesis argues that in the absence of alliance, progress of TFT-LCD industry stalled; by contrast, development took off when alliances began to form among the latter two institutions.

The material for case studies presented in this chapter came predominately from publicly available sources, including annual reports of firms, business journals and news clips, government-backed research reports and statistics, trade publications and the interviews conducted by the author during 2008−2009 (see list of interviewees in Appendix 1). Semi-structured interviews were used to explore interviewees’ views regarding the motivation of each institution in entering and forming the alliance. The subjects were purposefully selected from both public and private domains to better represent the view of each institution. Appendix 2 contains the information and letter of consent provided to participants.

5.2 Case study methodology

The adoption of the case study method is appropriate in understanding the institutional context that exists for technology diffusion and absorption between multinational and late-developing firms. An effective use of the case studies approach can capture the link between the institutional context and the strategic role of the firm

52 (Ray, 1998). The literature suggests that the choice of research method should be determined by the research question presented (Campell et al., 1982). This thesis examines institutional and organizational linkages that led to the successful establishment of flat panel industry among late-development firms in Taiwan, thus the choice of exploratory field-based studies that employ a qualitative approach is appropriate, as proposed by distinguished researchers in the field of organization and strategy studies (Pettigrew, 1990; Van de Ven, 1992). In addition, previous studies that explored relationship between institutions and innovation have also employed case study, along with statistical analysis to demonstrate the validity of the case study method (Akinwande et al., 2003; Linden & Murtha, 1998; Murtha, 2001).

The case study research strategy has the capacity to allow researchers to achieve an understanding of what may be a complex problem through an in-depth investigation of one case, or in-depth investigation of multiple cases (Eisenhardt, 1989, 1991; Pentland, 1999; Yin, 1989). Eisenhardt (1989) stated that the case study approach is a research strategy which focuses on understanding the dynamics present within signal settings. This method of research typically combines multiple data collection methods, including archives, interviews, questionnaires and observations (Yin, 1989). Insights derived from the case study can accomplish several goals, including providing description and testing or generating theory (Eisenhardt, 1991).

The primary data source is the annual publication of ‘The Ranking of Business Group in Taiwan’ by China Credit Information Service (CCIS). The publication CCIS surveys the largest 100 business groups in Taiwan (see Appendix 3) and is widely used as a major source of data and information. CCIS defines Taiwanese Business Groups as a cluster of three or more related corporations that mutually acknowledge their common memberships and have combined sales of at least US$500 million (CCIS, 2000).

Thus the three business groups selected for in-depth studies were Acer, Chi Mei and Formosa Plastic Group. Each case study presents the historical background of the firm and its rationale for investing in the TFT-LCD industry and ends with an analysis of why one group failed and the other groups succeeded.

53 Acer was incorporated in 1976 and reached sales of US$500 million by 1999 (CCIS, 2000). Acer began investing in TFT-LCD industry in the early 1990s and later merged with two other firms to create the largest LCD producer in Taiwan. The two other firms in the case studies, ‘Chi Mei’ and ‘Formosa Plastic’, both originated in petrochemical industry but specialized in different product sectors. Formosa Plastic Group specialized in the production of polyvinyl chloride (PVC); while Chi Mei specialized in acrylonitrile butadiene styrene (ABS). By 1990, both became one of the largest business groups in Taiwan.

5.3 First round of consortia (1990−1998)

5.3.1 Background of first consortia

The first round of TFT-LCD consortia involved the individual participation of five firms from 1990 to 1995. The development in this phase was constrained by the lack of technology know-how. By the early 1990s, Japanese multinationals became the largest suppliers of LCD panels to Taiwanese electronic manufacturers. The continual high demand gave Japanese multinationals incentives to discourage the entry of potential Taiwanese LCD manufacturers, by adopting the strategy of supplying LCD panel at below market prices to Taiwanese electronic manufacturers (Optolink Report, 1994). The low price ensured many Taiwanese firms scrapped plans to produce LCD on their own, relying instead on Japan for the supply. Since Taiwanese electronic manufacturers usually placed huge demands for LCD panels it allowed Japanese multinationals to remain profitable.

The strategy of Japanese multinationals did successfully discourage the Taiwanese electronic manufacturers from making investments in the production of flat panels, since Taiwanese firms came to realize that such ventures would prove technologically difficult, risky and costly to implement (The Economic Daily, June 30, 1991). Instead, participants of the first round consortia comprised the non-electronic local business groups who were seeking to diversify into microelectronic manufacturing and state-sponsored research institutions.

54 Local business groups in Taiwan had only limited participation in the microelectronic industry and information technology products until the late 1980s. The participation of local business groups brought a huge departure from the previous decades. In previous decades, firms that invested in electronic and IT were mostly smaller firms with dynamic entrepreneurs or firms under the support of the state. Throughout the 1980s, electronic manufacturing was widely viewed as too risky. Why then would large local firms seek entry into electronic manufacturing in the 1990s? Stan Shih, the founder and ex-chairman of Acer Group provides the motivation for entry into TFT-LCD:

What people don’t realize is that TFT-LCD is not a stand-alone business, but a base for growth in IT industry. This is an important distinction from other electronic producing nations. Our strategic vision was to install Taiwan made panels in Taiwan- designed consumer electronics, personal computers and other electronic products. (Shih, 1996: 23)

The rationale to develop an electronic components industry in general and the TFT-LCD industry in particular also coincides with the search for new growth opportunities by business community. As W.C. Yang, the chairman of that largest business group in Taiwan, ‘Formosa Plastic Group’, states:

Beginning in the 90s, the petrochemical industry will no longer be as profitable as it used to. Formosa Plastic Group as a whole needs to actively seek opportunities in emerging industries if it wishes to continue to grow for the next thirty years, we have located electronics components manufacturing as potential fields to enter” (United Economic Daily: Feb 24, 1990).

5.3.2 Formosa Plastic Group (FPG)

FPG was founded in 1954 by YC Wang and became the largest industrial group in Taiwan in 30 years. It began as a small producer of PVC, a raw material used in making a range of plastic products. From this humble beginning, it gradually diversified to form interlocking holding companies and manufacturing subsidiaries that focused on textile and petrochemical industries. By the early 1980s, it became the

55 largest global producer of PVC and by the mid 1980s the three largest firms in the group were Formosa Plastics Corp, Nan Ya Plastics Corp, and Formosa Chemicals & Fiber Corp, which together achieved aggregate revenues of approximate US$3 billion. As FPG became the largest global producer of PVC and a host of chemical products, it began to seek diversification and chose IT industry as the next growth opportunity.

The investments in IT industry began with one of the subsidiaries under Nan-Ya Plastics supplying necessary chemical materials to a host of local IC firms. Nan-Ya Plastics produced carbon fiber, which was the key ingredient for manufacturing of printed circuit boards. It also produced silicon for semiconductor manufacturers. Over time, the company became the largest subsidiary in the Formosa Plastics Group, and its electronics division emerged as a key element of the group’s long-term growth strategy. Nan-Ya’s first microelectronic venture entailed the production of printed circuit boards for computer manufacturers. By 1991, Nan-Ya Plastics produced 2.4 million printed circuit boards a year to supply to major PC firms in Taiwan. This runaway success of producing print circuit boards pushed the group to move into other related critical components areas, such as Flat Panel Display and semiconductor (The Economic Daily, October 14, 1992).

Between 1993 and 1995, Nan-Ya began its semiconductor production via technology transfer with Japan’s Oki Semiconductors, where it obtained 16-megabit and 64-megabit DRAM technology. The semiconductor sector took off when Nan-Ya successfully began volume productions of 16-megabit DRAM in 1996, which were also targeted to the local PC manufacturers. By 1999, Nan-Ya achieved sales of approximately US$987 million in both DRAM and PCB sectors and became one of the largest DRAM producers in Taiwan. Nan-Ya has since invested in host of DRAM and PCB-related firms, including Inotera Memories, Formosa Advanced Technologies Company, and Formosa Komatsu Silicon Corp (United Economic Daily, February 15, 2000).

Nan-Ya also licensed the Plasma Display Panel (PDP) technology from Japanese Fujitsu to begin production of 40- and 50-inch plasma panel. FPC’s venture into microelectronic industry was underscored by managing-director Winston Wang who summarized the progress of the group throughout the 1990s:

56 It has been done on a step-by-step basis over the past 10 years. There have been no sudden jerk, Only now that I have built up a core of over 200 top-class engineers do I feel more confident in making the more significant strides in this area that we are in the process of doing. (Euromoney, October, 1994: 100)

As the need for investment capital in IT sectors increased, FPG decided to spin off the electronic-components subsidiary in Nan-Ya Plastics to form the new firm called Nan-Ya Technology (NYT) in 1994. NYT formed Formosa Plasma Display Panel but was unable to transfer the dramatic success it enjoyed in DRAM and PCB to the LCD sector. In choosing a specific category of flat panels, NYT chose Plasma Display Panel (PDP) over TFT-LCD as its main product6. The PDP sector never quite received the same government support as seen in the TFT-LCD sector. Thus the firm suffered from the double disadvantages of poor infrastructure and lack of supporting industries from the beginning. The venture also suffered not just from this initial drawback, but also the internal management shuffle in both companies further stifled cooperation between two firms.

In the same way, firms that attempted to enter the flat panel display industry without foreign alliance failed badly and were forced to withdraw. Formosa Plastic Group’s plan to produce PDP demonstrated that even the most resourceful local business group could not succeed in the display industry on its own merit:

The most critical reason for the failed venture stem from lack of proprietary technology in making PDP. The production of PDP is not something that can be done on the basis of low-cost manufacturing ability alone. One had to have the proprietary technology in the PDP itself, and the proprietary knowledge on how to commercialize it (The Economic Daily, March 15, 1998).

NYT did try to resume the development of PDP through joint-venture arrangements. In 2002, NYT chose Japan’s Fujitsu as its partner for technology transfer to form Formosa PDP. The joint venture was to produce 42-inch PDP with NYT owning 77.5 per cent equity and Fujitsu owning 22.5 per cent equity (Annual Reports 2003, Nan-Ya Technology). However, NYT never became a critical producer

6 PDP is primarily used for display panels that are 40 inches and above that are display in the stadiums and outdoor places. The manufacturing of PDP is even more capital intensive than TFT-LCD, thus only very few firms can sustain the investments in this category.

