Regional development and sources of superior
performance across textile and IT sectors in Taiwan
YI-MIN CHENy and FENG-JYH LINz
yDepartment of Asia-Pacific Industrial and Business Management, National University of Kaohsiung, 700 Kaohsiung University Road, Nan-Tzu District, Kaohsiung 811, Taiwan; email: ymchen@nuk.edu.tw
zDepartment of Business Administration, Feng Chia University, Taiwan
The rapid growth and industrialization of Taiwan’s textile and IT sectors, mainly comprised of small and medium-sized enterprises, has prompted an array of explanations among academics, including neoliberalism, structural-institutionalism, flying geese patterns, regional networks and economic geography. Drawing on neoliberal, structural-institutional, regional networking and economic geographic views in that strong Taiwanese entrepreneurial culture is important to its textile and IT sector development, this study shares their positive perspectives in influencing the sources of profitability differentials among Taiwan’s textile and IT firms in international competitiveness. Researchers investigating the sources of performance differences among firms have focused mainly on the relative importance of industry and firm factors. Specifically, this study employs Taiwan’s business database to examine industry and firm effects on profitability differentials in these sectors using return on assets and the economic performance measures of economic value added and market value added. A variance components model is proposed, and findings indicate that firm effects dominate performance while industry effects have little impact. Our discussion reconciles results with those of previous studies.
Keywords: industry organization; textile sectors; economic performance; IT sectors;
resource-based view; Taiwan.
1. Introduction
Understanding the determinants of firm profitability is a key theoretical and empirical issue in the fields of industrial organization, economics and strategic management. Researchers investigating the sources of performance differences among firms have most often focused on the relative importance of industry and firm factors. Since the pioneering work by Schmalensee (1985), two streams of studies have come to dominate the literature. The first builds on Schmalensee (1985), such as in Montgomery and Porter (1991) and Wernerfelt and Montgomery (1988), and supports industry factors playing a central role in determining firm profitability while firm effects are insignificant. The second major area of work argues that other factors, such as firm or business unit effects, are the primary sources of performance differences while industry factors account for small or negligible effects. Motivated by Rumelt’s (1991) study, Powell (1996), Roquebert et al. (1996), McGahan and Porter (1997, 2002), Mauri and Michaels (1998), Brush et al. (1999) and Hawawini et al. (2003), whether measuring competitive advantage effects through business unit or business segment effects, found evidence of the dominance of firm-specific effects.
Entrepreneurship and Regional DevelopmentISSN 0898–5626 print/ISSN 1464–5114 online ß 2006 Taylor & Francis http://www.tandf.co.uk/journals
However, despite the advances and new insights concerning this debate, the generality of results has been limited by the sparsity of data and the narrowness of conventional performance measures (McGahan and Porter 2002). For example, Schmalensee (1985) and Rumelt (1991) used US Federal Trade Commission (FTC) data to measure how much industry, corporate and business-level factors affect firm performance, while further empirical studies along the lines of Rumelt’s (1991) work used the CompustatÕ database. Unfortunately, since the FTC and CompustatÕ databases include mostly US firms, previous research has neglected firms outside the USA in explaining the sources of performance differences among firms with respect to industry and firm effects. Additionally, in order to support deeper understanding of the processes that generate favourable and unfavourable effects at the industry, corporate parent and business-specific levels, McGahan and Porter (2002) suggested that detailed studies at the sectoral level are needed. This study assesses the relative importance of industry and firm factors on profitability differences among the firms byu employing Taiwan’s business database of the textile and IT sectors.
The question of whether firms’ performance is driven primarily by industry or firmu factors in international competitiveness is highly salient at a time when the globalization of markets has raised questions about the appropriateness of domestic industrialization policy and other forms of regional development policy towards competition. Taiwan is a particularly interesting setting in which to explore this question because it is frequently argued that the economic growth of Taiwan is simply the product of a unique ‘industrial ecosystem’ that brings together knowledge and technical and financial resources (Mathews 1997). Many Western observers, for example, have asserted that Taiwanese textile and IT industries are internationally competitive because of cooperation among Taiwanese rivals, sheltering from international competition, and selective government intervention in competition orchestrated by the Industrial Technology Research Institute (ITRI) and other governmental agencies (Kraemer et al. 1996, Mathews 1997, Thorbecke et al. 2002). Related to this, we will examine mainstream perspectives on Taiwan’s textile and IT sector development in the next section. Drawing on aspects of neoliberalism, structural-institutionalism, regional networks, and economic geography, this study shares their positive view of strong Taiwanese entrepreneurial culture to promote industrialization of the textile and IT industries. Thus a persistent theme of this entrepreneurial culture perspective in explaining the sources of performance differences among Taiwan’s textile and IT firms in international competitiveness has been on the role of benign regional industrialization policy (industry effects) and unique organizational processes (firm effects).
In addition to the traditional approach which relies on raw accounting values of returns on assets (ROA) as the performance measure, gauges of value-based management have been pioneered and adopted by Hawawini et al. (2003) to decompose the variance in firm profitability into components associated with industry and firm effects. The adoption of value-based measures, such as the consultancy Stern Stewart & Co.’s economic value added (EVA), has coincided with increasing pressure from capital markets and corporate control markets for managers to focus their strategies on value creation, i.e. economic performance (Haspeslagh et al. 2001, Hawawini et al. 2003). Value creation occurs only when firms earn returns greater than the cost of capital, which implies that value creation is a reasonable proxy for economic performance.
The objective of this study is to investigate the extent to which industry and firm factors produce profitability differences among Taiwan’s textile and IT firms using the alternative measures of performance EVA and market value added (MVA) along with ROA. EVA, indicating economic profit, reflects operating performance in a given year, while MVA, indicating market-to-book value, reflects the market’s expectations of the firm’s future operating performance (Hawawini et al. 2003). The difference between ROA and EVA (or MVA) is that examining what drives ROA is not equivalent to examining what drives value creation or economic performance, as Hawawini et al. (2003: 1) argued that ‘raw accounting values of measures such as ROA account neither for the cost of capital nor for the accounting policies that may distort the true value of the underlying measures, for instance, the value of assets’.
This paper is organised as follows. After a brief review of Taiwan’s textile and IT sector development, the next section considers the debates in economic development based on the determinants of profitability differences among firms. Previous research examining the effects of industry and firm factors largely followed the descriptive model used in Schmalensee (1985), Rumelt (1991) and McGahan and Porter (1997); section 3 provides a review of this literature. Section 4 proposes a similar qualitative model and discusses adjustments necessary to account for both capital costs and accounting conventions when calculating the alternative performance measures. Following the statistical methodology most commonly used by previous studies, this work employs the variance components procedure to examine sources of performance differences among textile and IT firms in Taiwan. The variance components approach and the data sample used are discussed in section 5 and the results are discussed in section 6. Firm effects are shown to be the primary source of performance differences among textile and IT firms in Taiwan for both the traditional and alternative performance measures. Finally, this study concludes with a discussion of the results and offers final remarks.
