INTRODUCTION

The competitiveness of US, Japanese, Taiwanese and Korean(1) multinational enterprises (MNEs) in different aspects of microcomputer(2) component markets can be analyzed from the perspectives of economics or organizational studies. Economists such as Ricardo (1817) as well as Heckscher-Ohlin's (1933) theory of comparative advantage suggested that MNEs specialized in microcomputer components in which they could produce most efficiently. Vernon's (1966) product cycle model, in the other hand, proposed a sequential process where new microcomputer components would be first manufactured in the USA, then in other advanced economies such as Japan, and eventually in less developed countries such as the Philippines. Organizational theorists Lieberman and Montgomery's (1988) first-mover advantage furnished a rationale for the competitiveness of microcomputer component MNEs in terms of preemption of product space. Furthermore, Wernerfelt (1984), as well as Prahalad and Hamel (1990), proposed that competitiveness among microcomputer component MNEs arose from individual organizational capabilities. Finally, Porter's (1990) diamond theory explained that MNEs' international success was a result of the competitive advantages embedded in their homebases.

Figure 1 provides an overview of static competitiveness(3) in relation to microcomputer components of high monetary value. A typology of components incorporated in microcomputers as differentiated by concept-intensity, capital-intensity and labor-intensity is shown in Table 1. Concept-intensive components encompass the designs of microprocessors and software. Capital-intensive components refer to the assembly of dynamic random access memories (D-rams) and printer engines while labor-intensive components include monitors, keyboards, and printed circuit boards. The highest demand for skilled labor is in the concept-intensive components. For example, it took IBM, Motorola, and Apple designers in the Somerset Joint Design Center, Texas, nearly two years to complete the Power PC Architecture, which was implemented in the Power PC601, 603, 604, and 620 microprocessors. The critical feature of capital-intensive components, on the other hand, is that they contained the highest level of minimum efficient scale and initial capital investment. For instance, the minimum efficient scale for hard disk drives as estimated from IBM's Thailand facility in 1997 was approximately 1,000,000 per month, whereas the capital requirement for building a 256-megabit semiconductor plant was estimated to be about US$3 billion in 1998. (Financial Times, 11/28/1996:4 and 6/23/1998:35). Finally, labor-intensive components have the longest product life cycle.

FIGURE 1
Microcomputer Component World Market Share Estimates(*) of Producers
headquartered in the USA, Japan, Taiwan, and Korea in the mid-1990s

a. Concept-intensive Components of Microprocessors and Software

% of total sales value

         Microprocessors   Software

USA            96%           74%
JAPAN          4%            18%
OTHERS                        8%

Notes:

(1.) Static competitiveness for microcomputer components in 1995 was
adopted; it corresponded to the dynamic competitiveness for
concept-intensive components since the 1970s and for capital-intensive
components and labour-intensive components from the mid-1980s onwards.

(2.) Microprocessors refer to the widely accepted CISC
microprocessors. For example, the 32-bit CISC microprocessor market was
worth US$11.974 million in 1995, while the 32/62-bit RISC
microprocessor was worth only US$1.355 million. It has further been
estimated that by 2000, the CISC market will reach US$30.357 million
as compared to the US$4.530 RISC market.

(3.) Software includes all systems and applications software,
languages, networking systems, application development tools, and
other software not embedded in hardware.

Source: Estimation based on Business Week Nov 6 1995 p40, Datamation
June 15 1996 p55 and ICEC 1996a 4.-22.

Note: Table made from bar graph.

[ILLUSTRATION OMITTED]

TABLE 1
A Typology of Microcomputer Component Technology

                                                 Capital-intensive
                          Concept-intensive          Component
                              Component         e.g. D-rams. printer
                            e.g. Software,         engines, hard
                           microprocessors          disk drives

Attributes
Labour skill intensity   High                   Medium
Minimum efficient
  scale                  Low                    High
initial capital
  investment             Low                    High
Product cycle            Short                  Short

                           Labor-intensive
                              component
                         e.g. monitors, mice
                              keyboards

Attributes
Labour skill intensity   Low
Minimum efficient
  scale                  Medium
initial capital
  investment             Medium
Product cycle            Long

Source: Author 1996.