57 in the TFT-LCD industry, and continued to focus on the production of PDP remaining a marginal flat panel producer in Taiwan.

Thus, Formosa PDP failed plausibly because it largely depended on its internal R&D and did not actively pursue alliance with multinationals. One of the interviewees observed:

The proportion of shareholding of these two firms and corporate culture was the root of the problem… since the Japanese partner has minority shareholding, it is reluctant to release the newest technology to Formosa PDP, thus causing Formosa PDP having to pay royalty fees on outdated technologies. (Author interview)7

5.3.3 Discussion

Table 5.1 lists business groups and affiliated firms that participated in the investments of LCD from 1990 to 1998, where these firms obtained their initial technology and their results since initiation. The result of the first phase of consortia was dismal. Only two of the original five firms continued operations. The failure was largely attributed to lack of experience. Linden (1995) argues that entering the production of flat panel display was a high-risk endeavor, since the display technology was not mature and could easily be undermined by a breakthrough within a few years; furthermore, the technology was hard to master. At least two to three years are required to bring mass production yields to acceptable levels (Linden, 1995). He concludes that Taiwanese firms did not have the quality and price to persuade their buyers – the local computer firms − to rupture their extant and stable supply relations with Japanese suppliers (Linden, 1995).

7 Interview with Chen, Yun-Chen, Column Reporter of TFT-LCD Industry. The United Economic Daily (Interview conducted in Dec, 2008).

58 Table 5.1: First round of consortia 1990–1998

Company Group Technology Investment Results Affiliation Source Amount (in US$ million) Chung-Hwa Tatung Self & ITRI, 110.00 by Picture Tube Continue operation Group Taiwan 1995 Inc. Nanya Formosa 74.00 by Technology Plastic Self Withdrew in 1999 1994 Corp. Group Picvue 42.00 by N/A Self Withdrew in 1999 Electronic Inc. 1993 Prime View Yue Fong ITRI, 110.00 by International Yue Continue operation Taiwan 1995 Inc. Group Unipac Invite Sharp Japan for United Optoelectronic ITRI, 30.00 by equity joint-venture, Mircro Corp. Taiwan 1994 later merged with Acer Corp. Display Inc. Source: United Economic Daily (compiled by author)

5.4 Second round of consortia (1999−2005)

5.4.1 Background of second consortia

The failure of the first round of consortia did not dissuade Taiwan from developing the TFT-LCD industry. The shift in global dominance in flat panel display industry from Japan to South Korea, due in part to technology maturity and rising wages in Japan, gave Taiwanese electronic firms opportunities for alliances (Hart & Linden, 1998). The state and local business groups in Taiwan took full advantage of this global shift by seeking partnerships with Japanese multinationals to jointly invest in the production of TFT-LCD.

The second round of consortia witnessed the beginning of ‘Triple-alliance’. The concrete form of ‘equity-based strategic alliances’ (Murtha, 2001: 183) became the most noticeable difference in the formation of the second round of consortia. Under such arrangements, the cooperation between Japanese and Taiwanese firms became

59 the dominant feature. The alliance usually takes the form of ‘technology transfer for royalty fees’ and/or ‘subcontracting’ –often a mix of both.

Why did Japanese firms seek out Taiwanese partners to form alliances even though they did not intend to transfer proprietary LCD technology to Taiwanese firms? The entry of Korean electronic firms into the global TFT-LCD industry perhaps explains why Japanese electronic firms became more willing to enter into alliance with Taiwanese firms. Linden (2000) argues that the protracted low-pricing strategies of Korean firms compelled many Japanese TFT-LCD producers to seek partners that could take on a more capital-intensive part of the production (Linden, 2000). The interview with an ITRI researcher confirms Linden’s observation:

Historically, Korean firms have imported technologies from Japan, and then took market share by aggressively lowering the price of their product. It has happened in the computer chip industry and they (could forecast) what happened in computer chip industry (would) also happen in LCD industry. You have to understand that Korean firms got their initial LCD technology from Japan as well, and thus TFT- LCD produced by Korea and Japan are identical. In such circumstances, price competition is unavoidable. You can even say it is fact of the life. (Author interview)8

The emergence of Korean firms in the global flat panel industry is evident with the global market share it gained since 1997. Before 2000, Korean-based electronic firms had 36.8 per cent of market share and Japan 49.5 per cent, but by 2001, Koreans firms had gained 43 per cent of global market share while Japan had declined to 33.5 per cent (ITIS, 2002). These adverse events led Japanese firms to contract with Taiwanese firms and transfer technology for the production of LCD devices. The Japanese strategic intent behind joining in a strategic alliance with Taiwanese firms was to counter the increasing threat from Korean firms. The creation of ‘equity-based strategic alliances’ allowed firms from Taiwan and Japan to gain what they needed in order to stay competitive in the global display industry. As one Toshiba executive pointed out:

8 Interview with Mr. Liu, Chung-Yuan. Assistant Researcher in Electronic Industry Industrial Technology Research Institute (Interview conducted in Dec, 2008).

60 We want to concentrate limited investment resources on certain types of products, yet at the same time, we need a stable supply of LCD. Thus by providing standard technology for production to Taiwanese firms. We can concentrate on high margin products, and if there is a big demand for standard LCD down the road, Toshiba may consider buying output from its Taiwanese counterparts. (United Economic Daily: July 2, 1999)

Taiwan’s Quanta Computer, the largest computer-manufacturing firm, became the first to form equity-based alliances. In 1999, Quanta Display Inc. was formed with joint investment of Quanta and Japanese Sharp Inc. The rationale behind the alliance was explained Quanta Chairman Barry Lam:

Taiwan is the world's largest manufacturer of notebook computers, all of which use flat panel screens. The demand for notebook computer worldwide can easily increase to about 23 million units by 2000. Sharp (from Japan) will contribute both technical expertise and capital, and this joint venture will enhance both company's growth and earnings potential, and allow both Quanta and Sharp to (have an exclusive) supply (chain of) TFT-LCD for its products in notebook computers, LCD monitors, and personal computers. (United Economic Daily: September 13, 1999)

The wide applications of the TFT-LCD in many of the product categories have attracted investments from Taiwanese electronic firms. Previously, Taiwanese firms were only interested in producing the 10- to 15-inch TFT-LCD panels to be used largely on the handheld electronic devices and Notebook PC it manufactured. Beginning in the late 1990s, Taiwanese firms began to produce large-sized panels of 20 to 40 inches, even as production costs began to fall. The production of large-sized panels opened a whole new product market, such as LCD TV and LCD monitors that required further investment. The trend toward wide application of TFT-LCD was confirmed in the interview with the manager at AUO, who pointed out that:

The application for TFT-LCD is increasingly spreading; all kinds of electronic products, such as TV, Mobile Phones, PC, will all require TFT-LCD technology. I

61 believe building TFT-LCD is the trend for the future of Taiwan’s IT industry. (Author interview)9

Seeing the surging demand for flat panel displays, local business groups began to fund investments in TFT-LCD. This is evident in the group affiliations of each participating company. Every participating firm during the second round of consortia had the backing of one of the 30 largest business groups in Taiwan. The major role these local business groups performed was funding the construction and expansion of production facility. By 2001, it was estimated that it cost approximately US$1−2 billion to set up one production line, and a firm would need to have at least three to four production lines in order to achieve necessary scale (TOSEA, 2002). While the business group could bear the financial demands, it did not possess any production technology or production know-how. This lack of proprietary knowledge and technologies compelled Taiwanese firms to seek concrete relations with multinational electronic firms from Japan. The most common form of institutional linkage between multinational and domestic firm was the adoption of an ‘equity-based alliance’ that saw local business groups and multinational partners both share equity stake in the newly formed LCD firm.

Alliances with multinational electronic firms proved to be a very successful strategy. Most local firms were able to begin production after their third year of establishment, compared to the industry average of 4.5 years (Optolink Report, 2004). Once a local firm began production, it started to develop its own ‘organizational capability’ and a series of mergers followed in the domestic market. Towards the end of 2005, the four largest TFT-LCD firms accounted for 70 per cent of the production capacity globally (OSL, 2006). The most notable corporate merger involved three separate firms: Acer Display, Quanwhe and Unipac Opto, which together formed AUO Display and became the largest LCD producer in Taiwan.

5.4.2 Acer group

9 Interview with Mr. Chang, Pao-Long, Assistant Researcher. The Photonics Industry & Technology Development Association (Interview conducted in Apr, 2009).

62 Taiwanese entrepreneur Stan Shih founded the Acer group in 1976. The firm initially focused on designing and manufacturing microprocessors. By making IBM- clone personal computers during the 1980s, Acer soon enjoyed tremendous success in manufacturing and distributing its own personal computers.

From the 1990s, Acer decided to make a series of strategic investments in critical components that were indispensable for its branded computers (Shih, 1996). Acer Display Technology (ADT) was formed with the support of the Acer group. ADT initially partnered with Philips Electronics from The Netherlands to produce Plasma Display, but soon switched to supply personal computers with the production of 17- and 19-inch TFT-LCDs. A steady supply of components was critical for the firm to continue its success as computer manufacturers. Acer Display Inc was to become an integral part of this supply chain. According to former Acer Group Chairman Stan Shih’s account (1996) on Acer’s motivation to invest in the production of TFT-LCD:

We invested in both semiconductor and TFT-LCD industry in early 90s. Both products are the critical components industries Acer chose to enter after it has successfully established itself as the computer manufacturer in the late 80s. Acer began its investment in semiconductors in 1989; however, due to the cyclical nature of the industry that depressed the price of products and lack of its own proprietary technology, Acer sold off its investments in semiconductor and merged it with Taiwan Semiconductor Manufacture Company in 1999, but we did continue to invest in TFT- LCD industry. ADT was initially placed under the group’s communication and multimedia division. The initial strategy of ADT was to supply LCD monitors to the group. Sensing the opportunities presented in display industry, the company decided more investments should be devoted for the TFT-LCD (Shih, 1996: 24−26).

In 1996, ADT completed a technology transfer agreement with IBM Japan to produce TFT-LCD, which later allowed ADT to produce large-size panels in Taiwan. Besides forming strategic alliance with multinationals, ADT also made the decision to build its production facilities in the new industrial science park located in Taichung.