2. Explaining development in Taiwan’s textile and IT industries Taiwan’s firms emerged as formidable global market players by successfully transforming themselves from producing mainly low-value, labour-intensive goods to producing many high-technology products that require significant business savvy and expertise (Song et al. 1997). Taiwan was one of the first developing countries to open up to international economic flows, first targeting export markets and then relying on direct investments from foreign multinationals, and the textile and clothing industry played a central role in Taiwan’s economic development. After an early phase of specialization in the labour-intensive textile industry, Taiwan experienced a remarkable structural transformation and shifted emphasis towards electronics and electrical machinery in the 1980s. During the 1990s, Taiwan achieved enormous success in the electronics industry and especially in the IT sector. For example, Taiwan Semiconductor Manufacturing Corporation (TSMC) used its superior manufacturing capabilities and production process technologies to establish the integrated circuits foundry model, breaking away from the traditional vertically-integrated modes of production in the semiconductor industry, previously mono-polized by the USA, Japan and South Korea, to establish Taiwan as one of the world’s leading semiconductor producers. Although Taiwan’s IT sectors have become very competitive and have demonstrated outstanding performance since the 1990s,
Taiwan’s textile industry was one of the most important business sectors beginning in the 1960s and continuing into the 1980s. Even in 1995 when the textile industry’s value-added performance fell behind the IT industry, it still employed a labour force numbering between 300 000 and 350 000 and exported products valued at US$15.6 billion (Young 1996).
2.1 Neoliberalism and structural-institutionalism
Taiwan’s rapid growth and industrialization elicited a wide range of explanations among academics. First, supporters of neoliberalism or the free market approach, the most popular perspective, argued that free-market free trade policies were responsible for the region’s rapid growth and industrial transformation. Proponents have therefore advocated that other countries adopt similar policies. According to this model, industrialization is a matter of ‘getting the prices right’ and specializing (Amsden 1991). Through the 1950s, Taiwan’s textile industries were set up by selected entrepreneurs such as Y.-C. Wang, under the instruction of K.-Y. Yin, as well as technocrats like K.-T. Li with which he surrounded himself (Wade 1990). These forceful and strong-willed entrepreneurs are distinguished from the overseas Chinese business networks by shared professional as well as ethnic identities and by their deep integration into the free market dynamics. Gilder (1989) and Callon (1995) argued that national economic success in Taiwan’s IT industries is evidence of the dynamism of free markets and identified high levels of human capital formation, domestic entrepreneurship and market competition as the bases of the successful technology industries. Based on the strong entrepreneurial tradition and close relations with California business circles and educational systems, Dr Morris Chang, the founder of TSMC, and Mr Stan Shih, the founder of Acer which ranks among the world’s top five branded personal computer (PC) vendors, leveraged themselves through mastery of single-loop and double-loop organizational learning to succeed in competitive markets and achieve world-class IT capabilities (Mathews 1997).
From the perspective of neoliberalism, low-wage countries are supposed to develop by exporting labour-intensive products. In the 1960s, if wages alone mattered, Korean and Taiwanese textile industries might have been able to compete with the Japanese textile sector. Even when Korea and Taiwan liberalized their economies and satisfied the exchange rate demands of Bretton Woods institutions, the superior infrastructure, production equipment and management of Japanese textile companies made it impossible for Korean and Taiwanese textile companies to compete with them exclusively on the basis of lower wages (Amsden 1991). Confronting the difficult situation that results when lower wages are insufficient to compete in labour-intensive industries against the higher productivity of a higher-wage country even with financial assistance and modern infrastructure, the governments of South Korea and Taiwan had to intervene to offset Japan’s higher productivity with a wide range of subsidies and by introducing either inward foreign direct investment (FDI) from more technologically advanced countries or further exchange rate devaluations.
Unlike the industrialization of Japan and Korea, in which large firms led export-oriented growth, Taiwan’s development was mainly led by small and medium-sized enterprises (SMEs), often involved in direct contract relations with US and European manufacturers. Hamilton and Biggart (1988) explained that the organizational patterns of Asian business – the particular network configurations of firms and not
the policy choices of Asian states – were pivotal in promoting economic success and demonstrated important differences in economic organization, particularly in the export sectors, among Japan, South Korea and Taiwan. They indicated that very large business groups in South Korea (the chaebol) dominate the export sector, whereas this sector is dominated by SMEs in Taiwan. Hamilton and Biggart (1988) then argued that the South Korean state actively sponsored the extraordinary growth of selected business groups in selected industries by careful planning and implementing strategic policy measures and that the spread of SMEs throughout Taiwan’s export sector could not be explained by these actions of the government of Taiwan alone. Instead, widespread entrepreneurship in Taiwan was better explained by social institutions common to Chinese society that flourished in the absence of state intervention, this state-business relationship being characterized as a ‘strong society’ in which there emerged a ‘separation of spheres’ between public and private sectors of the economy. Thus Hamilton and Biggart (1988) proposed three models of state/ business relationships in South Korea, Taiwan and Japan: South Korea’s ‘strong state model’ in which the state intervenes heavily in the economy, Taiwan’s ‘strong society model’ in which the state refrains from such action and Japan’s ‘strong intermediate power model’ in which the state promotes public-private cooperation. However, the government of Taiwan in the ‘strong society model’ is by no means weak, but rather it is precisely the kind of state that promotes free enterprise in the export sector and thereby created the conditions for Taiwan’s rapid economic development (Amsden 1985, Gold 1986).
The importance of the role of the state to successful industrialization leads to the emergence of structural-institutionalism, another approach used to explain Taiwan’s economic growth. The structural-institutionalist school, a smaller but increasingly influential group, argued that Taiwan’s economic development was the result of effective state direction of economic activity (Wade 1990, Amsden 1991, Rodrik 1995). After the initial period of industrialization, industry-targeting policies were used to advance specific industries by launching public firms as pioneers. Under import substitution policy, exchange control, import licensing, control over the establishment of plants and land use as well as industry targeting by Taiwan’s government, vibrant textile SMEs helped Taiwan’s exports to achieve vigorous growth. As a consequence, the textile industry played a leading role in early development of Taiwan’s economy since the 1960s.
This industrial policy not only strategically targeted the textile industry in the 1960s but also the semiconductor and IT industries in the 1980s. For example, both TSMC and United Microelectronics Corporation were founded with the partial support of public funds. Under the perspective of structural-institutionalism, the intervention of government agencies such as Taiwan’s ITRI, a government-sponsored outfit in Taiwan, explain the dynamism of the new industries such as the semiconductor and IT industries (Wade 1990, Kraemer et al. 1996, Mathews 1997, Thorbecke et al. 2002). For example, the Hsinchu Science-based Industrial Park (HSIP), which was modelled quite explicitly on California’s Silicon Valley, is entirely the creation of Taiwan’s government to offer IT firms that settle there attractive terms for setting up business as well as a range of taxation benefits and allowances, including low-interest loans, R&D matching funds, tax benefits (e.g. investment allowances) and special exemptions from tariffs and commodity and business taxes. Bringing together semiconductor firms, venture capital organizations, university-private sector partnerships and interactions with other components of the industrial ecosystem raised
Taiwan to the world’s fifth largest semiconductor producer in the mid-1990s, even ahead of industrial giants such as France and the UK. However, the problem of government intervention is how to insure that after an industry is subsidized for possibly lengthy periods, it achieves world-competitive levels of productivity and quality (Amsden 1991). This argument leads some economists to doubt the necessary role of industrial policy and the causal role of the outward-oriented reform in Taiwan’s early industrialization that started in the late 1950s (Rodrik 1995, Thorbecke et al. 2002).
