Paper to be presented at the 25th Celebration Conference 2008 on
ENTREPRENEURSHIP AND INNOVATION - ORGANIZATIONS, INSTITUTIONS, SYSTEMS AND REGIONS Copenhagen, CBS, Denmark, June 17 - 20, 2008
LATECOMER STRATEGIES: THE EXPERIENCE OF THE TFT-LCD INDUSTRY IN TAIWAN Wei Zhang Zhejiang University, China
[email protected] Yongjiang Shi Institute for Manufacturing,Cambridge University
[email protected] Xiaobo Wu School of Management, Zhejiang University
[email protected] Yusheng Teng Institute for Manufacturing,Cambridge University
[email protected]
Abstract: In this paper, we have reviewed the notable rise of Taiwanese TFT-LCD panel manufacturers to answer two research questions: what latecomer advantages and disadvantages they had and what strategies they employed to utilize the advantages or to surmount the disadvantages. Besides confirming the feasibility of some factors addressed by previous literature such as differentiation advantage, we find that the competition dynamics between incumbents worked as a major source of latecomer advantages for these Taiwanese firms to penetrate into the industry and keep growing. Moreover, we discuss the influence of technological regime and argue that the incremental technological advance rather than radical innovation provides latecomers with opportunity to utilize the learning curve effect. We also discover and discuss new forms of latecomer advantages and some subtle tactics, which enrich the existing knowledge about latecomers.
JEL - codes: O, -, -
Latecomer Strategies: the Experience of the TFT-LCD Industry in Taiwan Abstract: In this paper, we have reviewed the notable rise of Taiwanese TFT-LCD panel manufacturers to answer two research questions: what latecomer advantages and disadvantages they had and what strategies they employed to utilize the advantages or to surmount the disadvantages. Besides confirming the feasibility of some factors addressed by previous literature such as differentiation advantage, we find that the competition dynamics between incumbents worked as a major source of latecomer advantages for these Taiwanese firms to penetrate into the industry and keep growing. Moreover, we discuss the influence of technological regime and argue that the incremental technological advance rather than radical innovation provides latecomers with opportunity to utilize the learning curve effect. We also discover and discuss new forms of latecomer advantages and some subtle tactics, which enrich the existing knowledge about latecomers. Keywords: Latecomer advantage and disadvantage; Latecomer strategy; Catch-up; TFT-LCD
1. Introduction Since the 1980s, the rise of newly industrializing economies (NIEs) in East Asia as a global economic power has been one of the central dimensions of global transformation (World Bank, 1993). In some products areas, for example, microcomputer hardware, consumer electronics and semiconductors, latecomer firms within this region have gradually developed innovation capability and become remarkable rivals to MNCs from Japan, US and Europe (Hobday, 1995; Mathews and Cho, 2000). The spectacular success of Taiwanese firms in thin-film transistor liquid crystal display (TFT-LCD) industry provides a recent example. Taiwan accounted for only 1% of market share worldwide in 1998. But in 2007, led by flagship producers AU Optronics, Chi Mei Optoelectronics, Chunghwa Picture Tubes, Hannstar, and Innolux, Taiwan is projected to hold the highest global market share for TFT-LCD production capacity, at 43%, followed by South Korea with 40%. How can latecomers succeed in catching up with early movers and even take over the leadership position? The analysis on latecomer advantages and disadvantages probably is a pivotal approach to answer this question. Hobday (1995) defines the latecomer firm as “a manufacturing company (existing or potential) which faces competitive disadvantages in attempting to compete in exports markets”. Mathews and Cho (1999, 2000) also note the meagre resource and competitive disabilities of
latecomer. Therefore, to overcome these inherent disadvantages becomes the inevitable challenge for latecomers when they enter new sectors. Without solutions to cope with these disadvantages, they will have no chance to compete powerful and experienced incumbents and even no chance to survive. Nevertheless, the late entry also provide opportunities for latecomers since they can make use of free-rider effects and other industrial or technological dynamics. In short, what advantages and disadvantages latecomers had and what strategies they employed to utilize advantages or to surmount disadvantages can be vital to explain their success or failure in catch-up. The achievement of Taiwanese TFT-LCD manufactures has already drawn quite a few attentions from academia. In existing literature, the strategic position of Taiwanese TFT-LCD firms is described as classic or superior quick follower that focus on being a superior quick follower rather than challenging the technology leaders (Fuller et al. 2003; Hung, 2006). Regarded as a crucial strategy to overcome technological disabilities, their acquisition of the cast-off technology from Japanese leaders is wildly addressed (Fuller et al. 2003; Mathews, 2004). Some other strategies are also particularly explored. For instance, Mathews (2004) underlines the timing of strategic choices and argues that TFT-LCD firms in Taiwan entered and unleashed investments during industrial downturns which provide “window of opportunity” for latecomers. Based on case study, Chang (2005) discusses the benefits obtained from mergers that result in more complete product lines and integrated resources to enhance competitiveness. Besides the discussions at firm level, other literature examines wider economic and social context and considers the positive impact of institutional structures as a source of technical efficiency and market opportunity. The perspective of national innovation system (NIS) is used to examine key elements that support the implementation of the follower strategy, including government policy, human resources, capital investment, and a bridging institution (Hung, 2006). However some researchers believe that the role of local government has only been secondary in the development of Taiwanese TFT-LCD industry in contrast to Taiwan’s semiconductor development experience described by Hobday’s latecomer model (Hobday, 1995), which considers government as the key role and frontier at the initial stage for rapid learning and technological catch-up (Hsing, 2006). Comparative studies on competitiveness of Japanese, Korean and Taiwanese TFT-LCD firms are also conducted. Good manufacturing flexibility, strong demand from downstream manufacturers, and strong will to invest are regarded as the main advantages, whereas the disadvantages lie in the dependent on Japan’s technology license and key components imported from Japan and US (Chang, 2005; Hsing, 2006). However, to our knowledge little work has been carried out on the systematic examination of Taiwanese TFT-LCD panel manufacturers’ advantage and disadvantage as latecomers in the process of catch-up. The actual situation they faced and the drivers of their strategies remain unexplored. The existing literature on comparative studies has not paid particular attention to the effect of late entry and looks like a snapshot rather than a longitudinal review. Hence Taiwanese firms’ evolution from latecomers to major players has not been convincingly explained.