The 2001 merger of Acer’s ADT with United Microelectronics Inc’s Unipac Display Inc. created AU Opotronics (AUO) Inc. The firm became the largest TFT-

63 LCD in Taiwan and the third largest global producer, behind Korea’s Samsung Electronic and LG Electronics. The creation of AUO allowed Acer to raise the necessary financial capital to continue investment in its new production capacity, which saw its global market share steadily rise to become the second largest TFT- LCD producer in 2005 (OSL, 2006).

Beginning in early 2000, the TFT-LCD industry began to undergo a series of consolidations, which saw the creation of large-scale local manufacturers. The consolidation was motivated by several factors, including the announcement of the TFT-LCD industry specific policy (IEKC, 2001) and access to external capital that allowed smaller firms to finance its expansions. The creation of AUO demonstrates that by this stage merger and acquisition had become a preferred strategy for the industry. While the state and the private sector cooperated to gain global market share, the role and participation of multinationals did not diminish. The equity-based alliance was carried over from the second round of consortia to the third and current round, which witnessed many local firms partnering with MNCs as suppliers and OEMs (Murtha, 2001).

Local TFT-LCD firms have always faced the intrinsic obstacle of small scale and shortage of proprietary technologies for it to compete effectively with international TFT-LCD producers, especially its Korean counterparts. The consolidation of small LCD outfits offered several advantages, including diversified product lines, shared R&D personnel, expanded production capacity and ability to gain favorable material and component prices. Most importantly, the merger allowed local LCD producers to invest in the newest production line and increase its overall production capacity vis-à-vis rival Korean firms.

5.4.3 Chi Mei group

The Chi Mei group was founded by W.L. Hsiung in the late 1960s, and had become one of the largest producers of chemical material in the global petrochemical market by the late 1980s. The motivation for the group to enter into TFT-LCD was a gradual process (Wang, 2002). Initially, the group did not plan a full-scale TFT-LCD production, instead seeking to be the major supplier of components in the industry. In

64 1997, the group chose to invest in the production of color filter glass essential for the production of TFT-LCD.

The initial horizontal product diversification in color filter soon turned into the full production of TFT-LCD in 1998. Difficulties to find consistent demand for its production of color filters was the major reason why Chi Mei decided to move into the full production of TFT-LCD, since at the time only three firms produced TFT- LCD in sufficient quantities that warranted external suppliers for its components. These three firms had survived first round of consortia in 1990−1995: Chung-Hwa Picture Tube, Prime View International and Unipac Optoelectronic Inc (Wang, 2002).

In 1998, the group formed a joint-venture agreement with Japan Fujitsu to produce 14-inch TFT-LCDs. This partnership established mutually beneficial relationships between Chi Mei and its Japanese counterparts: it waived the need for Chi Mei to pay for technology licensing fees and various royalties fees, while Fujitsu secured a stable supplier of TFT-LCD for its consumer electronic divisions. While Chi Mei’s supplier-buyer relation with its Japanese counterparts allowed the firm to secure steady buyers for its TFT-LCD, Chi Mei suffered huge losses in the initial stage of the development. It is estimated that the group lost on average US$40 million a month when it first began to produce the 14-inch LCD on its own. The loss came mainly from the initial outlay on fixed investments, including machineries, materials and site construction. As the global TFT-LCD industry swung back to high demand in 2000, Chi Mei began to expand its production capacity via acquisitions. In 2001, Chi Mei acquired IBM Japan’s TFT-LCD production facilities. This acquisition not only helped Chi Mei to gain additional production facility, it helped the firm to expand its client lists. The account of Chi Mei Group’s founder W.L Hsiung demonstrated the versatility of large business groups:

We did not begin with producing LCD, initially we started off with the production of color filters. We decided to give it a try in the production of color filters, since we have been looking for the chance to diversify our investments by applying our knowledge in chemical industries to other industries. (Tian-xie: July 1999)

65 Chi Mei’s full scale entry into the production of TFT-LCD industry coincided with the establishment of a new science-based industrial park in Tainan, where the firm was based. Once Chi Mei acquired the necessary technical know-how from multinational firms, it cooperated with the state to allow the creation of TFT-LCD clusters in the industrial park. In 2005, Chi Mei became the second largest TFT-LCD producer in Taiwan, behind AUO (Wang, 2002). According to the interview with Mr. C.W. Hsu of Chi Mei, the reasons behind Chi Mei’s successful entry into TFT-LCD industry was due to its partnership with Fujitsu:

One of the benefits for Chi Mei to partner with Japanese was that is allowed us to acquire its TFT-LCD facilities. The direct acquisition of the production site allowed us to leverage Fujitsus’s existing resources and gain access to more clients from American multinational. Before the merger, most of Chi Mei’s clients are from Japan and Taiwan. Since the merger, Chi Mei’s client list has expanded and now includes Apple, Dell, HP and IBM. (Author interview)10

5.4.4 Contribution of multinational electronic corporations

Since the display technology has been under development for past decades, the most logical solution was to seek assistance with multinational firms that possessed such know-how. In searching for proprietary TFT-LCD technology abroad, Japan became Taiwan’s important ally and gave way to the formation of the ‘Triple- alliance’. An interview with the Deputy Director of Industrial Development Bureau demonstrated the shifting landscape of TFT-LCD industry that led to the alliance:

There is never an industry that is comparable to TFT-LCD industry. There are only three countries that are involved in this industry, namely Japan, South Korea and Taiwan. Japan is slowing exiting the market, thus leaving only Korea to compete directly with Taiwanese firms. (Author interview)11

10 Interview with Mr. Hsu, Chiung-Wen, Assistant Public Relations Manager. Chi Mei Display Inc. (Interview conducted in Mar 2009). 11 Interview with Mr. Lin, Yan-Kang, Display Industry Deputy Director. Industrial Development Bureau, Ministry of Economic Affairs (Interview conducted in Dec 2008).

66 In the case of the TFT-LCD industry, the technology came mostly from Japan. Major Japanese LCD producing firms, such as Sharp, Fujitsu, Toshiba and Matsushita, have all sought alliance with local firms. The alliance had been beneficial to both parties, especially to local LCD producers in Taiwan. Table 5.2 shows the alliances between Taiwanese and Japanese firms throughout 1998 to 2005. There are six newly established TFT-LCD firms in Taiwan and, without exception, they all have Japanese TFT-LCD manufacturers as an equity holder. The major difference between the first and the second round of investment is the degree of involvement by the multinational electronic firms.

Table 5.2: Second round of consortia 1999−2005

Affiliated Japanese Results Business Partners Groups

Acer Display Acer Group IBM Japan, Beginning production in 1999, Tech Japan later merge with Unipac and Quanhwe

Chunwha Tatung Group Mitsubishi, Beginning production in 1999 Picture Tube Japan

Unipac United Matsushita, Beginning production in 1999, Optoelec Microelectronic Japan later merge with Acer Display and Group Quanhwe

Chi Mei Chi Mei Fujitsu, Japan Beginning production in 1999 Optoelec Industrial Group

HannStar Walsin Linwha Toshiba, Beginning production in 2002 Display Group Japan

Quanhwe Quanta Sharp, Japan Beginning production in 2001, Display Computer later merge with Acer Display and Group Unipac.

Innoloux Foxconn Group Sharp Beginning production in 2003 Display Electronics, Japan

TPO Display Compal Philips Beginning production in 2004 Electronic Electronic, Group Dutch Source: United Economic Daily (compiled by author).

67

One of the multinational firms that made significant investments during this period was New York-based Corning Inc. Corning Display Technologies Company has a global market share of over 50 per cent and makes the AM-LCD glass substrate used in the production of TFT-LCD (Company Annual Report, 2008). The Corning Inc. built its first glass substrate production factory at TSIP in 1999, which was subsequently expanded into a fully-integrated glass substrate manufacturing plant in 2002. According to the company report of Corning:

Currently [in year 2002], the company's total annual revenue from sales of LCD glass substrate in Taiwan and Japan is approximately US$400 million. It is estimated that revenue will continue to grow at a 20 to 40 percent rate until 2006, with Taiwan being the primary growth area. With southern Taiwan having become an important center for the development of the optical industry, Corning is planning to invest an additional US$100 million in the island, and will continue this investment in the future. The company already has another plan to build a glass smelting plant in the Tainan Science-based Industrial Park. (Company Annual Report, 2003)

It was perhaps this critical investment made by Corning in Taiwan that provided Taiwan’s LCD makers the wherewithal to rapidly leapfrog into an industry with seemingly insuperable entry barriers.

5.5 Conclusion

The chapter demonstrates with case studies how Taiwan was able to create a TFT-LCD industry in a relatively short period of time. In essence, the alliances among the three institutions enabled latecomer firms to compensate the two major disadvantages they faced when attempting to develop technology-intensive industries: late entry to the competitive industry and lack of advanced technologies.

The infirmity caused by late entry into a highly competitive industry was perhaps greatly mitigated when the public and private sectors decided to cooperate and coordinate their resources to catch up with first movers. As argued in Chapter

68 Two, the state has provided the necessary infrastructure for the private sector; moreover, it has set specific policies to target the development of TFT-LCD industry and direct resources into the industry. The lack of advanced technologies was mitigated when the domestic business groups formed alliances with multinational electronic corporations – with alliances often taking the form of ‘equity-based’ partnership, which allowed both parties to profit from the venture. In summary, the results across all the case studies suggest a positive relationship that exists between the performance of the TFT-LCD industry in Taiwan and the existence of ‘Triple- alliance’.

The suggestion that there exists a positive correlation between the performance of TFT-LCD firms and the existence of ‘Triple-alliance’ from the three case studies demonstrated in this chapter will be examined with the performance of 100 electronic firms in Taiwan in the next chapter.

69 Chapter Six: Statistical analysis

6.1 Introduction

The object of this chapter is to examine the impact of institutional factors on market performance of latecomer firms in the TFT-LCD display industry in Taiwan through econometric modeling. Specifically it examines whether the ‘Triple-alliance’ construct has any significant impact on the TFT-LCD industry. Results from the testing of the propositions of ‘Triple-alliance’ model present the nature of relationships between the independent variables consisting of the participation of state, multinational corporations and local business groups with the dependent variable – market share. Section 6.2 reiterates the research question and proposes three hypotheses. Section 6.3 explains the source of data, and how they were collected for the study. In addition, it provides a detailed definition of both dependent and independent variables and the rationale for choosing them to test the hypothesis. Section 6.4 shows the result of regression and Section 6.5 discusses the finding and results. Section 6.6 summarizes the finding from the regression analysis.