Figure 1 shows that in the mid-1990s US MNEs captured phenomenal market shares of over 70% of concept-intensive components whereas Japanese and Korean MNEs had comparable market positions in capital-intensive components. Similarly, Taiwanese and Korean MNEs had market shares of approximately 70% in labor-intensive components. This figure demonstrates the fact that organizational capabilities are unequally distributed among the 4-nation microcomputer component MNEs and suggests the weaknesses of the aforementioned perspectives. The mobility of MNEs' manufacturing operations through the deployment of foreign direct investment means that they can extract the comparative advantages or competitive advantages of different localities, hence highlighting the inadequacy of applying these perspectives as means to explain the competitiveness of microcomputer component MNEs. Similarly, the MNEs' abilities to relocate their manufacturing activities to the most appropriate sites in accordance with the stages of the component product development cycle also indicates the incompleteness of Vernon's model. In addition, the concepts of first-mover advantage and organizational capabilities failed to capture the national pattern of microcomputer component production. Specifically, though MNEs' exploitation of scarce assets outside their national boundaries allows them a free-ride and challenges early movers' leading positions, the national patterns of catching up among later movers cannot be justified.

The purpose of this paper is to explain inter-country variance in microcomputer technology specialization by means of a framework built upon national culture and organizational capabilities. It is proposed that the values, beliefs, evaluations, and judgments among top decision makers of the 4-nation microcomputer component MNEs are conditioned by their respective cultural values. Hence, the distinct core values of the USA, Japan, Korea and Taiwan can hinder or enhance knowledge accumulation on component technology among national MNEs, which in turn will affect their capabilities for creating and sustaining economic rents when competing internationally.

The organization of this paper is as follows. In the next section, the Cultural Selectivity Hypothesis will be advanced to explain the country-specific competitiveness pattern within the microcomputer component industry. The second section tests the hypothesis with qualitative case studies of the industry participants in Europe as well as with summary statistics concerning the component categories. This will be followed by a discussion of the result. The final section contains some concluding comments.

THE SELECTIVITY OF NATIONAL CULTURE ON COMPETITIVENESS

The Cultural Selectivity Hypothesis posits a relationship between national culture and national competitiveness. National culture has been defined by Hoftstede (1980) in terms of core values such as individualism versus collectivism, strong versus weak uncertainty avoidance and masculinity versus femininity. In this paper, the focus is upon individualism, continuous improvement, and flexibility. These core values are selected on the premise that they contribute towards microcomputer component MNEs' competitiveness in areas of concept-intensive, capital-intensive and labor-intensive component technologies. The importance of national culture on MNEs' capabilities and consequently their competitiveness has been explained by Casson (1995). He (ibid:89) wrote that "since different values legitimate different objectives, and different objectives generate different kinds of problem, societies with different cultures will tend to focus on distinctive types of problem solving. `Learning by doing' is an important aspect of problem solving and so learning effects will give each culture a distinctive kind of problem-solving expertise".

It is therefore suggested that knowledge-based capabilities within the microcomputer component industry vary among MNEs headquartered in different countries. The differentials of US, Japanese, Korean, and Taiwanese MNEs' capabilities are derived from their heterogeneous learning abilities that are in turn influenced by the national culture of their headquarters, as expressed in individualism, continuous improvement and flexibility. Thus, US MNEs will excel in concept-intensive components where competitive advantages are consistent with their product design capabilities deriving from the core American value of individualism. Japanese MNEs will gain in capital-intensive components where competitive advantages hinge on their manufacturing process capabilities which are in turn derived from their core cultural values of continuous improvement. On the other hand, Taiwanese MNEs' strengths in labor-intensive components where competitive advantages are obtained from their people management capabilities built on core cultural values of flexibility. Finally, Korean MNEs which originated from a society having inherited cultural values from both China and Japan, is able to capitalize on their manufacturing processes and people management capabilities and compete in both capital-intensive and labor-intensive components.