This paper seeks to shed some light on the development process of Taiwanese TFT-LCD panel manufacturers, especially the early stage since their entry, by investigating their (dis)advantages resulting from the late entry and corresponding strategies. We intend focusing on firm-level competitive advantages and strategies stemming from late entry, thereby wider institutional contexts such as government policy are outside the scope of this paper. The paper is divided into sections dealing with introduction, literature review, case study, discussion on findings and is summed up with concluding remarks.
2. Latecomer disadvantages and advantages The research on latecomer advantages and disadvantages can be traced back to the work of industrial economics researchers who examine entry order effects and mainly focus how early entrants benefit from barriers that deter latecomers (Bain, 1956). Subsequently, first-mover advantages draw wider attentions in strategic management, marketing and economics. With different perspectives and approaches including resource-based view (RBV), transaction costs theory and cognitive economics, numerous studies have be done to explore the source, form and consequence of early mover advantages (Lieberman and Montgomery, 1988, 1998; Kardes et al., 1993; Muthukrishnan, 1995). Meanwhile, first-mover disadvantages and follower advantages also gradually attract significant attention (Schnaars, 1994; Berndt et al.,1995; Zhang and Markman, 1998). Particularly, along with the rise of NIEs some researchers shed light into these new entrants and investigate the feasibility of catch-up or even leapfrogging, especially from the perspective of technological changes (Brezis et al. 1993; Gruber, 1996). Cho et al.(1998) examine the advantages or early movers versus latecomers and group them into three areas: market, competition and the firms. Applying this framework, they investigate strategies employed by Japanese and Korean semiconductor firms to overcome latecomer disadvantages and utilize latecomer advantages. Their work can be regarded as a pioneering attempt to systematically analyze latecomer (dis)advantages and strategies but has some flaws. They have not paid particular attention to the characteristics of product and technology, hence their framework remains questionable to interpret the leadership changes in high-tech industries which are characterized by highly uncertain technology dynamics. Moreover, they have not distinguished between “latecomer” and “late entrant”. As argued by Mathews (2002), latecomer firms enter a industry not by strategic choice. They are condemned to be followers by history and have to make the best of their resource-poor initial situation. This notion fits the actual situation of East Asian latecomers much better than considering resourcefulness as latecomers advantage in Cho et al.’s framework. At least, resourcefulness may be enjoyed by some late entrants but evidently cannot be regarded as a generic advantage due to late entry. Based on above discussion, in this study we propose a conceptual framework which categorizes latecomer (dis)advantages into three aspects: demand, competition and the features of technology and product.
Demand One of the most addressed first-mover advantages or latecomer disadvantages in literature is the cost of switching (Liberman and Montgomery, 1988, 1998). Switching cost can result from tangible sources such as formal contracts and buyers’ initial investment or transaction cost to a particular product (Cho et al., 1998). It can also result from more intangible ones including customers loyalty due to their cognitive processes, uncertainty avoidance and accumulate experience with the pioneer’s product(Carpenter and Nakamoto, 1994). Especially, first-mover advantages can be significantly enhanced by the ‘network externalities” of pioneer’s product as the industry standard (Katz and Shapiro, 1994). Particularly, for latecomer SMEs the reputation disadvantage is more evident because these firms do not have legitimacy and stable links to customers and supporters (Stinchcombe, 1965; Eisenhardt and Schoonhoven, 1990; Gao et al., 2007). However, latecomers may also benefit from free-rider effects stemming from the evolution of market. They can avoid the costs of educating customers incurred by the first movers. In addition, the market changes over time and the direction of change is not entirely predictable. Changes in consumer tastes may offer acute latecomers opportunities to disrupt incumbents’ leading position (Kerin et al. 1992; Cho et al., 1998). Competition First-movers can benefit from preempting resources of various types. (Liberman and Montgomery, 1998). These resources include superior positions in geographic space, human resources, suppliers and raw materials (Cho et al., 1998). Preemption may also interact with market opportunities, particularly when first mover focus on exploiting the competitive vacuum in the market. In general latecomers have to face saturated market and fierce competition when they enter the industry. But in another point of view, competition and competitors may also provide opportunities for latecomers. Latecomers can gain differentiation advantage by simply doing things differently-for example, altering the configuration of value chain activities involved in designing, producing, marketing, delivering, and supporting a product (Buaron 1981). The irreversibility of investments in any of these areas might preclude the first-movers from adopting the strategies and tactics of later entrants that have the benefit of hindsight (Porter 1980). Besides the inertia of sunk resources, first movers may have another type of inertial force embedded in their organizational behavior, This path dependence is described as “incumbent inertia” (Lieberman and Montgomery, 1988), or “core rigidity” (Leonard-Barton, 1992) or “hard wired” routine (Nelson and Winter, 1982). Furthermore, latecomers also can achieve advantages by learning from the pioneer's mistakes in such areas as positioning (Hauser and Shugan 1983; Liberman and Montgomery, 1988). This learning advantage can also be regarded as a form of free-rider effect. Technology & Product Generally speaking, first-movers master the core technology and use it as a major entry barrier to latecomers. However, the source of (dis)advantages on technology area are much more than that. Lieberman and Montgomery (1998) argue that the magnitude
of first-mover advantages varies greatly across product categories. First-mover advantages or latecomers disadvantages may result from the characteristics of product and technology. For example, a steep learning curve enables first-movers to gain significant advantages over late entrants (Lieberman, 1987). This advantage can be enhanced even further when it is protected by patents. Actually, these product and technology characteristics can be systematically analyzed from the perspective of “technological regime”. Technological regime includes various elements such as technological opportunities, appropriability of innovations, cumulativeness of technical advances, the property of the knowledge base, accessibility to external knowledge flows, uncertainty of the technological trajectory, knowledge gap, and technological cycle time (Malerba and Orsenigo, 1996; Breschi et al.,2000; Lee and Lim, 2001). These elements have been considered to be significantly related to the possibility of latecomers’ catch-up. For instance, Gruber (1996) compares the leadership changes in DRAM and EPROM industries. He argues that the wider range of applications, longer interval and lower level of continuity between different generations of DRAM lead to the more significant scale effect and less significant learning curve effect and therefore provide opportunities for latecomers. Lee (2005) suggests that Korea and Taiwan has achieved higher levels of technological capabilities in such sectors