6.2 Proposed hypotheses

The basic premise of the thesis is that cooperation between three specific institutions – the state, multinational corporations and local business groups − can potentially assist latecomer firms to enter technology-intensive industries in electronics. The specific empirical question for the regression analysis can thus be stated as follows: “Does the conjoint involvement of three institutions − state, business groups and multinational electronic firms − have a significant impact on market performance of electronic firms in Taiwan?”

70 6.2.1 The role of the multinational corporations in ‘Triple-alliance’

The first hypothesis examines the role of multinational corporations (MNCs) in assisting latecomer firms in successfully entering a technology-intensive industry. In the case of the TFT-LCD industry, MNCs both function as a source of technology and act as customers for LCD firms. Previous studies (Matthews, 1997, 2002) have shown that latecomer firms have greatly benefited by playing a complementary and supportive role for advanced multinational firms, and subsequently learning how to become more productive. Taiwanese electronic firms took on the role of ‘original equipment manufacturers’ to complement Western multinational electronic firms which focused on branding and marketing. Firms from the case study show that the leading LCD manufacturers in Taiwan, such as AUO and CMO, are less interested in becoming brand names themselves and instead prefer to focus on becoming the most efficient and reliable supplier to other global LCD brand firms.

The most obvious relationship is revealed when there is a payment of technology fees and/or royalty to advanced MNCs by latecomer firms. In the case of LCD industry, the structural alliance is often an equity-based alliance (Murtha, 2001) that formally links MNCs with domestic electronic firms in Taiwan and allows both parties to profit from each other’s comparative advantages. The formal linkage is represented by the independent variable (ForeignAlliance). One major incentive for an MNC to form alliance is the flow of royalty payments it receives for its advanced technologies from Taiwanese firms, which perpetuates its technological dominance. Hence the role of MNCs is a multi-measure construct, captured by an independent (nominal) variable ForeignRoyaltyPayment and a dummy variable Foreign-Alliance.

Hypothesis 1: Structural alliance with MNCs is likely to have a significantly positive impact on latecomer firms’ market performance in the TFT-LCD industry.

6.2.2 The role of the state in ‘Triple-alliance’

The creation of microelectronic industry shows that state involvement was indispensable. Taiwan’s microelectronic industry evolved from its initial role as low- cost assembling location, to original manufacturing and then to become full-scale

71 designer and producers of advanced information technology products (Matthews, 1997). The state played a critical role in leading these transformations. For example, the state-sponsored organizations, such as ITRI and ERSO, were designated with the task of leading the transformation of microelectronic industries from subcontracting assembly to advanced components producers (Matthews, 1997; Wang, 1995).

In the context of Taiwan, the state-designed industry-specific policy that tailored the development of the TFT-LCD was widely known as ‘Two Trillions, Twin Stars’ policy (Hung, 2006; Wang et al., 2007). The state specifically responded to the problem of infant entrepreneurs in TFT-LCD technology in Taiwan with generous financial incentives and R&D assistance, later establishing industrial parks to encourage the formation of industry clusters specifically for TFT-LCD industry. It is widely believed that without the initial involvement of the state, the private sector would never have invested in TFT-LCD on such grand scale, thus the first hypothesis is the role of the state support in the ‘Triple-alliance’.

In the case of LCD industry, the structural alliance often came with various forms of financial and research support from the state, and formally linked the state to domestic electronic firms in Taiwan. Moreover, the state also provided functional support through the establishment of science-based industrial parks across Taiwan. The government’s direct financial and research support is represented by the independent (nominal) variable (StateSupport), whereas the functional support is represented by the independent (dummy) variable (IndustrialParkExistence). In short, the state demonstrated its support to the private sector by diffusing results of public- sector research and by establishing industrial parks for TFT-LCD firms.

Hypothesis 2: External assistance from government is likely to have a significant positive impact on latecomer firms’ market performance in the TFT-LCD industry.

6.2.3 The role of the local business groups in ‘Triple-alliance’

The third hypothesis predicts that local business groups were instrumental in absorbing foreign technology and applying its group-wide financial and human

72 resources to manufacture technology-intensive products at massive economies of scale. Previous studies have shown that these distinct forms of organization within East Asia resulted from a unique pattern of industrialization that happened in a specific institutional context (Amsden & Chu, 2003; Chung, 2001, 2005; Kim, 1997; Whitly, 1991, 2000). It assumes that economic relationship and activities are socially constructed, and thus a systematic analysis is needed to understand different forms of business system.

In the context of Taiwan, the ‘local business group’ has provided necessary and abundant organizational support in allowing its subsidiaries in successfully entering into the display industry in the absence of proprietary technology. The ‘Triple- alliance’ framework argues that the nature and size of firms can have a major impact on the sophistication of products the firm produces. Large local business groups in Taiwan were able to provide necessary finances and human resources to enter the TFT-LCD industry and, more importantly, unlike small-medium sized companies, large business groups possessed project execution skills to ensure complex productions of TFT-LCD are carried out with minimum defects. In addition, large business groups were also able to cross-subsidize the so-called LCD ‘loss leaders’ in the short run with their more profitable profit centers, in the hope that gestation periods would eventually be completed in the former product-markets and eventually make profits.

In the case of TFT-LCD industry, the increasing productivity of TFT-LCD panel coincides with the institutional linkage of local business groups in Taiwan, represented in this case by the independent variable (GroupAffiliation). Thus the third hypothesis is summarized as follows:

Hypothesis 3: Business group affiliation is likely to have a significantly positive impact on latecomer firms’ market performance in the TFT-LCD industry.

73 6.3 Research methodology

6.3.1 Characteristics of selected firms

The primary criterion for sample selection for this type of research is that the firms should be publicly listed companies in Taiwan. This excludes both privately owned microelectronic firms and microelectronic firms that operate in Taiwan but are listed elsewhere. The data for testing were derived from public firms listed on the Taiwan Stock Exchange (TSEC) in 2006, and were collected from company annual reports listed on TSEC in 2006. Both dependant and independent variables came from these annual reports of firms listed with TSEC. Firms had to meet the criteria of having been listed with TSEC for a minimum of one year and not be undergoing any corporate-level restructuring in 2006. This ensured any financial impairments or one- time gain were not included in the dataset to distort the analysis.

The sample study compares firms that are involved with LCD industry with firms that are not involved in LCD industry, in order to ensure LCD industry is represented and examined adequately. The rationale for this comparison is to ensure LCD firms are represented in the manner where it can be compared to non-LCD firm in a statistically significant manner, and reduce the potential for anomalies. The distribution of these two groups is given in Table 6.1.

Table 6.1: Distribution of companies by LCD activities

Firms Frequency Percentage

LCD Firms 46 45%

Non-LCD Firms 56 55%

Total 102 100%

The total sample (n = 102) represents flat panel display and other related firms as a proxy for a comparative study of Taiwanese latecomer firms in the LCS and

74 microelectronic industries. The dataset targeted publicly listed firms from the following industries: components (n = 13), computers and peripheral equipment (n = 16), consumer electronics (n = 12), flat panel display and display components (n = 46) and semiconductors (n = 15). Table 6.2 provides the summary of companies with respect to its respective industries.

Table 6.2: Distribution of companies by respective industries*

Firms Frequency Percentage

Components 13 13%

Computer and peripheral equipments 16 16%

Consumer electronic 12 11%

TFT-LCD and display components 46 45%

Semiconductor 15 15%

Total 102 100%

*Taiwan Stock Exchange

6.3.2 Construction of variables

The dependent variable here is market share (sales) and independent variables are the state, MNCs, business groups and firm-specific constructs. Firm specific variables are used mainly as controls: these are age, marketing expenditures and R&D expenditures.

Dependent variable

The performance of technology-intensive firms is best measured by individual firm-level performance (Lindt, 1995). Since the LCD industry has the characteristics of high technological turnover with screen size doubling every five years, the primary

75 determining factor of success lies in the differences in scale. The representative dependent variable, firm sales revenue, is preferred over operational profits. Sales revenue is a proxy for, and captures the effect of, a number of variables such as relative scale, size and market share vis-à-vis the industry – a measure widely used in the industrial organization literature (Caves & Ghemawat, 1992). The choice of ‘sales’ as the dependent variable is in accordance with the methodology adopted by CCIS in selecting 100 of the largest business groups in Taiwan based on sales, instead of profit (CCIS, 2000). The rationale for such choice is to avoid miscalculation of firm performance due to transfer pricing that might exist between the listed and non- listed subsidies as the preferred mechanism to reduce its end of year tax liabilities, thus leading to consistently underreporting its true operation profits (Numazaki, 1988, 1991). Sales revenue is therefore a robust measure of the impact of ‘Triple-alliance’ model vis-à-vis firms that do not have such alliance. The description of the dependent variable ‘sales’ is provided in Table 6.3.

Table 6.3: Definition of dependent variables

Variable name Label Variable definition

Sales Sales Firm-level sale is used for the dependent variable as it reflects the performance of the firm. Figures for Sales Revenue are derived from individual annual reports in 2006.

Independent variables

A total eight independent variables were employed to analyze factors that can explain the performance at the individual firm level. In accordance with the ‘Triple- alliance’ perspective discussed in the theoretical chapter, variables selected here reflect the alliance of three major institutions. These variables include foreign royalty payment, foreign alliance, existence in industrial park, government (state) support, group affiliations, and individual firms’ marketing and product research development expenditures as well as age. Table 6.4 provides the overview of all independent variables that were tested and gives the definitions of each variable in greater detail.

76 Table 6.4: Definition of independent variables

Variable name Label Variable definition

Age Age The length of period that the firm has been established. It is calculated by deducting the year of establishment of the firm from year 2006.

Existence in IndustrialParkExistence A ‘dummy variable’ - whether the firm is industrial park established in any of industrial parks in Taiwan or not – thus signifying whether it is receiving State support.

Foreign royalty ForeignRoyaltyPayment Log of technology licensing and royalty payment payments made to multinational firms in year 2006.

Foreign alliance ForeignAlliance A dummy variable of whether the firm is linked to multinational firms or not.

Government StateSupport A dummy variable whether the firm gains (state) support support of the state or not.

Group GroupAffilation A dummy variable of whether the firm is linked affiliation to any of multinational corporations either through joint ventures or strategic alliances.

Marketing MktingExpense Log of marketing expenditures of individual expenditures firm in year 2006.