2.2 Flying geese patterns
Despite disagreements over the role of state intervention, both the neoliberal and structural-institutional explanations of Taiwan’s industrial development demonstrate the benefits of export-led growth following the earlier experiences of Japan. Drawing on these aspects of neoliberalism and structural-institutionalism, the recent ‘flying geese’ theory of development, the third approach to explain Taiwan’s phenomenal economic growth, shares their positive view of capitalism’s power to promote industrialization in crediting regional economic dynamics, in particular those shaped by Japanese FDI, for the regional advance (UNCTAD 1993). The ‘flying geese’ approach of sequential economic development in a group of interactive economies led by a lead-goose nation explains the process as follows: competition and state industrial policies lead firms in advanced countries to shift to higher-value-added industries and, through capital mobility, help countries at lower stages of industrialization to develop their own industries (Hart-Landsberg and Burkett 1998). Japan helped South Korea and Taiwan, and all three countries enable the members of the Association of Southeast Asian Nations (ASEAN-3: Indonesia, Malaysia, and Thailand) and China to follow Japan and the newly industrialized economies (NIEs: South Korea, Taiwan, Hong Kong, and Singapore) in a sequentially structured take-off into sustained growth and development.
Without question, Taiwan’s textile and IT industries have reaped positive (and some negative) externalities such as technology transfers from Japan. Nevertheless, the ‘flying geese’ metaphor raises more questions than it answers (Amsden 1991, Hart-Landsberg and Burkett 1998, Buckley and Ghauri 2004). The major weaknesses in the ‘flying geese’ metaphor are that it inaccurately describes the way geese fly (different geese assume the leadership for different periods), it ignores vertical linkages across and between Taiwan and the rest of the world and it overplays the benign effects of leadership and underplays power relationships (Bernard and Ravenhill 1995, Hatch and Yamamura 1996, Hill and Fujita 1996, Hart-Landsberg and Burkett 1998, Tsui-Auch 1999, Edgington and Hayet 2000).
2.3 Regional networks and economic geography
Regional integration in the spatial development of the world economy and regional networks in Asia, the fourth key explanation for East Asian economic development, are important both theoretically (Markusen and Venables 2000) and empirically (see Yeung (2001) on Singapore’s firms in Southeast Asia, Ghauri and Prasad (1995) on Asian networks and Saxenian and Hsu (2001) on the significance of international
communities in transferring technology and organizing production between Silicon Valley and HSIP). Analyses of global production networks represent an important conceptual advance because these multinational networks demonstrate a powerful mechanism for industrial advances in remote locations such as Taiwan. For example, the success of Taiwan’s PC producers derives from their role as original equipment manufactures (OEM) for the lead creation, technology transfer and improved domestic capabilities (Borrus 1997, Dedrick and Kraemer 1998, Ernst 2000).
Recently, analysts have moved beyond the simple state-market debate to examine other determinants of economic performance such as the geography of production, the fifth perspective to explain the success of East Asian industrialization. Economic geography has a long history (Krugman 1991, 2000, Clark et al. 2000), is currently enjoying a renaissance (Scott 2000) and emphasizes endogenous agglomeration forces generated by the interaction of increasing returns and transport costs. Support for research developments in this area has been fostered by the recent trend towards an increasingly borderless world economy, including the integration of national economies within new trading blocks such as the European Union. As market integration dissolves economic barriers between nations, national boundaries no longer provide the most natural unit of analysis. Krugman (1991) provided the Hsinchu region in Taiwan as an exemplar of Marshallian external economies, in which the localization of skill, specialized materials and inputs and technological know-how generate cost reductions for individual firms and increasing returns to the region as a whole, a viewpoint which is in direct contrast with the neoclassical international trade theory based on constant returns and zero transport costs. In fact, the theoretical focus of economic geography, i.e. the cumulative process of industrial clustering, represents a revival of modelling increasing returns in spatial contexts advocated by early industrial organization and dynamic development theorists. Porter (1990, 2000) combined economics and geography and viewed Silicon Valley and Taiwan’s Hsinchu as industrial clusters in which competition and vertical co-operation among local firms account for rising productivity, innovation and new-firm formation to analyse geographical economic development.
Additionally, both Silicon Valley and Hsinchu boast high rates of entrepreneur-ship and hundreds of SMEs alongside larger technology companies with multiple backward and forward linkages. Callon (1995) and Ernst (2000) stressed that the performance of producers in these regions depends on their flexibility, speed and innovative capacity relative to their leading competitors. The most convincing accounts document how the decentralization of the industrial clusters of Silicon Valley and Hsinchu ensures the flexibility and innovative capacity needed to compete in a volatile market environment (Saxenian 1994, Hsu 1997). However, the concept of external economies overlooks the contributions of technological innovation to regional growth. For example, economic geography cannot explain why Taiwan outperformed Singapore in the IT industry in the 1990s when both economies were destinations for electronics FDI in the 1960s and 1970s or how the emergence of successful regions such as Taiwan’s Hsinchu are located far from established centres of technology and skill. Evidently, no single approach of the above mainstream perspectives – free markets, the role of the state, flying geese theory, regional networks and economic geography – can satisfactorily explain East Asian economic development or Taiwan’s rapid industrialization. Drawing on neoliberal, structural-institutional, regional networking, and economic geographic views in that strong Taiwanese entrepreneurial culture is important to its textile and IT sector development, two kernels of truth
reside in the strong entrepreneurial culture story as applied to Taiwan. The first is that Taiwan’s textile and IT industrialization patterns are shaped by the successive waves of Taiwanese entrepreneurs. For example, since they are characterized by individual freedom, risk-taking, initiative, thrift, frugality, and hard work, and often sources of new ideas, materials, processes, and services that larger textile firms may be unable or reluctant to provide, it is well recognized that small entrepreneurial firms play an important role in Taiwan’s textile sector development (Chen 1999). The creation of a series of fast-moving innovative IT industries and a position of the world’s leading producers is widely astonished as the product of a thriving group of Taiwanese new home-grown firms with strong entrepreneurial tradition (Ferguson and Morris 1993). The second is that the managerial philosophies and practices of Taiwanese firms differ substantially from their Western counterparts. For example, in comparing the Western and Taiwanese styles of management, Tzeng (1985) argued that Westerners try to develop a theoretical and methodological model before implementing it, while Taiwanese extract the essential features of a given philosophy but remain flexible in their application of it. Chen (1995) noted that Western managers want to control all aspects of the working environment while Taiwanese managers tend to adapt to a specific environment.
Management researchers (Kelley et al. 1987, Ralston et al. 1997) advocated that both cultural upbringing and immediate environment play significant roles in shaping individual attitudes and behaviours from the management and organizational perspectives. In this context it is natural to ask from the entrepreneurial culture perspective, with respect to the importance of both external environment, such as the free market policy, industrial policy and the role of the state, and internal environment, such as networking and manufacturing capabilities, to the development of Taiwan’s firms, what accounts for the relative effects on performance differences among firms in textile and IT sectors?
The picture emerging from our analysis of whether Taiwanese firms’ performance is driven primarily by industry or firm factors is thus quite different from that presented by mainstream theorists of industrialization. It is our hope that this different understanding of Taiwan’s sector development will assist the realization of the potential how the cultural context of Taiwan may influence the conditions for firms and industries in different sectors to find common ground as they work to build an international competitive economy.
3. Identifying the determinants of firm performance
Researchers in industrial organization and strategic management have long been interested in understanding the determinants of firm profitability to address the questions of how and why certain private enterprise firms build competitive advantage in environments of rapid technological change (Teece et al. 1997). Mason, the father of industrial organization economics, argued in the late 1930s that there is a rather deterministic association between market structure and profitability (Mason 1939). The logic of the argument rests on the premise that structural characteristics of the industry or market place constraints on the conduct or strategies that firms can pursue (Roquebert et al. 1996). These constraints lead to different performances among firms in the same industry (Mason 1939). Therefore, the central argument was that the
structural characteristics of the industry are the primary determinants of performance (Porter 1980).