as featured by short cycle time of technology whereas the advanced countries do significantly better in those sectors with longer cycle time. Besides, accessibility to external knowledge flows is evidently vital for latecomers that suffer the lack of necessary technology at the initial stage and resort to technology learning or transfer from first-movers. Moreover, if latecomers can efficiently acquire and absorb external technology they may enjoy significant economic advantage. For instance, research suggests that the cost of imitation is about 65% of the cost of innovation (Mansfield et al., 1981). Again, this is another type of free-rider effect. As a particular feature of technological environment, changes in technology can be a crucial source of advantage for latecomers. Tushman and Anderson (1986) classify technology discontinuities into two types: competence-enhancing and competence-destroying. Competence-destroying discontinuities render obsolete the expertise required to master the technology that it replaces and therefore provide opportunity for latecomers. Similarly, Henderson and Clark (1990) and Henderson (1993) assert that if the shift to the new generation is radical enough, incumbents may be unable to adapt. Bower and Christensen (1995) propose the concept of ‘disruptive technologies’, which create growth in the industries they penetrate or create entirely new industries through the introduction of products and services that are dramatically cheaper, better, and more convenient. These technologies are often overlooked by incumbents and adopted by latecomers as powerful weapons to challenge existing giants. In short, technology and product characteristics can provide latecomers with both disadvantages and advantages. we can presume that latecomers may face disadvantages in unfavorable technology context such as high cumulativeness of technical advances, long cycle time and high level of innovation appropriability. Meanwhile, latecomers
may also benefit from favorable environment including abundance of available knowledge source, short cycle time and the emerging of competitive technology. Since it is difficult to list all the possible advantages and disadvantages stemming from technological regime we only use unfavorable and favorable regime in Table 1, which summarizes the latecomer disadvantages and advantages discussed in this study. This category of advantage source viz. technology and product is a significant difference between our framework and Cho et al.(1998)’s one. For example, whereas Cho et al.(1998) classify learning curve effect to the category of ‘the firm’ we argue that the magnitude of learning curve effect is actually determined by technology and product characteristics within specific industry and then can be capitalized by any first-movers. Hence we regarded it as a latecomer disadvantage in technology and product area. Besides, as we mentioned before, resourcefulness and shared experience or assets can be competitive advantages for any firms no matter first movers or latecomers so they cannot be regarded as latecomer advantage - a special case of competitive advantage resulting from market entry timing (Kerin et al., 1992).
Source Demand
Table 1: Latecomer disadvantages and advantages Latecomer disadvantages Latecomer advantages Switching cost Mature consumer
Competition
Reputation Incumbents’ preemption of input factors
Changes in consumer tastes Differentiation Incumbent inertia
Already fierce competition Technology & Product
Lag in technology
Skipping incumbents’ errors Technology spillover
Learning curve effect
Changes in technology
Unfavorable regime
Favorable regime
3. Case study: The TFT-LCD industry in Taiwan Methodology We conducted case analysis of the Taiwanese TFT-LCD panel makers to investigate what advantages and disadvantages they had as latecomers and what strategy they employed to utilize the advantages or to surmount the disadvantages. This suitability of the case study method for examining these objects is recognized by Schramm (1971): “…the case study asks‘What happened?’ …The essence of a case study, the central tendency among all types of case study, is that it tries to illuminate a decision or set of decisions: why they were taken, how they were implemented, and with what result.”
Data for this study were collected from both interviews and archives. In-depth interviews with executives of panel manufacturers, individual consultants and scholars in local universities were conducted. All the interviewed executives are senior marketing or R&D managers with at least 10 years experience in their current firms so their reviews of the evolutionary path are fairly reliable. In addition to interviews, archival data were also derived from published reports, journals, and government agency website about these industries and firms. Since AU Optronics (AUO) is the currently largest TFT-LCD panel makers in Taiwan and we had specific access to its detailed information via personal relationship, we use AUO as the major example to illustrate our research findings. Overview of Taiwan’s TFT LCD industry Taiwan’s LCD industry started by manufacturing TN LCD and STN LCD which are small-sized and low end displays mainly used in telephones, cell phones, automotive electronics, instruments and meters. In 1990, Industrial Technology Research Institute (ITRI) successfully developed the technology of 3-6 inch TFT panel. In 1994, Unipac Optoelectronics began to set up fabrication plant for small-sized TFT-LCD. In 1997, Chunghwa Pictures Tubes (CPT) signed technology transfer contract with Mitsubishi to obtain third-generation (G3) TFT panel technology. Subsequently, other Taiwanese firms such as Acer Display Technology (ADT), Chi Mei Optoelectronics (CMO), Unipac, Hannstar and Quanta Display (QD) also established cooperation with Japanese firms including IBM Japan, Fujitsu, Matsushita, Toshiba and Sharp. TFT-LCD panel production capacity began to expand rapidly after aggressive investment in Taiwan. In 2004, Taiwan surpassed Korea and became the largest supplier of TFT-LCD panel. In 2007, Taiwan comprises around 43% of the global market. Overview of AUO AUO is the largest manufacturer of large-sized TFT-LCD panels in the world, with approximately 20.4% of global market share and revenues of US$14.81billion in 2007. The company was formed in September 2001 by the merger of Acer Display Technology, Inc.(ADT) and Unipac Optoelectronics Corporation. In October 2006, AUO merged with Quanta Display Inc., forming the company as of today. AUO provides customers a full range of panel sizes and comprehensive applications, offering TFT-LCD panels in sizes ranging from 1.5 inches to 65 inches. Additionally, AUO is the first pure TFT-LCD manufacturer to successfully list at the New York Stock Exchange. Case study results Demand In our analytical framework based on literature review, we assume that customers’ switching cost and lower confidence on new entrants may be the major disadvantages for latecomer firms. However, AUO didn’t encounter significant disadvantages on these aspects. Actually, the entry of AUO and other Taiwanese firms to TFT-LCD industry was not simply the result of their own decision. After the Asian financial crisis, Japanese TFT-LCD firms decided to suspend investment and reduce cost to cope with industry downturn and Korean competitors. At that time, Taiwanese electronic