Research & R&DExpense Log of research & development expenditures Development each firm spend in year 2006. expenditures

The dummy variables are designated either 1 or 0. As per the descriptive statistics Table 6.5, 42 out of 102 exist in one of the science-based industrial parks, 44 out of 102 firms form an alliance with multinational electronic firms, 68 out of 102 firms receive state support in either reduction of tax liabilities or direct financial

77 support, and 42 out of 102 firms are affiliated with one of the largest 100 business groups. Table 6.5 shows the frequency of these characteristics for 2006.

Table 6.5: Qualitative variables and descriptive statistics (n = 102)

Variable Frequency Percentage

Existence in industrial park 42 41%

Foreign alliance 44 43%

Government (state) support 68 57%

Group affiliation 42 41%

Table 6.6 shows both the mean and standard deviation of the five quantitative variables for 2006: age, foreign royalty payment, marketing expenditures, R&D expenditures and firm-level sales.

Table 6.6: Quantitative variables and descriptive statistics (n = 102)

Variable Mean Standard deviation

Age 15.7 11.704

Foreign royalty payment 7.5448 1.5184

Marketing expenditures 5.1917 0.93839

R&D expenditures 5.4694 0.80063

Sales 6.9994 0.82017

78 6.4 Analytical approach

The model whether ‘Triple-alliance’ model explains the success of firms in the microelectronic industry is tested here with hierarchical regression analysis. Hierarchical regression allows individual examination of the hypotheses (Pallant, 2007). This research strategy allows for the exploration of relationships among a set of intertwined variables (Raudenbush & Bryk, 2002) and provides logical conceptualizations to ensure the order of variables are in line with proposed hypothesis (Gelman & Hill, 2006). Individual firm-level sales in 2006 is the dependent variable. Independent variables are entered in a hierarchical and sequential manner. SPSS version 17 was employed to test the hierarchical regression analysis. The following set of four equations explains each stage of hierarchical regression analysis and henceforth is referred to as the four models of our econometric analysis.

Sales = ß0 + ß1 Foreign Royalty Payment + ß2 Foreign Alliance ...... (1)

Sales = ß0 + ß1 Foreign Royalty Payment + ß2 Foreign Alliance + ß3 State Support +

ß4 Industrial Park Existence ...... (2)

ß0 + ß1 Foreign Royalty Payment + ß2 Foreign Alliance + ß3 State Support + ß4

Industrial Park Existence + ß5 Group Affiliation ...... (3)

Sales = ß0 + ß1 Foreign Royalty Payment + ß2 Foreign Alliance + ß3 State Support +

ß4 Industrial Park Existence + ß5 Group Affiliation + ß6 Age + ß7 R&D Expenditure +

ß8 Marketing Expenditure ...... (4)

Both Correlation and Collinarity tests are conducted to ensure the degree of relationship among variables is within acceptable limits. Table 6.7 presents the Pearson Product-Moment correlation coefficients of all the variables with both Tolerance-level and Variable Inflation Factor (VIF) for all independent variables. Statistical problems can occur at a high bivariate correlation of 0.90 and above (Gujarati, 1995). In another words, any independent variables that have VIF greater than 10 and Tolerance-level less than .10 need to be excluded from the regression model (Tabachnick & Fidell, 2001). Table 6.8 presents the Collinearity statistics table, which shows that independent variables used for analysis are within the boundary of both Tolerance-level and VIF coefficients.

79 Table 6.7: Pearson correlation coefficients

1 23456 7 89 1 Sales 1 2 ForeginRoyaltyPayment 0.569* 1 3 ForeignAlliance 0.564* 0.4* 1 4 StateSupport 0.3* -0.044 -0.066 1 5 IndustrialParkExistence 0.634* 0.226* 0.415* 0.272* 1 6 GroupAffiliation 0.623* 0.381* 0.436* 0.162* 0.442* 1 7 Age 0.453* 0.361* 0.317* 0.036*** 0.358* 0.345* 1 8 R &D Expenditure 0.827* 0.492* 0.416* 0.399* 0.502* 0.619* 0.308* 1 9 MarketingExpenditure 0.875* 0.455* 0.507* 0.266* 0.62* 0.6* 0.45* 0.772* 1 ***significant at p<=0.01 level; **significant at p<=0.05 level; *significant at p<=0.10 level

1 ForeignRoyaltyPayment 5 GroupAffiliation 2 ForeignAlliance 6 Age 3 StateSupport 7 R & D Expenditure 4 IndustrialParkExistence 8 Marketing Expenditure

Table 6.8: Collinearity statistic table

1 2 3 45 67 8 Model 1 Tolerance 0.84 0.84 VIF 1.19 1.19

Model 2 Tolerance 0.834 0.707 0.885 0.733 VIF 1.199 1.415 1.13 1.365

Model 3 Tolerance 0.783 0.665 0.869 0.684 0.673 VIF 1.278 1.505 1.15 1.462 1.487

Model 4 Tolerance 0.63 0.632 0.705 0.562 0.547 0.731 0.29 0.289 VIF 1.588 1.582 1.418 1.779 1.827 1.369 3.451 3.466

1 ForeignRoyaltyPayment 5 GroupAffiliation 2 ForeignAlliance 6 Age 3 StateSupport 7 R & D Expenditure 4 IndustrialParkExistence 8 Marketing Expenditure

80 6.5 Results

Table 6.9 shows the regression results for each model. Model 1 tests the first hypothesis on the effect of the involvement of multinational firms. Model 2 tests the second hypothesis on the effect of the involvement of the state. Model 3 tests the third hypothesis on the effect of the involvement of the local business groups. The last model tests the effect of firm-specific expenditures, though it is not included in the hypotheses. A more thorough discussion of each model is provided in the following section.

Table 6.9: Regression results table

Model 1 Model 2 Model 3 Model 4 Variables ForeignAlliance 0.087 0.087 0.063 0.084 ForeignRoyaltyPayment 0.744*** 0.602*** 0.537*** 0.184* IndustrialParkExistence 0.248*** 0.215*** 0.118* StateSupport 0.177** 0.162** 0.034 GroupAffiliation 0.179** 0.025 Age 0.046 Marketing Expenditure 0.287*** R & D Expenditure 0.347*** 2 Adjusted R 0.637 0.73 0.749 0.836 Δ R2 0.644 0.097 0.021 0.088 SE 1.137 0.981 0.946 0.764 ***Significant at p<. 001 level; **significant at p<.01 level; *significant at p<.05 level; †significant at p<.10 level

Model 1 tests the effect of the involvement of ‘multinational firms’ with two independent measures, namely Foreign alliance and Foreign royalty payment, to test the first hypothesis which proposes that firms that form structural alliances with multinational corporations are likely to have success in TFT-LCD industry. The result of the regression test shows Model 1 has an Adjusted R Square of .637, which supports our first hypothesis. Foreign alliance (β =0.087) demonstrates no significant relationship with the dependent variable ‘sales’. The relationship between dependent

81 variable ‘sales’ and independent variables Foreign royalty payment (β=0.744) was positive and significant at the level of p < .001. In other words, the regressions show significance of variables that represent the involvement of multinational corporations. In summary, the test shows that the amount of ‘Foreign royalty payments’ has a positive and significant impact on whether the firm is likely to succeed in technology- intensive sectors. Previous studies (Bair, 2005; Gereffi, 1999) measured the degree of linkage of local firms in developing countries with multinational firms from developed economics as the evidence of sophistication in organizational capabilities local firms have achieved in the former group of countries (Chen, 2002; Wang, 2002). These results validate, and using multi-measure constructs extend for a much a larger dataset the earlier studies of Chen (2002) and Wang (2002) on the impact of foreign knowhow, and alliances with multinational corporations on firm performance.

Model 2 tests the effect of the involvement of the state with two variables (Industrial park existence and State support). The results of the regressions show an adjusted R Square of .73, which supports the validity of the second hypothesis. State support (β=0.177) and Industrial park existence (β=0.248) is positive and significant at the level of p < .01 and p < .001, respectively. The test shows that with state support and being located in industrial parks, local firms are likely to be more successful in technology-intensive sectors. Conversely, it is plausible that latecomer firms, suffering from technology backwardness, are less likely to succeed in TFT- LCD without external assistance from government. In summary, the test shows that the involvement of the state bureaucracy is positively correlated to the success of late- development firms. Previous studies (Spencer et al., 2003; Tung, 2001) also showed that support from the state institutions accelerate the creation of new industries that are often too technology-intensive for the private sector to develop alone. One of the major mechanisms to facilitate the assistance of the state to the private sector is the creation of an industrial park that allows firms to cluster in one geographic area for the purpose of production and innovations (Akinwande et al., 2003; Weiss, 1994; Wong & Matthews, 1998).

Model 3 tests the effect of the involvement of the local business groups with the variable (group affiliation) proposing the hypothesis that firms that come from local business group have greater likelihood of success in TFT-LCD industry. The result of

82 Model 3 shows an Adjusted R Square of .749. The relationship between dependent variable ‘sales’ and the independent variable at this model was positive with β = 0.179 for Group affiliation. The value is significant at p < .01. Previous studies suggested the business group affiliation is vital for firms to succeed in technology- intensive industry (Chen & Sewell, 1996; Chung, 2005; Matthews, 1995, 2005). The result validates Chung’s (2005) research on the effectiveness of group affiliation in the success establishment of technology-intensive industry.

Model 4 tests the effect of firm-specific variables, including length of establishment (Age), amount of expenditures for marketing and R & D. These three variables identify the degree of contribution at firm level with the consideration of the ‘Triple-alliance’. The result of Model 4 shows an Adjusted R Square of .836. The relationship between dependent variable ‘sales’ and independent variables at this model was positive with β = 0.046 for Age, β = 0.347 for R&D expenditures, and β = 0.287 for Marketing expenditure. Both R&D expenditures and Marketing expenditure are significant at the p < .001 level. The regression test shows firm level factors including both R&D and marketing efforts are indispensable in the performance of the firm.