During the 1970s and 1980s, with respect to the unit of analysis and the relative focus across fields, there were major shifts in the industrial organization economics and strategic management fields. The main reason for the shifting theoretical orientation was the inability of industrial organization economics to explain intra-industry heterogeneity in firm profitability. In other words, while industrial organization economics, proposing the structure-conduct-performance framework, considers industry as the main unit of analysis that would render purely deterministic theories irreconcilable, strategic management focuses increasingly on the individual firm to explain intra-industry performance differentials.
Following the arguments of strategic management, the resource-based theory of the firm has ‘reemerged’, proposing that firm-specific idiosyncrasies in the accumulation of valuable, rare and specialized resources create sustained competitive advantages for those firms possessing them (Rumelt 1984, Wernerfelt 1984, Barney 1991, Collis 1991, Conner 1991). The ‘reemergence’ reflects that earlier research seeded the birth of the resource-based theory of the firm. In fact, in the late 1930s, Nourse and Drury (1938) suggested that firm-specific influences, such as management skills, basically determined firm advantages and performance. Firms were not simply seen as identical ‘black boxes’ in a given industrial structure but as dynamic collections of idiosyncratic attributes representing sources of competitiveness and relative performance. Since company strategies differ between firms within an industry, the bundle of idiosyncratic attributes that each firm possesses comes to differ (Nelson 1991).
Previous research does not suggest that the resource-based theory of the firm in strategic management should supplant the traditional and well-established industrial organization framework but rather that the former supplements the latter. In light of this argument, strategic management has recently tried to identify the relative impacts of industry and firm effects on firm profitability (Roquebert et al. 1996).
Since the pioneering work by Schmalensee (1985), two streams of studies have come dominate the literature. Unsatisfied with 80% of the variance in industry rates of returns on assets left unexplained and the use of only a single year’s data in Schmalensee (1985), Rumelt (1991) extended the study to cross-sectional and time series data using four years of data and including not only corporate effects – what Schmalensee (1985) called firm effects – but also business unit effects. Rumelt’s (1991) landmark study allowed for the inclusion of a composite term to measure business unit effects and attempted to clarify Schmalensee’s (1985) large degree of error. Rumelt (1991) concluded that industry effects explained around 9% of the variance in long-term performance and business unit effects accounted for more than 44% of the variance in business unit profits. Not surprisingly, the initial works of Schmalensee (1985) and Rumelt (1991) reignited the debate on the relative influences of industry, corporate and business unit factors as determinants of firm profitability for sectors of the US economy over the last decade.
With such robust support, it appears safe to conclude that industry structure does not matter as much for firm profitability in US firms. In other words, there may be little value in further empirical study seeking to determine the relative impact of industry and firm factors for these firms. However, Khanna and Rivkin (2001) showed that the relative importance of the effects on business group performance differed radically in emerging markets, and reliable and comparable data on the accounting profits of firms
in other parts of the world may yield significant insight into questions about the relationships between the national economic environment and industrial performance (McGahan and Porter 2002). As a consequence, the most direct opportunity for further research resides in exploring new data in settings outside the US. It is also necessary to resolve the question that Hawawini et al. (2003) asked: is ROA as used in past research a reliable performance indicator of economic value? McGahan and Porter (2002) also suggested the value in exploring additional measures of firm performance.
Since the resource-based view of the firm and industrial organization are probably the best candidates to explain firm performance across Taiwan’s textile and IT sectors, to explore the contribution of resources and capabilities of firms and structural characteristics of sectors on textile and IT firm performance, this study decomposes the relative impact of industry and firm effects on performance differentials in terms of the raw accounting measure ROA and the economic performance measures EVA and MVA for textile and IT firms in Taiwan.
4. The model and its operationalization 4.1 The model
Recognising that Taiwan’s firms are not large corporations and not substantially diversified, this study uses the term ‘firm’ to designate an autonomous competitive unit within an industry in contrast with the term ‘company’ or ‘corporation’ – a legal entity considered in previous research which owns and operates one or more business units (Rumelt 1991, McGahan and Porter 1997, 2002, Mauri and Michaels 1998, Brush et al. 1999). Thus the term ‘firm effects’ indicates both intra-industry dispersion among theoretical ‘firms’ and differences among ‘firms’ which are not explained by their patterns of industry activities. In this regard, this study uses the term ‘firm effects’ to include what other research refers to as ‘corporate effects’ and ‘business unit effects’. Working with this terminology, this study posits the following descriptive model of firm accounting or economic profit:
rijt¼...þiþjþtþ"ijt ð1Þ
where i ¼ 1, . . . , m indexes industries, j ¼ 1, . . . , ni indexes firms (ni is the number of
firms within industry i), t indexes years and rijtis the accounting or economic profit in
year t for firm j in industry i. The first right-hand-side term is . . ., which is the
average profit over the entire period for all firms (the three dots indicate averages over indices i, j and t). The next three terms i, jand trepresent the random industry,
firm and year effects, respectively, and the final term "ijt is the random error term.
This essentially descriptive model takes the industries and firms as given and is estimated using dummy variables to represent industry, firm and year effects. In particular, the model offers no causal or structural explanation for profitability differences across industries, firms or years – it simply posits the existence of differences in profitability associated with these categories.
The main effects (i, jand t) and error term ("ijt) are assumed to be normally
and independently distributed with means zero and variances 2, 2, 2and 2. These four sources of variation in business returns represent industry factors, firm effects,
the effect of yearly macroeconomic fluctuations and random error. The industry effect reflects the influence of structural characteristics of industries on the performance of firms. The firm effect includes both corporate and business unit effects and reflects the influence of firm-specific factors such as heterogeneity in tangible and intangible assets among firms due to differences in reputation, operational effectiveness, organizational effectiveness, organizational processes and managerial skills (Hawawini et al. 2003). The year effect captures the impact of broad economic trends.
Although the model in equation (1) of firm accounting or economic profit is similar to the descriptive models of Schmalensee (1985), Rumelt (1991), McGahan and Porter (1997, 2002) and Hawawini et al. (2003), there are two key differences between the current and previous descriptive models for the analysis of the effects of industry, firm and year factors. First, all the ‘interaction effect’ terms have been discarded because the model would be overspecified if we equally represented transient industry effects (the industry-year interaction) and transient firm effects (the firm-year interaction). Second, the data set of Taiwan’s textile and IT firms does not allow for distinctions between corporate effects and other firm-specific effects such as business-level effects.
4.2 The operationalization
There are three performance measures used in this study: the raw accounting value ROA and the value-based measures EVA and MVA. As McGahan and Porter (1997) and Hawawini et al. (2003) indicated, there are shortcomings to the accounting measures of profit. First, because accounting conventions such as Generally Accepted Accounting Principles (GAAP) exclude intangible assets from the balance sheet, accounting measures of assets may be understated for some firms. Second, the use of operating income (i.e. earnings before interest and taxes) divided by total assets (ROA) excludes the effects of differences in financing, such as the cost of capital. As a result of these shortcomings, ROA conveys neither the cost of capital nor adjusts for accounting policies that may distort the true values of the measure, such as asset values.
The above accounting measure of profit relies on two distinct financial statements under the accounting conventions – an income statement and balance sheet – whereas the economic model of value uses only one: sources and uses of cash. The economic model of value holds that share prices are determined by smart investors who care about just two things: the cash to be generated over the life of a business and the risk of the cash receipts (Stewart 1991). Therefore, based on the existence of different accounting policies and conventions, Harcourt (1965) and Fisher and McGowan (1983) argued strongly against the use of ROA as a proxy of economic profitability. Stewart (1991) introduced value-based economic measures of performance: EVA and MVA. EVA is the operating profits less the cost of all of the capital employed to produce those earnings; MVA is the market value less the book value of capital employed. The capital employed is the sum of the equity capital and debt capital. In fact, as Haspeslagh et al. (2001) suggested, the adoption of the EVA performance measure and the practice of value-based management have coincided with increasing pressure on managers to focus their strategies on economic performance.