manufactures had already gain reputation of cost advantage with competitive product quality. In 1998, IBM Japan transferred TFT-LCD panel technology to ADT and treated ADT as a strategic supplier and OEM partner to capitalize its low cost. Matsushita transferred its technology to Unipac in the same year. Hence, at the initial stage AUO didn’t need to carve out customers from incumbents. Actually, AUO are still doing a lot of OEM works for Japanese partners today. TFT-LCD panel is a quite standardized product with universal interface so buyers do not have significant transaction cost in replacing existing suppliers by new ones. Moreover, interviewed managers of AUO and other TFT-LCD firms asserted that although there exists certain degree of customer loyalty, price is still the most important determination for customers since a lot of suppliers are present in this market. Taiwanese manufacturers are famous for their capability on cost reduction. Furthermore, nearly all of these panel makers built fabrication plants in mainland China to utilize the labor cost advantage. For instance, as one of the interviewed manager said, the monthly salary of a assembly operator in Korean LPL (joint venture of LG and Phillips), Taiwanese firms’ plants in Taiwan and in mainland China is about $1,5000, 800 and 200 respectively. Therefore, it’s not too difficult for Taiwanese panel makers to sell products to other price-sensitive customers besides Japanese partners. For AUO, it has an additional advantage to weaken latecomer disadvantage. Though AUO met reputation disadvantage to some extent, as a part of ACER Group, it had vast demand from downstream business within the group such as TV set, cellphone, notebook and desktop monitor. Meanwhile, the reputation of ACER Group also helped AUO gain customer confidence. Given these external and internal features, as a latecomer, disadvantages stemming from switching cost and low confidence didn’t appear notable to AUO, whereas it obviously benefited from free-ride effects stemming from the market, viz avoiding the costs of educating customers. Competition When AUO entered TFT-LCD industry, the competition was already fierce. However the fierce competition also provided opportunity for latecomers. As we mentioned before, AUO and other Taiwanese TFT-LCD firms were helped by Japanese firms to enter this market. Shocked by the Asian financial crisis, Korean manufacturers which entered large-sized TFT-LCD market in 1995 started a worldwide price war to gain US dollar and expand market share. In 1997, the average price of LCD panel decreased by 50% and Korean firms captured 15.6% of the whole market. Therefore Japanese TFT-LCD firms which had been suffering from economic decline since early 1990s met serious challenge and began to seek external partners to transfer TFT-LCD panel technology. By transferring technology, Japanese firms could obtain stable panel supply with competitive cost. Furthermore, Japanese firms also could earn a huge amount of royalty or license fee to raise capital for developing next generation display technologies such LTPS TFT-LCD and OLED. Therefore, Taiwanese firms entered as strategic partners and worked with Japanese leaders to compete with Korean rivals, rather than fighting on their own. For AUO, technology transfers from IBM Japan and Matsushita provided a solid foundation for the
subsequent development. Table 2: Technology source of Taiwanese TFT-LCD firms Year mass Major technology Imported technology Companies Parent company production source ADT Acer 7/99 IBM Japan, Manufacturing, design, Fujitsu MVA* technology Unipac UMC 10/99 Matsushita Construction, manufacturing, design, whole-set technology GD Quanta Q4/01 Sharp Whole-set technology CMO
Chimei
8/99
CPT
Tatung
5/99
Fujitsu, IBM Japan Mitsubishi
Hannstar
Walsin
2/02
Toshiba
MVA technology, Joint venture Construction, manufacturing, design, whole-set technology Construction, manufacturing, design, whole-set technology
MVA: multi-walled vertical alignment Source: Compiled from various sources
Although AUO and other Taiwanese firms got the technology from Japanese, the lack of skilled manpower to manage necessary developing and manufacturing work still remained as a challenge for them in the beginning. Taiwanese panel makers resorted to Japanese engineers to learn from their techniques. Due to the downturn, these engineers were made to retire early by the age of 50. Hence they went to Taiwan to continue working in this industry. Though they were mainly individual consultants, the spread of mouth led to more Japanese coming into Taiwan and playing significant roles in the initial stage. Later on, Taiwanese firms also succeeded to hire Korean engineers. Interestingly, these Korean were attracted by the Taiwanese culture. In Korea, the business are run by huge-sized ‘Chaebols’ with more levels and rigorous hierarchy which frustrates the employees as they have less tendency to promote. Plus, the Korean wives prefer the Taiwanese culture which pays more respect to the female society. Therefore a lot of Korean engineers come to Taiwan in the prime of their life at the age of around 30. At the moment, Hsin Chu Science Park, one of three TFT-LCD clusters in Taiwan, has more than 200 Korean engineers. This talents transfer helped Taiwanese to surmount the latecomer disadvantage caused by pioneers’ preemption of experienced people as a critical input factor. As a latecomer AUO also has employed differentiated positioning to survive and then to grow up in the industry. Unlike most Japanese and Korean leaders pursuing rapid generation upgrade to produce larger and larger size panels, AUO not only makes great efforts in the large size competition but also keeps on expanding production of
small-and medium-sized panel (smaller than 10 inch). Currently AUO is running four G5, two G6 and two G7 fabrication plants for large-sized panels along with four G3.5 and one G4 plants for small- and medium-sized ones. Compared with large-sized panel, small-and medium-sized TFT-LCD panel has wider range of applications and the price is more stable. Therefore, when most firms suffered from the unpredictable demand and price changes of large-sized panel in the industry cycles AUO could achieve scope economies and effectively offset the lost by profiting from the business small- and medium-sized panel such as products for digital cameras, mobile phones and portable games. In 2004, when AUO’s world ranking was No.3 in market share of panels for notebooks and monitors and No.5 in LCD-TV, it had already become the No.1 in the panels for digital cameras and No.3 for car navigations. In recent years, the surge of demand from mobile phone, portable DVD, MP3 and other consumer electronics attracts a lot of TFT-LCD firms. However, the product lines for G5 and above are mainly designed for large-sized TV panels and the specifications such as resolution and thickness of glass substrate are quite different from small- and medium-sized ones. Therefore it is not easy and economical for firms to adjust manufacturing process for smaller panels but AUO can simply utilize its existing fabrication lines to grasp this opportunity. As a result, in 2007 AUO’s shipment of small-and-medium-sized panels surged 80.7% from 2006 and reached 143.07 million when the shipment of large-sized panels increased by 65.9% and reached 80.9 million. For the competition of large-sized panels, we can also witness AUO’s differentiated strategy. While Korean and Japanese firms are focus on large panel AUO endeavors to improve the performance of smaller products. Take the LCD-TV sector in 2004 as a example. AUO didn’t follow aggressive Korean firms but mainly concentrated on panels that were smaller than 40 inch (Figure 1). Meanwhile, AUO launched products with better performance on key attributes such as response time and brightness (Table 3). Therefore it seems that AUO tried to produce differentiated products in mature market and avoid head-to-head competition in new generations. Statistic data shows that the majority of demand was panels smaller than 30-inch at that time and even in 2005 the mainstream LCD-TV screens was still 32-inch. Therefore we infer that AUO preferred more rational strategy to meet the current mainstream demand rather than adventuring in high-end but highly uncertain market. In addition, we can also find quite similar strategy in CMO, the second largest TFT-LCD panel manufacturer in Taiwan.