6.6 Summary

Previous studies have demonstrated that the emergence of TFT-LCD producers in developing economics is the result of cooperation across different institutions (Akinwande et al., 2003; Amsden & Chu, 2003; Hart & Linden, 1998; Matthews, 2005; Tsai, 2006). However, none of the previous studies highlighted the critical role business groups played in bolstering Taiwan’s LCD industry. What is more no other study tested the ‘Triple-alliance’ model by way of econometric modeling. This study not only confirms the validity of previous studies by theorizing the public and private relationship in the form of ‘Triple-alliance’ but also confirms the hypothesis that the simultaneous involvement of three institutions is essential to the successful creation of the technology-intensive TFT-LCD industry in Taiwan. The empirical results show that, for microelectronic firms based in Taiwan, institutional linkage to the state and multinational firms does lead to higher sales. In summary, the findings of this chapter

83 support the analysis of previous chapter in that the institutional alliance of the state, local business group and multinational corporations are essential to facilitate the creation and expansion of TFT-LCD industry.

84 Chapter Seven: Conclusion

7.1 Introduction

This thesis examined the key institutional determinants of success of latecomer firms in newly industrializing economies, with special reference to flat panel display products. Its findings suggest that for latecomer firms to develop competitive advantage in the TFT-LCD industry, it is critical to function under a package of vertical strategies, encompassing firm-specific systems, the state and multinational corporations. It is further demonstrated that the cooperation must take the form of an institutional alliance, where latecomer firms partner with the state and multinational corporations to jointly invest in the new industry. Finally, it is shown that firms attempting to enter the flat panel display industry must also have the backing of business groups in order to develop competitive advantage.

This concluding chapter summarizes the major findings of this thesis, and discusses the contribution and implication of this study for future research and policy formulation. The chapter concludes by addressing the limitations of this study and suggestion the direction of future research.

7.2 Summary of major findings

In summary, the key findings of this research are that technologically backward latecomer firms are less likely to succeed in TFT-LCD without external assistance from the government, technology transfers from multinational electronic corporations and the backing of big business groups. As demonstrated both in the qualitative case studies and econometric analyses, successful local electronic firms in the production of TFT-LCD are those that combine firm-specific technological effort with technological assistance from the state and multinational electronic firms. These three institutions together form the so-called triple alliance that facilitates the transfer of advanced technologies from multinational electronic firms to local business groups. In so doing, this study contributes to the international business literature that focuses on

85 the process and path of development firms in late-development economies. Whereas previous researchers have examined differences in firm specific strategies, comparing developed nations such as the US and Japan to developing nations in the Asia-Pacific region (Amsden & Chu, 2003; Hart & Linden, 1998; Linden, 2000), this thesis specifically highlights the combined role of the public and private institutions in facilitating high-tech development in a specific industry; it elucidates the linkage of local business groups and multinational corporations for organizational resources. In short, the development of the TFT-LCD industry in Taiwan is accentuated by equity- based alliances and technology linkage with multinational electronic firms where local Taiwanese firms pay royalty fees for multinational corporation’s proprietary technologies. The participation of multinational firms is necessary, because the firms that entered the industry without alliances did not succeed.

In terms of state-business interactions, previous studies have shown that governments in late-development nations performed a leadership role in establishing high-tech industries (Amsden, 1989; Amsden & Chu, 2003; Lall 1995; Matthews, 2002; Wade, 1991; Weiss, 1994). Other studies argue that the state largely focused its attention on providing functional macro-level institutions through providing policy guidance and infrastructure support (World Bank, 1993). This study shows that behind successful creation of high-tech industries, policy support must be in the form of a vertically coordinated package of assistance to the private sector in the form of financial and R&D assistance, along with the establishment of the state-sponsored and designated industrial parks across Taiwan. Furthermore latecomer firms must be assisted by the state to link up with multinational technology providers in the OECD countries.

In terms of the role the ‘large business groups’ perform in advancing technology-intensive industries, the findings show that the involvement of ‘large business group’, both local and multinational, is essential to the success of developing technology-intensive products. Previous studies have shown that business groups have the best chance of success in microelectronic-related products, since these organizations have the financial and organizational capacity to apply to this sector (Chandler, 2005; Linden, 2000). This study has demonstrated that microelectronic firms that are affiliated with one of the top 100 local business groups in Taiwan, and

86 have linkages with multinational microelectronic firms, have the highest chance of success in manufacturing indigenous technology-intensive products. Indeed, firms that succeed in the TFT-LCD industry are those affiliated with ‘business groups’. This is in spite of some of these business groups having no prior experience in microelectronic industry. Previous studies in the role and function of business groups developed by Chandler and related scholars are useful in understanding why some firms become successful in microelectronic industry. Finally, although the role of the state has diminished somewhat over the years, the involvement of the state still remains vital when it come to the provision of infrastructures and R&D capabilities.

7.3 Contributions of the study

In recent years, the ‘institutional-based view’ has become an important pillar in the field of International Business, in conjunction with traditional insights from industry and the resource-based views (Peng, 2008). Interest in how some emerging economies adopt a unique kind of institutional arrangement to gain competitive advantage over traditional multinationals in developed countries has led to the development of new conceptual frameworks (Fuller, 2002; Tsai, 2006). Emerging paradigms argue that dynamic interactions between institutions and organizations allow firms’ strategic choices to be driven by not just industry conditions (Porter, 1980, 1990) or firm capabilities (Barney, 1991), but also by the particular institutions which these firms draw on and interact with.

The acceptance of the institutional-based view (Powell & DiMaggio, 1991) as the legitimate perspective in the field of International Business is demonstrated in this thesis with the case study of the creation of the TFT-LCD industry in Taiwan. More importantly, the thesis demonstrates the utility of the ‘triple-alliance’ perspective (Evans, 1977, 1982), where multinationals, state and local business cooperate to enter the technology-intensive industry, such as TFT-LCD ,and become globally competitive.

The thesis also adds to the discussion on industrial policy and makes several contributions for policy decision makers: First, it demonstrates that the mode of

87 industrialization is increasingly shifting toward a more institution-based alliance across borders, as opposed to the traditional view that the nation must monopolize production and substitute foreign imports with domestic-made products. Second, the involvement of large business groups is likely to determine the success of the industrialization effort. The influence of business groups on the economy is likely to increase as the state partners with business groups for continual technology upgrading. The quality of interactions between institutions is also important. It must be emphasized that institutions, in this case, should not be viewed in abstract form, but rather as distinct social-economic organizations that are capable of interacting with each other effectively to foster the creation of technology-intensive products.

7.4 Limitations of the study

In terms of quantitative studies, this study is limited by the sample scope and size of 100 electronic firms that traded publicly in 2006. This study does not take into account the performance of the same firms in other years, or the impact of other privately held firms; therefore, the generalizability of the results is somewhat limited. As well, caution needs to be exercised when drawing implications for other latecomer firms because the study’s scope is rather narrow. Moreover, firms that previously performed above average may no longer be able to sustain similar performance in the light of the recent global financial crisis. Thus, a valid research question for future studies is whether the ‘triple-alliance’ model is applicable to firms that have been wracked by the global financial crisis and whether, on average, firms that had such alliance performed better than firms with no such alliance in place.

In addition, while the case evidence clearly shows the benefit of such an alliance to firms in late-industrializing countries, the question of whether such a strategy remains beneficial to both multinational corporations and the state in the long run remains unanswered. In summary, the data limitations and the narrow scope of the study could potentially limit the validity of the proposed conceptual framework. Thus caution needs to be exercised when generalizing the applicability of the study to the latecomer firms and industrializers.

88

7.5 Future directions

This study has drawn heavily from disciplines including development economics, political science and sociology. The nature of this study suggests several pathways for future research. The first proposition focuses on the understanding of how technology transfer takes place in the context of the triple-alliance. While the formation of alliances among institutions induces technology transfers, the actual process of technology transfer and the process of learning to master these technologies require more detailed studies.

In terms of qualitative studies, research could focus on one of the large TFT- LCD firms in Taiwan. This narrow and in-depth approach would allow researchers more intimate observations of the mechanism of the technology transfer to, and assimilation in, latecomer firms. Finally, the empirical design of this study could be extended to other sectors and segments of the electronic industry. The empirical design of the study could also be extended to private owned firms in the electronic industry.

89 References

Akamatsu, Kaname (1932) A historical pattern of economic growth in developing countries. The Developing Economies, 86, 196-217.

Akinwande, Akintunde; Fuller, Douglas; Sodini, Charles (2003) Leading, Following or Cooked Goose? Innovation Successes and Failures in Taiwan's Electronics Industry. Industry & Innovation, 10(2), 179-196.

Amsden, Alice H. (1989) Asia’s Next Giant: South Korea and Late Industrialization. Oxford: Oxford University Press.

Amsden, Alice H. (1991) Diffusion of development: the late industrializing countries and greater East Asia. AEA Papers and Proceedings, 81(2), 282-286.

Amsden, Alice H. (2001) The Rise of “The Rest”: Challenges to the West From Late- Industrializing Economies. New York: Oxford University Press.

Amsden, Alice & Chu, Wan-wen (2003) Beyond Late Development: Taiwan’s Upgrading Policies. Cambridge: MIT Press.

Bair, Jennifer (2005) Global capitalism and commodity chain: Looking back and looking forward. Yale University Press.

Balasssa, Bela (1982) Development Strategies in Semi-Industrial Economies. Baltimore: John Hopkins University Press.

Balassa, Bela (1988) The Lessons of East Asian Development: An Overview. Economic Development and Cultural Change, 35(3), 273-290.

Barney, Jay (1991) Firm Resources and Sustained Competitive. Advantage. Journal of Management, 17, 99-120.

Bernard, Mitchell & Ravenhill, John (1995) Beyond Product Cycles and Flying Geese. World Politics. January, 47, 171-209.

90 Biggart, Nicole & Hamilton, Gary (1988) Market, Culture, and Authority: A Comparative Analysis of Management and Organization in the Far East. The American Journal of Sociology, 94, 52-94.

Breznitz, Dan (2005) Development, flexibility and R & D performance in the Taiwanese IT industry: capability creation and the effects of state-industry co-evolution. Industrial and Corporate Change, 14(1), 1-35.

Boulton, William (1997) WTEC Panel Report on Electronics Manufacturing in the Pacific Rim. World Technology Evaluation Center. May, 1997.

Campell, J.P., Daft, Richard & Hulin, C.L. (1982) What to study: Generating and developing research questions. California: Sage Publication.

Caves, Richard E. & Ghemawat, Pankaj (1992) Identifying Mobility Barriers. Strategic Management Journal, 13(1), 1-12.

CCIS (2000) The Largest 100 Corporations in Taiwan. Published by China Credit Information Service, Taipei, Taiwan.

Chandler, Alfred D. Jr. (1977) The Visible Hand: the Managerial Revolution in American Business. Cambridge: MIT Press.