Stern Stewart & Co. (1997) produced a list (Stern Stewart Performance 1000) of annual EVA and MVA estimates beginning in 1982, but its coverage is limited to the largest
1000 companies and dominated by US firms. Stewart (1991) emphasized that the EVA definition should not be uniformly applied to all firms. Due to shortcomings in conventional GAAP accounting, he identified a total of 164 performance measurement issues in estimating EVA methodology. There are so many adjustments in computing EVA that most firms that adopt EVA restrict the number of adjustments to fewer than 10 to make performance systems manageable (Hawawini et al. 2003). Additionally, Martin and Petty (2000) and Young and O’Byrne (2001) provided an overview of common adjustments that involve corrections for distortions caused by accounting policies that can understate the true level of invested capital (also referred to as a correction for successful efforts accounting) and for those caused by accounting for operating leases, mergers, goodwill, marketing expenses and R&D expenses. As a result, the accounting adjustments are a focal point of the EVA formulation.
To apply EVA and MVA and examine sources of profitability differentials among firms in other countries, we need a tractable and relatively accurate method similar to Stewart’s (1991) methodology to compute their values. Due to limitations of the Taiwan Economic Journal (TEJ) data set, the number of common accounting adjustments made to calculate EVA for Taiwan’s textile and IT firms is restricted to six as shown in table 1. These adjustments are R&D expenditure, marketing expenditure, deferred income tax reserve, construction in progress, bad debts and other reserves and marketable securities. All of these modifications are intended to reflect economic consequences better than more traditional accounting measures and to provide appropriate adjustments in the formulation of EVA for Taiwan’s firms (Wang and Liu 2001, Liang et al. 2002).
We take the R&D expenditure as an example to explain the necessity of adjusting the net operating profit after taxes (NOPAT) and invested capital employed (ICE). Under GAAP, accountants are required to record R&D outlays as if the potential R&D contribution to value is always exhausted in the period incurred. However, Scherer (1965) indicated that R&D activity has the characteristic of time delay. In addition, current and previous R&D expenditures can inflate the value of the firm (Chauvin and Hirschey 1993, Sougiannis 1994). Hatfield (2002) suggests that R&D
Table 1. Adjustments made to calculate Economic Value Added (EVA) for
Taiwan’s textile and IT firms.
Adjustments NOPAT ICE
R&D expenditure þR&Dt0.8 R&Dt10.2 þR&Dt0.8 þ R&Dt10.6
R&Dt20.2 R&Dt30.2 þR&Dt20.4 þ R&Dt30.2
R&Dt40.2
Marketing expenditure þMEt0.8 MEt10.2 þMEt0.8 þ MEt10.6
MEt20.2 MEt30.2 þMEt20.4 þ MEt30.2
MEt40.2
Deferred income tax reserve
Increase (decrease) in the liabilities of deferred income tax reserve
Liabilities of deferred income tax reserve
Decrease (increase) in the assets of deferred income tax reserve
Assets of deferred income tax reserve
Construction in progress No adjustment Construction in progress
Bad debts and other reserve Increase (decrease) in the
bad debts and other reserves
þBad debts and other reserves
Marketable securities No adjustment Marketable securities
investment continuously influences the economic profit of the firm for five years. Following this result, this study considers 5 years of R&D investment. As a consequence, EVA is the financial performance measure that comes closer than any other to capturing true economic profit, or the amount by which earnings exceed or fall short of the required minimum rate of return that shareholders and lenders could get by investing in other securities of comparable risk.
In theory, a company’s MVA at a point in time is equal to the discounted present value of all of the EVA it can be expected to generate in the future. Both EVA and MVA possess the following two features. First, they simultaneously reflect the concept of residual income, i.e. income that is adjusted for capital costs and hence risk as well as the time value of money. Second, both EVA and MVA are usually not bound by accounting conventions that tend to distort performance measures like ROA (Hawawini et al. 2003).
5. Data sample and statistical methodology 5.1 Data
To use EVA and MVA to measure economic performance with the above adjustments and to examine the sources of profitability differentials for Taiwan’s textile and IT firms, this study needs a reliable database since Stern Stewart & Co. does not provide data on these measures for non-US firms. The database used is the TEJ, which covers macroeconomic indicators and financial reports for 678 companies listed and traded on the Taiwan Stock Exchange (TAIEX).
The industrial classification system in Taiwan is different from that of the US, one feature of the TEJ is that it does not offer business-level data at the 4-digit SIC code level – a traditional taxonomy that classifies firms into particular industry groups – but provides firm-level data at a 3-digit level of industry classification. The TEJ data set suffers from a lack of specificity for industry classification, and this has two consequences for this work. First, since we assign firms according to their primary industry classifications, the analysis underestimates industry effects to the extent that the firm is diversified beyond its primary industry, even when there are only a few cases. Second, the firm effects are likely to reflect both corporate- and business-level effects.
5.2 Sample
The sample of Taiwan’s textile and IT sectors covers the 6-year period from 1998 to 2003. The sample was screened in various ways. This study excluded firms that did not contain a primary industry classification, firms that reported results with missing values and firms that were not reported to be active in the same industry classification over the 6-year period. The final sample contains 210 observations for 35 firms across four industry classifications of the textile sector, and 714 observations for 119 firms across eight industry classifications of the IT sector. According to Brooksbank’s (1991) classifications of SMEs, firms with fewer than 500 employees dominate the sample set of Taiwan’s textile and IT industries – 69% and 70%, respectively – even though all the companies in this sample are listed and traded on the TAIEX. Taiwan’s textile
and IT industry classifications collected from the TEJ database are based on the classifications of the TAIEX, which is the government-monitoring authority of stock transactions on Taiwan’s listed firms. The industry classifications and the numbers of final sample firms are shown in table 2.
5.3 Statistical methodology
In general, past studies used two main statistical methods to decompose the variances of profitability: fixed-effects and random-effects ANOVA. In fixed-effects ANOVA, research typically estimates a null regression model of no effects on the response variable other than a constant term, and then progressively adds variables that represent each effect in the model. While the fixed-effects version of ANOVA is common, the variance components approach, sometimes termed random-effects ANOVA, is also a popular method to estimate industry, firm and year effects on business profitability. Estimation of random effects incorporates the assumption that each effect represents a random sample of the true population effect and that each effect is independent of the other effects in the model (Neter et al. 1996). Additionally, in SASÕ packages, a variance components analysis can control for biases that arise
from the order of entry of predictor effects by rotating entries and adjusting estimates of the variables.
The variance components model is developed using the descriptive model in equation (1) by decomposing the total variance in the response variable (the performance measure) into its components as follows:
Sr2 ¼S2þS2þS2þS2 ð2Þ
The response variable rijt in the descriptive model has constant variance and is
normally distributed because it is a linear combination of independent normal random variables. This work uses the VARCOMP procedure in SASÕto estimate the different variance components. The variance components estimation is particularly suited to this study because it does not require a data set covering the whole population while at the same time allowing the results to be generalized. This is useful since it is impossible to construct a data set that covers all industries and all firms in each industry.
One inherent disadvantage of variance components estimation is that the procedure does not provide reliable tests for significance of the predictor
Table 2. Taiwan’s textile and IT industry classifications and the number of
sample firms.