Source: MIC (2004)
Figure 1: Comparison of product positioning
Size
20.1 inch
26/27 inch
Table 3: Comparison of product performance Companies Solution Response Saturation Contrast time LG Phillips 640 x 480 25 72 400:1 Sharp na na na 500:1 Samsung 640 x 480 16 72 800:1 CMO 640 x 480 16 75 600:1 640 x 480 16 75 600:1 AUO 500:1 800 x 600 16 75 LG Phillips 1280 x 768 12 72 600:1 Sharp 1366 x 768 12 na 800:1 Samsung 1280 x 768 16 72 800:1 CMO 640 x 480 16 75 600:1 1280 x 768 75 600:1 AUO 12
Brightness 450 430 500 450 500 500 600 450 450 550 600
Source: MIC (2004)
Size Percentage Size Percentage
Table 4: Sales volume of TFT-LCD TVs in Q1/2004 20 – 29 inch 30 – 39 inch < 20 inch 42.5% 15% 30/32 37 20/21 22/23/26/27 41.5% others inch inch inch inch 26% 16% 0.5% 13.7% 1.3%
>40 inch 1%
Source: MIC (2004)
Free-rider effect deriving from learning pioneers’ mistake does not appear evident for AUO. According to the opinion of interviewed managers, pioneers’ failures are usually hidden by the companies and not shared. The latecomer advantage caused by pioneers’ irreversible investment does not seem significant in this case. Though building a new generation plant requires totally new equipments rather than improving existing ones, it’s not a problem for Korean or Japanese to afford the vast investment. Technology Among all the features of technological regime, the accessibility to external knowledge flows tends to predominate when it comes to explaining the successful catch-up of Taiwanese TFT-LCD firms. Besides the initial import of technology from IBM Japan and Matsushita, AUO continued acquiring and learning external knowledge, mainly from Japanese firms. Even though Japanese government formulated a National IP Strategy and enacted a Basic Law on Intellectual Property in November 2002 to pursue leading position in innovation, the knowledge inflow from Japan to AUO is not interrupted but the relationship appears more interactive. Table 5: Chronology of AUO’s technology acquisition Company involved Agreement contents Term IBM Japan Transfer and cooperative Mar, 1998-Mar, 2005 R&D of TFT-LCD technology Matsushita Cooperative R&D and Jun, 1998-Oct, 2003 authorization of TFT-LCD technology Fujitsu Limited Authorization of MVA Jun, 2000-Jun, 2005 LCD module patents Fujitsu Display Transfer of MVA Premium Oct, 2002-Oct, 2005 Technologies Corporation LCD technology Fujitsu Display Authorization of MVA From Mar, 2003 Technologies Corporation Premium LCD patents and technology Fujitsu Display cooperative R&D of MVA From Mar, 2003 Technologies Corporation Premium LCD patents and technology Fujitsu Display Acquisition of FDTC’s From Mar, 2003 Technologies Corporation stock
Semiconductor Energy Laboratory Co., Ltd. Hitachi Display IBM Sharp Samsung Electronics
Sharp
Mutual authorization of TFT-LCD patents Mutual authorization of TFT-LCD patents Authorization of about 170 TFT-LCD patents Authorization of PC-use LCD patents Mutual authorization of TFT-LCD and OLED patents Mutual authorization of TFT-LCD patents
Sep, 2003-Dec, 2008 Nov, 2004-Jun, 2009 From Jul, 2005 Jul, 2005-Jun, 2010 From Jan, 2006
na
Source: Compiled from AUO’s annual reports
Based on technology acquisition, AUO also tried to develop their own R&D capability. In the interviewed managers’ view, patent is not really an issue due to the structural feature of TFT-LCD panel. AUO can reverse engineer the product and then make some incremental changes and finally claim it as their own patent. However, this approach also led to a lot of IP infringement issues. AUO has been accused by Japanese firms such as Sharp and Semiconductor Energy Laboratory but surprisingly it does not affect the relationship with the IP transfer. According to our interview, this is because they had worked together as partners for such a long time. It is difficult and uneconomical for each other to find a substitute. This created certain barriers for new entrants as well because they will find it hard or even impossible to find someone to sell them the technologies. One of the most distinct features of TFT-LCD panel technology is the continuous and rapid evolution of new generations. The first generation (G1) of panel was launched in 1990 and G2, G3, G4 were put into mass production with intervals from three to five years. The upgrade is accelerated in 21th century and new generations are launched in almost every year. However, the major difference between each generation of plants is the size of glass substrate, whereas the technology and process stay quite similar. In other words, there are no revolutionary technology changes and most of the progress are the incremental improvements on the production process. The learning curve effect is significant enough for manufacturers to improve productivity step by step. At the start, AUO didn’t have experiences in managing the process. Due to the lack of relevant experience and capability, the design of the factory was not ideal and the productivity was low. It took AUO a long time to integrate the components and equipments bought from various suppliers to achieve smooth production. However, since AUO had limited R&D risks by just simply copying mature technology which had already tested by Japanese leaders and had stable demand as well, it’s possible for AUO to concentrate on improving production to gradually offset the initial disadvantage on process knowledge. AUO employed specific tactics to facilitate this
improvement. While other firms such as Sharp and Samsung built single plant with large scale output, AUO successively built two G5 plants with less output. Based on the accumulated experience in building and running the first plant, the next one can be rapidly constructed and achieve higher yield rate. Therefore due to the learning curve effect and subtle planning, AUO are getting better in terms of mastering the production process of the panels. AUO and other Taiwanese panel manufacturers had another latecomer advantage due to learning curve effect. Since the manufacturing equipment for TFT-LCD panel is small batch and customized product, by installing equipment for first-movers the equipment makers also accumulated experience and were able to offer better products and tutorship of operation for latecomers. This effect was utilized by Taiwanese firms to enhance their manufacturing competitiveness. In order to understand more about AUO’s situation and corresponding strategies after entering TFT-LCD industry, it is necessary to delve into the structure of panel (Figure 2). Raw materials or key components take the major portion on total cost in TFT-LCD panel production, accounting for about 60% - 70% (This percentage varies due to different size, generation and yield rate etc.) Among all materials and components, glass substrate, polarizer, color filter, back light unit and driver IC are the most costly ones, taking up about 75% - 80% of total materials cost (Figure 3). In addition, most of these key components are ‘design-in’ products which must match the panel manufacturers’ specific needs and specifications from the beginning of TFTLCD design. Production of LCD panels is impossible to complete if any component is missing or unfit. Hence, holding the key components is doubtless vital to gain competitive advantages in this industry.