Chandler, Alfred D. Jr. (1990) Strategy and Structure: Chapters in the History of the Industrial Enterprise. Cambridge: MIT Press.

Chandler, Alfred D. Jr. (2005) Commercializing High-Technology Industries. Business History Review, 79(3), 595-605.

Chang, H.J & Rowthorn, B. (1995) The Role of the State in Economic Change. World Institute for Development Economics Research. Oxford University Press.

Chang, Tsou P. (2008) Development Trends in Taiwan’s Opto-electronics Industry. Technovation, 22, 161-173.

Chen, Cheng-Fen & Sewell, Graham (1996) Strategies for technological development in South Korea and Taiwan: the case of semiconductors. Research Policy, 25, 759-783.

91 Chen, Shin-horang (2002) Global Production Networks and Information Technology: the case of Taiwan. Industry & Innovation, 9(3), 249-265.

Cheng Sen-Hao (2007) A Study of Industrial Technology Research Institution of Taiwan. R&D Management, 37(3), 31-47.

Cho, Tong-song & Matthews, John (2000) Tiger Technology: The Creation of a Semiconductor Industry in East Asia. Cambridge University Press.

Chou, T.L. & Kirkby, R.J.R. (1998) Taiwan’s Electronics Sector: Restructuring of Form and Space. Competition and Change, 2, 331-358.

Chu, Y. P & Hu, S.C. (1999) The Political Economy of Taiwan’s Development into the 21st century. Berkeley: University of California Press.

Chu, Yun-Han (1999) The Institutional Foundation of Taiwan’s Industrialization: Exploring the State-society Nexus in Chu, Y. P & Hu, S.C. (ed) The Political Economy of Taiwan's Development into the 21st century. Berkeley: University of California Press.

Chung, Chi-nien (2001) Markets, culture and institutions: The emergence of large business groups in Taiwan, 1950s-1970s. Journal of Management Studies, 38, 719-745.

Chung, Chi-nien (2005) Beyond Guanxi: Network Contingencies in Taiwanese Business Groups. Organization Studies, 27(4), 461-489.

Chung, Chau-Chang & Ernst, Dieter (2000) Taiwan enterprises in global perspective. M. E Sharpe Inc. Publication.

Cumings, Bruce. (1987) The Origins and Development of the Northeast Asian Political Economy, in Deyo, F.C. (ed), The Political Economy of the New Asian Industrialism. Ithaca, NY: Cornell University Press.

Dedrick, Jason & Kraemer, Kenneth L (1998) Asia’s Computer Challenge: Threat or Opportunity for the United States & the World. Oxford University Press.

92 Deyo, Frederic C. (1987) The Political Economy of the New Asian Industrialism. Ithaca, NY: Cornell University Press.

Dunning, John H. (1997) Alliance Capitalism and Global Business. London: Routledge Press.

Einhorn, Bruce (2006) Why Taiwan Matters? Business Week. May 16th. McGraw-Hill Publishing.

Eisenhardt, Kathleen M (1989) Building Theories from Case Study Research. The Academy of Management Review, 14(4), 532-550.

Eisenhardt, Kathleen M (1991) Better Stories and Better Constructs: The Case for Rigor and Comparative Logic. The Academy of Management Review, 16(3), 620-627.

Ernst, Dieter. (2000) Inter-organizational knowledge outsourcing: What permits small Taiwanese firms to compete in the computer industry? Asia Pacific Journal of Management, 17, 223-255.

Evans, Peter (1977) Multinationals, state-owned corporations, and the Transformation of Imperialism: A Brazilian case study. Economic Development and Cultural Change, 26(1), 43-64.

Evans, Peter (1982) Reinventing the Bourgeoisie: State Entrepreneurship and Class Formation in Dependent Capitalist Development. The American Journal of Sociology, 88, 210-247.

Evans, Peter B (1985) Bringing the State Back in. Cambridge University Press.

Evans, Peter (1995) Embedded Autonomy: States and Industrial Transformation. Princeton, New Jersey: Princeton University Press.

Fransman, Martin (1993) The market and beyond: information technology in Japan. Cambridge University Press.

93 Fuller, Dogulas (2002) Globalization for nation building: industrial policy for high- technology products in Taiwan. Industrial Performance Centre Working Paper. Cambridge: MIT Press.

Gelman, Andrew & Hill, Jennifer (2006) Data Analysis Using Regression and Multilevel/Hierarchical Models. Cambridge University Press.

Gereffi, Gary (1999) International trade and industrial upgrading in the apparel commodity chain. Journal of International Economics, 48, 37-70.

Global Competitiveness Report 2005. Published by World Economic Forum, Sweden.

Gold, Thomas (1986) State and Society in the Taiwan Miracle. Armonk, NY: M.E. Sharpe.

Granovetter, Mark (1985) Economic Action and Social Structure: The Problem of Embeddedness. The American Journal of Sociology, 91(3), 481-510.

Gujarati, Damodar N. (1995) Basic econometrics. McGraw-Hill Publication.

Haggard, Stephan (1990) Pathways from the Periphery: The Politics of Growth in the Newly Industrializing Countries. Ithaca, NY: Cornell University Press.

Hamilton, G. Garry & Kao, Che-Shu. (1987) Max Weber and the Analysis of East Asian Industrialization. International Sociology, 2(3), 289-300.

Hart, Murtha & Linden, Gregory (1998) Flying geese as moving targests: Are Korea and Taiwan catching up with Japan in advanced displays? Industry and Innovation, 5(1), 11-34.

Hart-Landsberg, Martin & Burkett, Paul (1998) Contradictions of capitalist industrialization in East Asia: a critique of “Flying Geese” theories of development. Economic Geography, 74(2), 87-110. Hobday, Michael. (1995) Innovation in East Asia: the challenge to Japan. Edward Elgar Press.

94 Hong, Sung-gul (1997) The political economy of industrial policy in East Asia: the semiconductor industry in Taiwan and South Korea. London: Edward Elgar Publication.

Hsu, J.Y. & Saxenian A.L. (2001) The Silicon Valley-Hsinchu Connection: Technocial Communities and Industrial Upgrading. Industrial and Corporate Change, 10(4), 893-919.

Hung, Shiu-wan (2006) Competitive Strategies for Taiwan TFT-LCD Industry. Technology in Society, 28, 349-361.

Johnson, Chalmers A. (1982) MITI and the Japanese Miracle: The Growth of Industrial Policy, 1925-1975. Stanford University Press.

Johnson, Chalmers (1995) Japan: Who Governs? The Rise of Developmental State. New York: W.W. Norton & Company.

Kim, Eun Mee (1997) Big Business, Strong State. Albany: State University of New York Press.

Kim, Linsu (2003) The dynamics of technology development: Lesson from the Korean experience. pp. 142-167 in Competitiveness FDI and Technological Activity in East Asia. Cheltenham: Edward Elgar.

Kojima, Kiyoshi (2003) The Flying-Geese Theory of Economic Development. Bunshindo, Tokyo.

Krugman, Paul. (1994) The Myth of East Asian Miracle. Foreign Affairs, 73(6), 62-78.

Lall, Sanjaya (1994) The East Asian miracle: Does the bell toll for industrial strategy? The World Development, 22(4), 645-654.

Lall, Sanjaya (1995) Developing industrial technology: lessons for policy and practice. World Bank operations evaluation study. World Bank: Washington, D.C.

95 Lall, Sanjaya (1996) Paradigms of Development: The East Asian Debate on Industrial Policy, in Lall, S. (ed.) Learning from the Asian Tigers Studies in Technology and Industrial Policy. St. Martin’s Press.

Lall, Sanjaya (1999) Promoting industrial competitiveness in developing countries: lessons from Asia. Commonwealth economic papers, Commonwealth Secretariat.

Lall, Sanjaya (2005) Understanding FDI-Assisted Economic Development. Routledge Press.

Lall, Sanjaya & Latsch, Wolfram (1998) Import Liberalization and Industrial Performance: The Conceptual Underpinnings. Development and Change. 29(3), 437-465.

Lall, Sanjaya & Latsch, Wolfram (2006) The “sophistication” of exports: A new trade measure. World Development, 34(2), 222-223.

Left, Nathaniel (1978) Industrial Organization and Entrepreneurship in the Developing Countries: The Economic Groups. Economic Development and Cultural Change, 26(4), 661-675.

Li, Kuo Ting (1995) The Evolution of Policy behind Taiwan's Development Success. World Scientific Press.

Linden, Gregory (1995) Taiwan’s Flat Panel Display Policy. University of California Graduate School of Public of Policy.

Linden, Gregory & Murtha, Thomas (1998) Flying geese as moving targests: Are Korea and Taiwan catching up with Japan in advanced displays? Industry and Innovation, 5(1), 11-34.

Linden, Gregory (2000) Industrial Policy, Technology, and Performance: Lessons from the East Asian Electronics Industry. PhD Dissertation. University of California Graduate School of Public of Policy.

Lindt, Gregory M (1995) Taiwan’s flat panel policy. Working paper series. Graduate School of Public Policy, UC Berkeley.

96 Little, Ian (1979) Economic Growth and Structural Change in Taiwan. Ithaca: Cornell University Press.

Little, Ian (1982) Economic Development: Theory, Policy, and International Relations. New York: Basic Boos Publication.

Matthews, John (1995) High-technology industrialization in East Asia: the case of the semiconductor industry in Taiwan and Korea. Chung-Hua Institution for Economic Research, Taiwan.

Matthews, John A. (1997) A Silicon Valley of the East: Creating Taiwan’s Semiconductor Industry. California Management Review, 39(4), 26-66.

Matthews, John A (1999) A silicon island of the East: Creating a semiconductor industry. Journal of World Business, 34(2), 139-156.

Matthews, John A (2002) Dragon Multinational: A New Model for Global Growth. New York: Oxford University Press.

Matthews, John A (2003) Competitive dynamics and economic learning: an extended resource-based view. Industrial and Corporate Change, 12(1), 115-145.

Matthews, John A (2005) Strategy and crystal cycle. California Management Review, 47(2), 6-32.

Matthews, John A (2006) Strategizing, Disequilibrium and Profit. Stanford, CA: Stanford University Press.

Miau, Fong Chang (1992) Creating Win-Win: Story of Mio. Taipei: Mitac Publishing.

Morikawa, Hidemasa (1992) Zaibatsu: the rise and fall of family enterprise groups in Japan. University of Tokyo Press.