Textile industry classification No. of firms IT industry classification No. of firms
Textile-Cotton spinning 7 Electronics-Communication 5
Textile-Garment 12 Electronics-Electronic Component 22
Textile-Synthetic fibre 12 Electronics-Electronic Channel 8
Textile-Wool spinning 4 Electronics-IC Produce 22
Electronics-Motherboard 26
Electronics-Network Modem 9
Electronics-Photoelectric/IO 21
effects (Hawawini et al. 2003). Since the effects are random, the null hypothesis that some of the variance parameters are zero lies on the boundary of the parameter space. This characteristic presents a non-standard problem for producing test statistics. In this regard, Schmalensee (1985), Rumelt (1991) and McGahan and Porter (1997) resolve the limitation in variance components’ estimation using nested ANOVA techniques that consider the effects to be fixed. The nested ANOVA approach generates F-statistics for the presence of predictor effects. In this way, the fixed-effect ANOVA transformation resolves the testing problem of the variance components procedure.
6. Empirical results
Using the TEJ database, we test whether the magnitude of firm and industry effects are sensitive to the performance measure. Table 3 shows the variance component estimates for the predictor variables that sum to the variance in the response variable for EVA, MVA and ROA. Table 4 reports the percentages of the total variance of the response explained by the predictor effects of the descriptive model. All estimates were evaluated at the 5% level by the nested ANOVA procedure for statistical significance. The results in table 4 indicate that the performance measures EVA, MVA and ROA
Table 3. Absolute values of the variance contributed by predictor variables for
years 1998–2003.
Variance estimate for variable
Textile industries IT industries
Variance component EVA MVA ROA EVA MVA ROA
Year effect 8.95Eþ10 2.06Eþ18 0.000118 3.09Eþ11 2.35Eþ20 0.000789
Industry effect 1.36Eþ11 4.50Eþ19 0.000495 2.43Eþ11 7.08Eþ20 0.000170
Firm effect 5.29Eþ12 5.99Eþ20 0.002043 1.62Eþ13 6.45Eþ21 0.004649
Error 2.06Eþ12 2.37Eþ20 0.003665 7.87Eþ12 3.77Eþ21 0.005519
EVA: Economic Value Added, MVA: Market Value Added, ROA: Return On Assets.
Table 4. Industry and firm effects as percentages of total variance of the response
variable for years 1998–2003 based on the data reported in table 3.
Textile industries IT industries
Variance component EVA MVA ROA EVA MVA ROA
Year effect 1.18 0.23 1.87 1.25 2.11 7.11
Industry effect 1.80 5.10 7.82 0.99 6.34 1.53
Firm effect 69.83 67.84 32.33 65.79 57.78 41.88
Model 72.81 73.17 42.02 68.03 66.23 50.28
Error 27.19 26.83 57.98 31.97 33.77 49.72
explain 72.81%, 73.17% and 42.02% of the total variance in textile firm profits and 68.03%, 66.23% and 50.28% of the total variance in IT firm profits, respectively.
From the results, it is evident that firm effects dominate long-term performance irrespective of whether performance is measured by EVA, MVA or ROA. Firm effects for EVA, MVA and ROA explain 69.83%, 67.84% and 32.33% of variance in textile firm profits and 65.79%, 57.78% and 41.88% in IT firm profits, respectively. In comparison, the corresponding figures for industry effects are 1.80%, 5.10% and 7.82% in textile firms and 0.99%, 6.34% and 1.53% in IT firms. The results provide strong support for the idea that the firm-related resources and capabilities of Taiwan’s textile and IT firms have important influences on profitability. In contrast, industry factors, such as industry membership and structural characteristics of textile and IT sectors, have little impact on performance, and the finding is robust across various performance measures. Therefore, when seeking to explain the sources of performance differentials among textile and IT firms in Taiwan, firm-related factors dominate industry effects. Year effects for EVA, MVA and ROA explain 1.18%, 0.23% and 1.87% of variance in textile firms and 1.25%, 2.11% and 7.11% in IT firms, respectively. By definition, these effects are macroeconomic fluctuations that affect all firms to the same degree in a particular year.
There is a degree of consistency in the results across the three performance measures. The first reason for this is that in this cross-sectional and longitudinal study, discrepancies resulting from different accounting measurements might even out over a period of time (Kay 1976). The second reason could be that while the results are similar, the processes that lead to the results might vary. Results across the three measures indicate only that firm factors are relatively more important than industry effects, but we cannot say whether the firm factors that drive performance in terms of EVA, MVA and ROA are the same (Hawawini et al. 2003).
Table 5 contains the comparable figures from Schmalensee (1985), Rumelt (1991), Roquebert et al. (1996), McGahan and Porter (1997, 2002) and Hawawini et al. (2003) on the various effects. There is a relatively moderate amount of error associated with the model in this study, specifically 26.9 to 33.8% in terms of EVA and MVA measures, in contrast with large errors reported in the past studies, which range from Rumelt’s (1991) 44.8% to Schmalensee’s (1985) 80.4%, as shown in table 5. One reason for the superior model fit may be that the number of Taiwan’s textile and IT sectors is relatively smaller than that of US sectors used in previous studies.
7. Discussion and concluding remarks
Our interpretation of Taiwan’s textile and IT industrialization diverges from mainstream perspectives, neoliberalism, structural-institutionalism, flying geese patterns, regional networks and economic geography in that we do not start from nations, but rather from generic conceptions of sector development, and from a strong entrepreneurial culture context in influencing the sources of profitability differentials among Taiwan’s textile and IT firms in international competitiveness has been on the relative importance of benign regional development policies (industry effects) and unique organizational processes (firm effects). At the same time, this study revisits questions of whether firms’ performance is driven by industry or firm factors, extending recent research in two major ways. First, this work considers Taiwan’s textile and IT sectors exclusively. Second, this investigation tests industry and firm
Table 5. Comparison of percentages of total variance of the component effects. Variance component Schmalensee (1985) Rumelt* (1991) Roquebert et al.** (1996) McGahan & Porter (1997) McGahan & Porter (2002) Year effect N/A N/A 0.4 2.4 0.4 Industry Effect 19.6 4.0 10.2 18.7 10.3 Firm effect y 0.6 45.8 55.0 36.0 47.6 Error 80.4 44.8 32.0 48.4 41.7 Hawawini et al. (2003) Current textile industries Current IT industries EVA/CE MVA/CE ROA EVA MVA ROA EVA MVA ROA Year effect 1.9 1.3 1.0 1.2 0.2 1.9 1.3 2.1 7.1 Industry effect 6.5 11.4 8.1 1.8 5.1 7.8 1.0 6.3 1.5 Firm effect y 27.1 32.5 35.8 69.8 67.8 32.3 65.8 57.8 41.8 Error 60.3 51.9 52.0 27.2 26.9 58.0 31.9 33.8 49.6 *Only the results of sample B in Rumelt (1991) are reproduced here. ** Only a verages across samples in Roquebert et al. (1996) are reproduced here. yFirm effects include both corporate and business-level effects. EVA: Economic Value Added; MVA: Market Value Added; ROA: Return on Assets; CE: Capital Employed.
effects using the raw accounting measure ROA and the economic performance measures EVA and MVA.