Source: DisplaySearch (2006)
Figure 2: Components of TFT Panels
Source: DisplaySearch (2006)
Figure 3: The cost structure of 32 inch TFT-LCD panels However, Taiwanese TFT-LCD firms lacked the capability on the key components in the initial stage due to the late entry. Most key components were monopolized by foreign firms, mainly Japanese. For instance, the color filter is the one of the most costly components of a panel, accounting about 15% of total cost. Two Japanese firms Toppan and DNP held the largest market share, accounting for 34% and 29% in fourth quarter of 2001. Considering polarizer representing approximately 9% of total cost, Three Japanese suppliers Nitto Denko, Sanritz and Sumitomo capture dominated the market with 47%, 20% and 15% in 2001. In addition, the key components of polarizer such as TAC and PVA film which require in-depth knowledge and experience in optics were mastered by Japanese leaders including Fujifilm and Konica. Before the second half of 1999, 100% of the polarizers that Taiwanese panel firms used were imported. For AUO, the interviewees admitted that 90% of their materials were bought from foreign suppliers in the early stage so its profit was quite limited. Furthermore, given the cyclical and fluctuant trajectory of the TFT-LCD industry, AUO and other Taiwanese panel markers could not hold stable component supply and thereby were at a disadvantage in the fierce competition while Japanese competitors could preempt materials from domestic suppliers due to long-term business relationships or ‘Keiretsu’ affiliations. In order to surmount this disadvantage, AUO employed aggressive vertical integration strategy. AUO’s integration modes varies based on the features of components and market dynamics. For example, color filters have a large portion on total cost and high interdependence with panels. In addition, the cost and risk of transporting color filters for new generation panels surges due to the larger size. Therefore AUO internalized the production of color filter. The IC industry in Taiwan is so mature that cooperation and network governance is more economical than internalization. Hence AUO just relies on the affiliated IC vendor and invested on several design houses focusing on small-and medium-sized panel segment. This vertical integration strategy is also employed by CMO and other Taiwanese panel makers.
Table 6: Taiwanese panel manufacturers’ vertical integration on key components Companies Color Driver Polariz Back light CCFL LED filter IC er AUO In-house Novatek Daxon In-house Wellypower Wellypower (group) (group) (group) (group) Darwin Raydium (group) & Orise (invest) CMO In-house Himax CMT In-house GIO In-house (group) (group) CLT(group) (group) Forepi (invest) CPT In-house Forward Sintronic (group) (invest) Hannstar In-house Cheertek (group) Innolux In-house In-house Source: Compiled from various sources
Up to now, AUO’s vertical integration strategy is quite effective. The Darwin, affiliated with AUO, can produce back light unit; the Wellypower, another affiliated firm, can provide CCFL and LED which are key components of back light unit; the Daxon which was invested by AUO, can produce polarized panel; the Novatek, also affiliated with AUO, is the largest TFT-LCD Driver IC manufacturer in the world. Moreover, AUO can self-provide color filters. On the whole, the capability of key component production of Taiwanese panel manufacturers had been notably enhanced within a short period.
Source: IEK (2004)
Figure 4: The self-sufficiency rate of key components of Taiwanese TFT-LCD panel manufacturers
Source Market
Table 7: Latecomer (dis)advantages and strategies of Taiwanese TFT-LCD panel manufacturers Latecomer Latecomer advantages Strategies disadvantages Not significant Existing customer Pure-play foundry;
Competition Not significant
Aid from Japanese leaders
Price orientation Close cooperation with Japanese
Differentiation advantage A full range of panel sizes
Technology & Product
Lack of experience in managing the process
Available external technology source
Dependence on external components
Incremental improvement rather than radical innovation
Patents do not matter
Differentiated performance on specific sizes Simply copy mature technolog Concentration on improving production Two successive smaller plants instead of single large one
Improved equipment Vertical integration with various modes
4. Discussion The overall findings from this case study are summarized in Table 7. Our findings from this case study resonate with some previous arguments but also reveal some different crucial issues which have not been emphasized in the research area of latecomer (dis)advantages. Some (dis)advantages addressed in previous research do not actually exist in this case. For instance, changes in consumer tastes which are considered as an important source of latecomer advantage do happen in the TFT-LCD sector. However the main change is simply the panel size expansion in every application and this trend is so apparent that everyone can see it and cannot provide particular opportunity for latecomers. The latecomer advantage stemming from learning pioneers’ mistake is also not significant because they didn’t make notable mistake in deed. Similarly, switching
cost and reputation which are argued to be latecomer disadvantages by previous literature do not matter in the TFT-LCD industry. Other (dis)advantages mentioned before such as avoiding consumer education are found in this context but they don’t appear to be crucial ones. What we found as the vital source of advantages for latecomers and want to underline in this research is the competition dynamics. Taiwanese panel makers’ entry is not only the result of their own strategic choice but also facilitated by the battle between Japanese and Korean pioneers. Considering the industry downturn and challenge from aggressive Korean competitors, Japanese firms had no other choice but to outsource production to reduce cost. Taiwanese firms are the ideal partner with competitive manufacturing capability, especially on cost down. Hence it was very reasonable for Japanese to ally with these “potential but probably minor competitors” to defeat “major rivals”. For Taiwanese firms, the cooperation with Japanese caused by the international competition has broad and continuous benefits. For the initial stage, Japanese provided design and manufacturing technology which acted as the foundation for Taiwanese firms’ development. Without the technology transfer from Japanese caused by this situation, it’s almost impossible for Taiwanese firms to edge in the industry. In addition, orders from Japanese partners ensured Taiwanese firms a relatively stable demand. Hence, this advantage stemming from competition dynamics also significantly influenced both the situation of market demand and technology regime. From the longitudinal view, even though Taiwanese latecomers have already grown up today, they are still receiving technology because Japanese firms have taken a strategy to avoid head-to-head scale competition and concentrate