Murtha, Thomas P. (2001) Managing New Industry Creation: Global Knowledge Formation and Entrepreneurship in High Technology. California: Stanford University Press.

Noble, Gregory W. (1998) Collective Action in East Asia: How Ruling Parties Shape Industrial Policy. Ithaca: Cornell University Press.

97 Numazaki, Ichiro (1988) Network of Taiwanese Big Business: A Preliminary Analysis. Modern China. 12(4), 487-534.

Numazaki, Ichiro (1991) State and Business in Postwar Taiwan: Comments on Hamilton and Biggart. The American Journal of Sociology, 96(4), 993-999.

Onis, Ziya (1990) The Logic of the Developmental State. Comparative Politics, 24(1), 109-126.

Orrus, M; Biggart, N; & Hamilton, G. (1997) The Economic Organization of East Asian Capitalism. Thousand Oaks, CA: Sage Publications.

Pallant, Julie (2007) SPSS Survival Guide: A guide to data analysis using SPSS for Version 15, 3rd edition. Allen & Unwin Press.

Peng, Mike W (2008) An Institution-based View of International Business Strategy: a Focus on Emerging Economies. Journal of International Business Studies, 39, 920-936.

Pentland, Brian T (1999) Building Process Theory with Narrative: From Description to Explanation. The Academy of Management Review, 24(4), 711-724.

Pettigrew, Andrew M (1990) Longitudinal field research on change. Organization Science, 1(3), 267-292.

Poon, Teresa (1998) Inter-firm Networks and the Dynamic Process of Industrial Upgrading: A Case Study of the Information Technology Industry in Taiwan. University of Sydney Press.

Porter, Michael E. (1980) Competitive Strategy. New York: Free Press.

Porter, Michael (1990) The competitive advantage of nations. New York: Free Press.

Porter, Michale & Sakakibara, Mariko (2004) Competition in Japan. Journal of Economic Perspective, 18(1), 27-50.

Powell, Walter W. & DiMaggio, Paul J. (1991) The new institutionalism in organizational analysis. University of Chicago Press.

98 Raudenbush, Stephen W & Bryk, Anthony S. (2002) Hierarchical linear models: applications and data analysis methods: Advanced quantitative techniques in the social sciences, 2nd edition. Sage Publication.

Ray, Pradeep K (1998) The Development of Telecommunication Industry in India. PhD Dissertation. University of Western Australia.

Schive, Chi (1990) The foreign factor: the multinational corporation's contribution to the economic modernization of the Republic of China. California: Hoover Institution Press.

Scott, John (1991) Social Network Analysis: a Handbook. London: Sage Publication.

Shih, Tsen Zhoung (1996) Remaking Acer. Taipei: Commonwealth Publishing

Simon, Denis (1992) Taiwan’s Strategy for Creating Competitive Advantage: The Role of the State in Managing Foreign Technology in N. T. Wang (ed) Taiwan enterprises in global perspective. M. E Sharpe Publication.

Spencer, Jennifer W; Murtha, Thomas P; Lenway, Stenfanie A (2003) How government matters to new industry creation. GW Center for the Study of Globalization. George Washington University.

Stiglitz, Joseph (1996) Some Lessons from the East Asian Miracle. World Bank Research Observer, 11(2), 151-178.

Tabachnick, Barbara G. & Fidell, Linda S. (2001) Using multivariate statistics. Allyn and Bacon Press.

Tezuka, H (1997) Success as the source of failure? Competition and cooperation in the Japanese economy. Sloan Management Review, Winter, 83-93.

Tsai, Terence (2006) The Silicon Dragon: High-Tech Industry in Taiwan. London: Edward Elgar Publication.

Tung, An-Chi (2001) Taiwan’s Semiconductor Industry: What the State Did and Did Not. Review of Development Economic, 5(2), 226-288.

99 Van de Ven, Andrew H (1992) Suggestions for studying strategy process: A Research Note. Strategic Management Journal (special issue). 13, 169-191.

Vernon, Raymond (1971) Sovereignty at Bay: The Multinational Spread of U.S. Enterprises. New York: Basic Book.

Wade, Robert (1984) Developmental States in East Asia. IDS Bulletin, International Development Study Institute, United Kingdom.

Wade, Robert (1990) Governing the market: economic theory and the role of government in East Asian industrialization. New Jersey: Princeton University Press.

Wade, Robert (1995) Resolving the State market Dilemma in East Asia. In Ha Joon Chang & Robert Rowthor (eds) The role of state in economic change. Oxford, Clarendon Press.

Wang, Chi-ren (2002) Growth Strategies of Chi Mei Group. Occasional Working Paper Series. Department of Industrial Management, Cheng-kong University.

Wang, J.-C. Lin, C.-H. Tsai, K.-H. (2007) Science and technology manpower policy and an estimation of high tech manpower demand for the regional operation centre: the case of Taiwan. International Journal of Technology Management, 38(3), 268-291.

Wang, Vincent Wei-cheng (1995) Developing the Information Industry in Taiwan: Entrepreneurial State, Guerrilla Capitalists, and Accommodative Technologists. Pacific Affairs, 68(4), 551-576.

Weiss, Linda M. (1994) Innovation alliances in Taiwan: A coordinated approach to developing and diffusing technology. Journal of Industry Studies, 1(2), 91- 101.

Weiss, Linda M. (1995) Governed Interdependence: Rethinking the government- business relationships in East Asia. The Pacific Review, 8(4), 25-50.

100 Weiss, Linda M. (2003) States in the Global Economy: Bringing Domestic Institutions Back in. Cambridge: Cambridge University Press.

Whitley, Richard (2000) Divergent Capitalisms: the Social structuring and change of business systems. Oxford University Press.

William, Boulton R. (1997) WTEC Panel Report on Electronics Manufacturing in the Pacific Rim. World Technology Evaluation Center Research Paper. California, USA.

Wong, Poh K. & Matthews, John (1998) Competing in the Global Flat Panel Industry. Industry & Innovation, 5(1), 1-10.

World Bank (1993) The East Asian Miracle: Economic Growth and Public Policy. Oxford University Press.

Wu, Se-Hwa (1992) The dynamic cooperation between government and enterprise: The development of Taiwan’s integrated circuit industry in Wang, N.T (ed) Taiwan enterprises in global perspective. M. E Sharpe Publication.

Yin, Robert K (1989) Case study research. California: Sage Publications.

101 Publicly available sources for case studies

Government Publications and Industry Reports (in Mandarin, Chinese)

Council for Economic Planning and Development, Executive Yuan, Taiwan Appraisal for Ten Major Development Projects (Published 1979). 2. Taiwan Statistical Data Book (Published for various years). 3. Annual Economic Development Report, R.O.C., Taiwan (Published from 1990 to 2007). 4. Council for Economic Planning and Development, Annual Report, 2005. Published by Executive Yuan. Taipei, Taiwan.

Institute for Information Industry, Taiwan 1. Information Industry Yearbook, Taipei, Taiwan (Published 1990 to 2005). 2. Taiwan Microelectronic Industry overview, Institute of Information Industry, 2007.

Industry & Technology Intelligence Service, Taiwan (ITIS) 1. Developments and Prospect for the FPD Industry. IEK industry commentary, published March 26, 2003. 2. Taiwan’s Flat Panel Industry: Outlook for 2010. IEK industry report, published April, 17, 2008. 3. Developments and Prospects for the Flat-Panel Display Industry. Published by Industrial Economic Knowledge Center (IEKC) from 1995 to 2003.

Industrial Technology Research Institute, Taiwan 1. The Investigation of Optical Electronic Industry. Published by Optoelectronic and Systems Laboratories (OSL) from Year 1990 to Year 2007. 2. The Investigation of Taiwan’s Optical Industry. Published by Taiwan Optoelectronic and Semiconductor Equipment Association (TOSEA) from Year 1990 to Year 2007.

102 Ministry of Economic Affairs, Executive Yuan, Taiwan (MOEA) 1. Industrial Technology Developing Strategy, MOEA Publication. Published 1998. 2. Taiwan Economic Statistic, MOEA Publication. Published from 1990 to 2005. 3. Technology Development Program, MOEA Technology Development Program Publication. Published 1996. 4. Hsinchu Science Park Administration Annual Report, 1995-2001. (HSPA). Published by Ministry of Economic Affairs. Taipei, Taiwan. 5. Science & Technology White Paper for 2005. Published by Ministry of Economic Affairs. Taipei, Taiwan.

The Photonics Industry & Technology Development Association (PIDA) 1. Optolink Reports. Published from Jan. 1993 to Jan. 2007.

Market Itnelligence Center (MIC) 1. MIC (2003) Annual Global TFT-LCD industry outlook report. Market Intelligence Center, Institute for Information Technology Publication.

MITAC 1. MITAC (2003) Annual company report for Mitac incorporated. Mitac Inc. investor relations service.

China Credit Information Service (CCIS) 1. CCIS (2000) Annual survey of the largest one hundred corporations in Taiwan. China credit information service.

Newspapers and Periodicals Business Weekly (in Mandarin, Chinese) [From 1989 to 2006] Far Eastern Economic Review (FEER) [From 1985 to 2004] Fortune Magazine [From 1994 to 1996] The Economist Daily (in Mandarin, Chinese) [From 1990 to 2006] Tian-xie Magazine (in Mandarin, Chinese) [From 1989 to 2006] United Economic Daily (in Mandarin, Chinese) [From 1990 to 2006]

103 Appendix 1: List of interviewees

Mr. Chang, Pao-Long Assistant Reseracher The Photonics Industry & Technology Development Association

Mr. Chen, Yun-Chen Column Reporter of TFT-LCD Industry The United Economic Daily

Mr. Hsu, Chiung-Wen Assistant Public Relations Manager Chi Mei Display

Mr. Lin, Yan-Kang Display Industry Deputy Director Industrial Development Bureau, Ministry of Economic Affairs

Mr. Liu, Chung-Yuan Assistant Researcher in Electronic Industry Industrial Technology Research Institute

104 Appendix 2: Participant information statement and consent form

THE UNIVERSITY OF NEW SOUTH WALES SCHOOL OF ORGANIZATION & MANAGMENT

PARTICIPANT INFORMATION STATEMENT AND CONSENT FORM

Project Title: Evolution of Taiwan’s Electronic Industry: Case of TFT- LCD Industry.

Dear