The results of the dominance of firm effects on the performance differentials among firms explains how much a resource-based theory of the firm matters in Taiwan’s textile and IT sectors and emphasizes firms’ internal characteristics rather than the external environment (Barney 1991). From this perspective, firm resources and capabilities have a significant impact on their strategies because their firm-specific idiosyncrasies, such as strong entrepreneurial culture, effective operational processes and organizational patterns, in the accumulation of valuable and rare resources and capabilities create sustained competitive advantages (Rumelt 1984, Wernerfelt 1984, Barney 1991, Collis 1991, Conner 1991) and suggest basic directions for decisions (Grant 1991, Oliver 1997). For example, the empirical evidence of dominant firm effects is consistent with the observation of Taiwan’s ‘strong society model’ mentioned by Hamilton and Biggart (1988), who indicated the organizational patterns of Taiwan’s SMEs – the particular network configurations of firms – were pivotal in promoting economic success and demonstrated important differences in Taiwan’s export sectors. The extraordinary growth and the dominance of SMEs in Taiwan’s textile and IT sectors could not be explained by those industrial targeting policies of Taiwan government alone but is better explained by widespread entrepreneurship in Taiwan. Chinese cultural value orientations and mediating environmental factors play significant roles in shaping the attitudes and behaviour of Taiwanese owner-managers and the marketing practices of Taiwanese SMEs (Siu et al. 2004). Additionally, Taiwanese management is significantly different from Western manage-ment (Chen 1995, 1999). For example, Chen (1999) highlighted four key features of Taiwanese management that are distinctive from Western management and are persistent in most of Taiwan’s textile firms: human-centeredness, family-centeredness, centralization of power, and small size. However, globalization and the need to integrate local human resources into the organization forced Taiwan’s IT firms to break away their traditional family-centered management style (Chen 2003).
The above conventional view explains the importance of entrepreneurship, organizational patterns and management styles to Taiwan’s textile and IT industrialization, what else are the essential firm-specific activities, resources, capabilities, competencies, or knowledge to the successes of Taiwanese SMEs? Taiwan’s textile and IT firms are sensitive to market trends (Song et al. 1997, Siu et al. 2004, Siu 2005) and aggressive in resource creation and utilization (Hamilton and Biggart 1988, Li 1989, Lin 1998, Chen 1999, Chiao et al. 2004), in networking (Mathews 1997, Lin 1998, Chen 1999, Saxenian and Hsu 2001, Chang 2003), and in internationalization (Wade 1990, Levy 1991). Despite these similarities, however, the sector development of Taiwan’s textile industry is qualitatively different from the sector development of Taiwan’s IT industry.
The outstanding success of the textile industry propelled it to become the core industry in the Taiwanese specialization model during the 1960s and continuing into the 1980s. What factors affected Taiwanese firms operating in traditional labour-intensive industries since the 1980s? The sharp appreciation of the Taiwan dollar, the severe shortage of labour and the consequent escalation of wages, the loss of the US Generalized System of Preference status which grants textile products imported from Taiwan either duty-free access or a tariff reduction, the rise of real estate prices and the aggressive competition from the Korean chaebol in the late 1980s each played a role. Under these certain challenges, why there is a significant differential
performance among textile firms? From the perspective of the resource-based theory of the firm, company strategies can differ between firms within an industry (Nelson 1991). In fact, these increasingly competitive pressures led to a remarkable restructuring in the labour-intensive textile industry. Many Taiwanese textile SMEs were compelled to shift production abroad (mainly to Southeast Asia and China) to maintain competitiveness. The remaining textile enterprises had to redirect their business towards more skill-intensive, R&D-oriented products to find new product niches and new market areas in order to survive.
Indeed, Taiwan’s textile and clothing sector has undergone a rapid process of upgrading from a few traditional spinning and weaving products to capital- and technology-intensive man-made fibres and fashionable clothing. Currently, garment firms that continue to produce in Taiwan are all specialized in high-end products with strong design content. At the same time, the overall industrial structure has diversified remarkably towards higher technology products and sectors. The IT industry in Taiwan followed a totally different path of development. While the textile industry received little FDI, Taiwan’s IT industry depended heavily on international markets and access to foreign technology. A key explanation for the success of Taiwanese firms competing in globalized high-tech industries is the co-evolution of domestic and international knowledge linkages (Chang 2003, Guerrieri and Pietrobelli 2004). In other words, inter-firm and inter-institution linkages were built to provide Taiwan’s IT firms with the necessary external connections to cope with the dual challenge of knowledge creation and internationalization (Chen 2003).
Nevertheless, why is there a substantial profitability difference among IT firms driven by firm factors? Similarly, from the strategic view of the resource-based theory of the firm, Chang et al. (2003) indicated that compatibility of quality capabilities and business strategy is necessary for a firm to achieve superior performance and suggested that Taiwan’s IT firms should invest resources and time to develop quality capabilities that fit into their business strategies. For example, first, Taiwan’s IT firms with preemptive or first-mover strategies should be aware of the effect on their profit in the short term when developing high-quality capabilities. Most R&D projects of new products in high-tech industries involve huge expenditures over several years. It is likely that the large amounts of R&D investment cannot be fully recovered in the short term, which explains the insignificant effect of high performance on net profit in the short term. Interestingly, dealing with the slow investment return of developing cutting-edge products, some IT manufacturing firms in Taiwan collaborated with government-supported R&D institutes such as the ITRI and/or universities in new product designs, pilot tests and process design. Chang et al. (2003) found that Taiwan’s IT companies were able to improve net profit with high-quality products through such collaborative efforts.
Second, IT firms’ service ability dimension with a differentiated or follower strategy does not lead to higher profits. One reason is that service ability has become an ‘order qualifier’ for most high-tech OEMs in Taiwan. With increasingly severe competition from nearby countries such as China, South Korea and Singapore, it is common practice that business agreements with major US and European customers include attractive after-sale services. Such service agreements can encourage more customer orders and sales at the expense of net profit for Taiwan’s IT firms unless they follow suit. As in most descriptive studies of this nature, there are difficulties in formulating a model and interpreting results. The data did not allow us to distinguish between
corporate- and business-level effects, therefore our results may underestimate industry effects. Nevertheless, this study provides deeper understanding of the processes linked to minor and dominant effects at the industry and firm levels. Our results present compelling evidence of the relative magnitudes to which these factors influence firm performance and of business strategy implications for marketing, R&D and after-sale services among textile and IT firms in Taiwan.
Acknowledgements
The authors thank two anonymous reviewers and Bengt Johannisson for their valuable comments.
References
Amsden, A. 1985 The state and Taiwan’s economic development, in Evans, P., Rueschemeyer, D. and Skocpol, T. (eds), Bringing the State Back In (Cambridge, MA: Cambridge University Press) pp. 78–106. Amsden, A. 1991 Diffusion of development: the late-industrializing model and greater East Asia, American
Economic Review, 81 (2): 282–286.
Barney, J. 1991 Firm resources and sustained competitive advantage, Journal of Management, 17 (1): 99–120. Bernard, M. and Ravenhill, J. 1995 Beyond product cycles and flying geese: regionalization, hierarchy and
industrialization in East Asia, World Politics, 47 (1): 171–209.
Borrus, M. 1997 Left for dead: Asian production networks and the revival of US electronics, in Naughton, B. (ed.), The China Circle: Economics and Technology in the PRC, Taiwan and Hong Kong (Washington, DC: Brookings Institution Press) pp. 139–209.
Brooksbank, R. 1991 Defining a small business—a new classification of company size, Entrepreneurship & Regional Development, 3 (1): 17–33.
Brush, T., Bromiley, P. and Hendrickx, M. 1999 The relative influence of industry and corporation on business segment performance: an alternative estimate, Strategic Management Journal, 20 (6): 519–547. Buckley, P. and Ghauri, P. 2004 Globalisation, economic geography and the strategy of multinational
enterprises, Journal of International Business Studies, 35 (2): 81–98.
Callon, S. 1995 Different Paths: The Rise of Taiwan and Singapore in the Global Personal Computer Industry (Stanford, CA: Asia-Pacific Research Center, Stanford University).