their resources on developing high value displays for new markets. Based on this experience from the case, we advise other latecomers to pay close attention to the competition between pioneers, understand their strategic intention and try to seize the opportunities caused by their choice such as alliance or retreat. One of the major source of latecomer (dis)advantages emphasized in this paper is the technological regime. Besides the accessibility of external technology source, the features of technical advances seem crucial to explain the successful catch-up of Taiwanese panel firms. The technical advances of TFT-LCD panels are mainly incremental improvements rather than radical innovations and therefore provide latecomers enough time to accumulate experience. However, given the numerous cases of successful catch-up during technological discontinuities whether the features of technological regime can be regarded as sources for latecomer advantage should be analyzed by linking demand and competition dynamics. In our opinion, the strategies to utilize advantages or to surmount disadvantages perhaps are more important than (dis)advantages themselves in explaining latecomers’ success or failure. From this case study, we found that Taiwanese firms also adopted specific strategies or approaches besides the acceptance of foreign technology. Some of these strategies have been noticed in previous literature. For example, unlike foreign competitors that focusing on large-sized sector AUO employed differentiated positioning by providing both small- and medium-sized and large-sized panels. And AUO also tried to achieve better performance on particular attributes in their
large-sized products. We also found other subtle approaches such as building two comparatively smaller fabrication plants successively to accelerate learning curve effects, which is also adopted by other Taiwanese electronics manufacturers including Foxconn. What is more impressive is AUO’s aggressive vertical integration driven by the desire of surmounting competitors’ preemption on key components and continuous cost reduction. These strategies have been approved to be effective and may also be feasible in other industries with similar context. We also underline the importance of understanding the structure of product in exploring latecomers’ catch-up. As illustrated by above analysis, the great portion of key components in total cost of panel production explains the Taiwanese firms’ unfavorable situation at the beginning and their subsequent endeavor in vertical integration. Without such knowledge, discussions and understandings about latecomers’ disadvantage and strategies will probably remain shallow. Moreover, we would like to address the difficulty of classifying latecomer (dis)advantages. For instance, the free-rider effects can be found on market/consumers and competition aspects in Cho et al.(1998)’s framework, in terms of saving the cost of educating customers and learning pioneers’ mistake. And in the framework proposed by Kerin et al. (1992), some first-mover advantages such as the marketing cost asymmetry as an economic factor and consumption experience asymmetry as a behavioral factor seem highly correlated. In our case study, as we mentioned above the advantage of receiving external aid caused by competition dynamics also led to the availability of existing customer in the demand area and availability of external technology source in the area of technology regime. Therefore, we are inclined to believe that (dis)advantages affect interactively on latecomers rather than working separately. In the paper presented at the DRUID Summer Conference 2004, Mathews (2004) underlines the importance of timing in strategizing in TFT-LCD sector in that latecomers including Korean and Taiwanese successfully entered only during industry downturns. However, according to our interview with managers of leading Taiwanese panel manufactures, we find that the timing of their entry may not be subject to strategic choice, but purely because of the emerging of opportunities, especially the external technology source. In other words, they were helped by Japanese firms to enter this market in the downturn rather than choosing this “window of opportunity” intentionally.
5. Conclusion To conclude, this paper advances the existing literature on latecomer advantages and disadvantages by reviewing the spectacular rise of Taiwanese TFT-LCD panel manufacturers, and reveals the importance of some unexplored source of latecomer advantage such as the competition dynamics which is proved to be vital for Taiwanese firms to successfully penetrate into the market. We also have discovered new forms of latecomer advantages, for example the free-rider effect of improved equipment due to suppliers’ own learning curve. This paper also illustrates the necessity of adequate
exploration on technology and product features to understand firms’ strategic behaviors. Moreover, the subtle tactics found in this case study may enlighten other latecomers. However, our study leaves a number of questions unanswered which warrant more attention. In the first place, the theoretical arguments and experiences from this single case is not easily extendable to the other industries. Broader and further investigation including cross-case study and quantitative method may be helpful to discover the mechanism between latecomer advantage, strategy and performance. In the second place, the study of latecomer (dis)advantages and strategies should take the firm features into account to get a whole picture. For example, whether SMEs have specific latecomer (dis)advantages and strategies is worth more exploration. Last but not the least, the examination of interactions between latecomer (dis)advantages and more detailed analytical framework are needed.
Acknowledgements The authors gratefully acknowledge the support from the Joint Program of National Natural Science Foundation of China and the Royal Society of the United Kingdom (grant no. 70711130230) and the Integrated Knowledge Centre (IKC) Commercialisation Laboratory sponsored by EPSRC.
Reference Bain, J.S. 1956. Barriers to New Competition. Cambridge, MA: Harvard University Press. Berndt, E. R., Bui, L. Deiley, D. and Urban, G. L. 1995. Information marketing and pricing in the U.S. anti-ulcer drug market, American Economic Review, 85(2):100-105. Bower, J. and Christensen, C. 1995. Disruptive technologies: Catching the wave. Harvard Business Review, 73(1):43-53. Breschi, S., Malerba, F. and Orsenigo, L. 2000. Technological Regimes and Schumpeterian Patterns of Innovation. Economic Journal, Vol. 110, 388-410. Brezis, E. Krugman, P. R. and Tsiddon. D. 1993. Leapfrogging in international competition：A theory of cycles in national technological leadership. The American Economic Review,83(5): 1211-1219. Buaron, R. 1981, New-Game Strategies, McKinsey Quarterly (Spring): 24-40. Carpenter, G. S. and Nakamoto, K..1994. Reflections on “consumer preference formation and pioneering advantage, Journal of Marketing Research, 31: 570-573.
Chang, S.C. 2005. The TFT–LCD industry in Taiwan: competitive advantages and future developments，Technology in Society, 27: 199-215 Cho, D.S., Kim, D.J., and Rhee, D.K., Latecomer strategies: Evidence from the Semiconductor Industry in Japan and Korea, Organization Science, 9(July-August): 489-505 DisplaySearch. 2006. TFT LCD and Component Market Trend, www.displaysearch.com Eisenhardt, K.M. and Schoonhoven, C.B. 1990. Organizational growth: linking founding team, strategy, environment, and growth among US semiconductor ventures, 1978–1988, Administrative Science Quarterly, 35: 504-529. Fuller, D.A., Akinwande, A.B. and Sodini, C. 2003. Leading, Following or Cooked Goose? Innovation Successes and Failures in Taiwan's Electronics Industry, Industry & Innovation, Volume 10, Issue 2 June, 179-196 Gao, X.D, Liu, J.X., Chai, K.H., Li, J.Z. 2007. Overcoming 'latecomer disadvantages' in small and medium-sized firms evidence from China, Int. J. Technology and Globalisation, Vol. 3, No. 4, 364-383. Gruber, H. 1996. Trade policy and learning by doing: The case of semiconductors , Research policy, 25:723-739. Hauser, J. R. and Shugan, S.M. 1983. Defensive Marketing Strategies, Marketing Science, 29 (Fall): 319-60. Henderson, R. 1993. Underinvestment and incompetence as responses to radical innovation: Evidence from the photolithographic alignment equipment industry, Rand Journal of Economics, 24(2), 248–270. Henderson, R.M. and Clark, K.B. 1990. Architectural innovation: the reconfiguration of existing product technologies and the failure of established firms, Administrative Science Quarterly, 35: 9-30. Hobday, M. 1995. Innovation in East Asia, The Challenge to Japan, Edward Elgar. Hsing, Y.L. 2006. Being Faster Movers: New Technology Strategy and Business Innovation of Taiwanese TFT-LCD Makers, IAMOT 2006. Hung, S.W. 2006. Competitive strategies for Taiwan’s thin film transistor-liquid crystal display (TFT-LCD) industry, Technology in Society, 28: 349-361. IEK (Industrial Economics and Knowledge Center). 2004. http://int.iek.itri.org.tw Kardes, F., Kalyanaram, G. Chandrashekaran , M.and Dornoff, R. 1993. Brand retrieval, consideration set composition, consumer choice, and the pioneering advantage, Journal of Consumer Research,20(1): 62-75. Katz, M. and Shapiro, C. 1994. System competition and network effects’, Journal of Economic Perspectives, 8(2): 93-115.
Kerin, R.A. Varadarajan, P. R. and Peterson, R.A. 1992. First-Mover Advantage: A Synthesis, Conceptual Framework, and Research Propositions , Journal of Marketing, Vol. 56, No. 4 (Oct.): 33-52 Lee, K. 2005. Making a Technological Catch-up: Barriers and Opportunities, Asian Journal of Technology Innovation, 13, 2, 97-131 Lee, K. and Lim, C. 2001. Technological Regimes, Catching-up and Leapfrogging: the Findings from the Korean Industries, Research Policy, 30:459-483. Leonard-Barton, D. 1992. Core Capability & Core Rigidity: A Paradox in Managing New Product Development. Strategic Management Journal, 13:111-125. Lieberman, M.B. 1987. The Learning Curve, Diffusion, and Competitive Strategy, Strategic Management Journal, Vol. 8, No. 5 (Sep. - Oct., 1987): 441-452. Lieberman, M.B. and Montgomery, D.B. 1988, First-Mover Advantages, Strategic Management Journal, 9 (Summer): 41-58. Lieberman, M.B. and Montgomery, D.B. 1998. First-mover (dis)advantages: retrospective and link with the resource-based view, Strategic Management Journal, 19: 1111-1125 Malerba, F. and Orsenigo, L. 1996. Schumpeterian patterns of innovation are technology-specific, Research Policy, Volume 25, Issue 3, May, 451-478. Mansfield, E., Schwartz, M. and Wagner, S. 1981. Imitation Costs and Patents: An Empirical Study, Economic Journal, 9 1 (December), 907- 18. Mathews, J.A. 2002. Competitive Advantages of the Latecomer Firm: A Resource-Based Account of Industrial Catch-Up Strategies, Asia Pacific Journal of Management, 19: 467-488 Mathews, J.A. 2004. Industrial Dynamics and Corporate Strategy: the Case of the Flat Panel Display Industry, the DRUID Summer Conference 2004. Mathews, J.A. and Cho, D.S. 1999. Combinative Capabilities and Organizational Learning by Latecomer Firms: The Case of the Korean Semiconductor Industry, Journal of World Business, 34(2), 139-156. Mathews, J.A. and Cho, D.S. 2000. Tiger Technology: The Creation of a Semiconductor Industry in East Asia. Cambridge, UK: Cambridge University Press. MIC (Market Intelligence Center). 2004. http://mic.iii.org.tw/intelligence Muthukrishnan, A. V. 1995. Decision ambiguity and incumbent brand advantage, Journal of Consumer Research, 22: 98-108. Nelson, R.R. and Winter, S.G. 1982. An Evolutionary Theory of Economic Change, Cambridge, MA: Harvard University Press Porter, M. J. 1980. Competitive Strategy. New York: The Free Press.
Schnaars, S. P. 1994. Managing Imitation Strategies: How Later Entrants Seize Markets from Pioneers. Free Press, New York. Schramm, W. 1971. Notes on Case Studies of Instructional Media Projects. Stinchcombe, A.L. 1965. ‘Social structure and organizations’, in James, G.M. (Ed.): Handbook of Organizations, Chicago, Rand McNally, 142–193. Tushman, M.L. and Anderson, P. 1986. Technological Discontinuities and Organizational Environments , Administrative Science Quarterly, Vol. 31, No. 3 (Sep), 439-465. World Bank. 1993. The East Asian Miracle: Economic Growth and Public Policy, New York: Oxford Univ. Press. Zhang, S. andMarkman, A. B. 1998. Overcoming the early entrant advantage: The role of alignable and nonalignable differences, Journal of Marketing Research, 35(4): 413–426.