Chang, Y.-C. 2003 Benefits of co-operation on innovative performance: evidence from integrated circuits and biotechnology firms in the UK and Taiwan, R&D Management, 33 (4): 425–437.
Chang, S.-C., Lin, N.-P., Yang, C.-L. and Sheu, C. 2003 Quality dimensions, capabilities and business strategy: an empirical study in high-tech industry, Total Quality Management, 14 (4): 407–421.
Chauvin, K. and Hirschey, M. 1993 Advertising, R&D expenditures and the market value of the firm, Financial Management, 22 (4): 128–140.
Chen, M. 1995 Asian Management Systems: Chinese, Japanese and Korean Styles of Business (New York: Routledge).
Chen, T.-J. 2003 Network resources for internationalization: the case of Taiwan’s electronics firms, Journal of Management Studies, 40 (5): 1107–1130.
Chen, W.-H. 1999 Manufacturing strategies of network-based small firms: observations on the textile industry in Taiwan, Journal of Small Business Management, 37 (2): 46–62.
Chiao, Y.-C., Yang, K.-P. and Yu, C.-M. 2004 Performance, internationalization, and FSAS of SMEs in newly-industrialized economies, Academy of Management Proceedings, IM: O1–O6.
Clark, G., Feldman, M. and Gertler, M. 2000 Economic geography: transition and growth, in Clark, G., Feldman, M. and Gertler, M. (eds), The Oxford Handbook of Economic Geography (Oxford: Oxford University Press) pp. 3–17.
Collis, D. 1991 A resource-based analysis of global competition: the case of the bearings industry, Strategic Management Journal, 12 (4): 49–68.
Conner, K. 1991 A historical comparison of resource-based theory and five schools of thought within industrial organization economics: do we have a new theory of the firm?, Journal of Management, 17 (1): 121–154.
Dedrick, J. and Kraemer, K. 1998 Asia’s Computer Challenge: Threat or Opportunity for the United States and the
World?(New York: Oxford University Press).
Edgington, D. and Hayet, R. 2000 Foreign direct investment and the flying geese model: Japanese electronics firms in the Asia pacific, Environment and Planning A, 32 (2): 281–304.
Ernst, D. 2000 Inter-organizational knowledge outsourcing: what permits small Taiwanese firms to compete in the computer industry, Asia Pacific Journal of Management, 17 (2): 223–256.
Ferguson, C. and Morris, C. 1993 Computer Wars: The Fall of IBM and the Future of Global Technology (New York, NY: Times Books).
Fisher, F. and McGowan, J. 1983 On the misuse of accounting rates of return to infer monopoly profits, American Economic Review, 73 (1): 82–97.
Ghauri, P. and Prasad, S. 1995 A network approach to probing Asia’s interfirm linkages, Advances in International Comparative Management, 10: 63–77.
Gilder, G. 1989 Microcosm: The Quantum Revolution in Economics and Technology (New York: Simon & Schuster).
Gold, T. 1986 State and Society in the Taiwan Miracle (New York: Sharpe).
Grant, R. 1991 The resource-based theory of competitive advantage: implications for strategy formulation, California Management Review, 33 (3): 114–135.
Guerrieri, P. and Pietrobelli, C. 2004 Industrial districts’ evolution and technological regimes: Italy and Taiwan, Technovation, 24 (11): 899–914.
Hamilton, G. and Biggart, N. 1988 Market, culture, and authority: a comparative analysis of management and organization in the Far East, American Journal of Sociology, 94: S52–S94.
Harcourt, G. 1965 The accountant in a golden age, Oxford Economic Papers, 17: 65–80.
Hart-Landsberg, M. and Burkett, P. 1998 Contradictions of capitalist industrialization in East Asia: a critique of ‘flying geese’ theories of development, Economic Geography, 74 (2): 87–110.
Haspeslagh, P., Noda, T. and Boulos, F. 2001 Managing for value: it’s not just about the numbers, Harvard Business Review, 79 (7): 64–73.
Hatch, W. and Yamamura, K. 1996 Asia in Japan’s Embrace: Building a Regional Production Alliance (Cambridge, MA: Cambridge University Press).
Hatfield, G. 2002 R&D in an EVA world, Research Technology Management, 45 (1): 41–47.
Hawawini, G., Subramanian, V. and Verdin, P. 2003 Is performance driven by industry- or firm-specific factors? A new look at the evidence, Strategic Management Journal, 24 (1): 1–16.
Hill, R. and Fujita, K. 1996 Flying geese, swarming sparrows or preying hawks? Perspectives on East Asian industrialization, Competition and Change, 1 (3): 285–298.
Hsu, J.-Y. 1997 A late industrial district? Learning networks in the Hsinchu science-based industrial park, Ph.D. dissertation, Department of Geography, University of California at Berkeley, Berkeley, CA. Kay, J. 1976 Accountants too could be happy in a golden age: the accountant’s rate of profit and the
internal rate of return, Oxford Economic Papers, 28: 447–460.
Kelley, L., Whatley, A. and Worthley, R. 1987 Assessing the effects of culture on managerial attitudes: a three-culture test, Journal of International Business Studies, 18: 17–31.
Khanna, T. and Rivkin, J. 2001 Estimating the performance effects of business groups in emerging markets, Strategic Management Journal, 22 (1): 45–74.
Kraemer, K., Dedrick, J., Hwang, C.-Y., Tu, T.-C. and Yap, C.-S. 1996 Entrepreneurship, flexibility, and policy coordination: Taiwan’s computer industry, Information Society, 12: 215–249.
Krugman, P. 1991 Geography and Trade (Cambridge, MA: MIT Press).
Krugman, P. 2000 Where in the world is the ‘new economic geography’?, in Clark, G., Feldman, M. and Gertler, M. (eds), The Oxford Handbook of Economic Geography (Oxford: Oxford University Press) pp. 49–60. Levy, B. 1991 Transaction costs, the size of firms and industrial policy, Journal of Development economics,
34 (1/2): 151–178.
Li, W. 1989 Entrepreneurial role and societal development in Taiwan, in Tai, H. (ed.), Confucianism and
Economic Development: An Oriental Alternative?(Washington, DC: Washington Institute Press) pp. 128–148.
Liang, D.-R., Chow, C. and Wu, A. 2002 The feasibility of applying economic value added on Taiwan’s firms, Accounting Research Monthly, 205: 110–118.
Lin, C. Y. 1998 Success factors of small- and medium-sized enterprises in Taiwan: an analysis of cases, Journal of Small Business Management, 36 (4): 43–56.
Markusen, J. and Venables, A. 2000 The theory of endowment, intra-industry and multi-national trade, Journal of International Economics, 52 (2): 209–234.
Mason, E. 1939 Price and production policies of large scale enterprises, American Economic Review, 29 (1): 61–74.
Martin, J. and Petty, J. 2000 Value Based Management (Oxford: Oxford University Press).
Mathews, J. 1997 A Silicon Valley of the East: creating Taiwan’s semiconductor industry, California Management Review, 39 (4): 26–54.
Mauri, A. and Michaels, M. 1998 Firm and industry effects within strategic management: an empirical examination, Strategic Management Journal, 19 (3): 211–219.
McGahan, A. and Porter, M. 1997 How much does industry matter, really?, Strategic Management Journal, 18 (1): 15–30.
McGahan, A. and Porter, M. 2002 What do we know about variance in accounting profitability?, Management Science, 48 (7): 834–851.
Montgomery, C. and Porter, M. 1991 Strategy: Seeking and Securing Competitive Advantage (Boston, MA: Harvard Business School Publishing).