Individualism is manifested in loose human relationships within a society and is fostered by advocating individual freedom. An individualistic society encourages its members to be self-reliant, unique, and competitive in order to achieve their goals. Individualistic cultures emphasize personal identity and hence encourage the creation of unique ideas. For instance, US business schools emphasize the importance of completely original ideas in starting new businesses. Hudson (1966) examined creativity in terms of the cognitive styles of convergent thinking and divergent thinking and suggested that the latter style was associated with creativity.(4) In other words, US MNEs that are entrenched in the strong notion of an independent construct of self are more inclined towards the acquisition of product design capability for the construction of abstract ideas. America's individualism enhances creativity and facilitates the accumulation of product design capabilities among concept-intensive MNEs, and they compete via unique product attributes. Penrose (1959:106) explained that the creative process "seems to be most highly developed" in the United States. In the US, "a kind of `competition in creativity' has become a dominant motif in the pattern of competitive behavior in many industries, where consumers and producers alike are caught up in an almost compulsive obsession for that which is `new.'" Jones (1970) stated that software designers were obliged to use current information to predict a future outcome and to work backwards to test the effect of their assumed designs. Creativity is therefore an important ingredient in the production of concept-intensive software. Commenting on the US microprocessor industry, Ferguson et al (1993:122) stated that "the American advantage in the architectural context is based on a massive advantage in rapid innovation, conceptualization, rather than detailed engineering." US MNEs have exhibited long traditions of promoting individualistic work practices advocated by management theorists in the areas of reward schemes, pet projects and dress codes ... For instance, one of the early management thinkers, Frederick Taylor, advocated the wage differential system to reward workers for performance on an individual basis.

Continuous improvement is the constant step by step improvement towards perfection, which is critical in the acquisition of manufacturing process capabilities in relation to capital-intensive components. Manufacturing process capabilities can be considered as tacit skills that hinge upon technological knowledge drawn from the constant refinement of work procedures in maximizing the quantity of output for each unit of input employed in a production cycle. The manufacturing process capabilities of continuously improving Japanese and Korean microcomputer component MNEs enable them to implement stringent quality control in capital-intensive component production. The concept of quality control was brought to the Far East by the Americans Deming and Juran in the 1950s. As the notion of improvization coincides with Japan's core cultural value of continuous improvement, it was not surprising that Japanese MNEs embraced and refined the quality control concept which is merely an extension of the country's cultural values. Continuous improvement is a philosophy as well as a way of life in Japan and Korea; it is embodied in the attention to details, the emphasis on proper procedure, and the impeccably arranged workplace. It has permeated Japanese and Korean MNEs of all industrial sectors. For example, the Japanese MNE, Nissan, regarded continuous improvement as "applied anywhere, anytime, anyplace". It summarized the essence of continuous improvement and stated that "our world is far from perfect and we are not going to change it overnight. But, step-by-step, we can improve that small part with which we come into contact. Whether it is at home, at work or in our social activities, the principles of kaizen apply" (Garrhan and Stewart, 1992:61). Such drive for perfection is also exhibited in Korean MNEs, such as Samsung, whose founder stated that "money is not what I pursued. Instead I have only striven to be the best in whatever business I choose" (Business Korea, 1987:47).

Flexibility refers to the willingness to take action so as to maximize the benefits derived from altered conditions; it arises from being able to perceive the situation from divergent or even dichotomous perspectives. As labor inputs are far less homogenous than mechanical inputs, flexibility contributes to the understanding of managing people that enables Taiwanese and Korean labor-intensive microcomputer component MNEs to extract the maximum productivity from the workforce in low cost localities, which are hampered by particular economic constraints, such as low levels of general education or power shortages. The CEOs of both Acer (Taiwan) and Samsung (Korea) exhibited people management capabilities. When Acer's CEO offered his resignation as a result of consecutive financial losses in 1992, the board refused it and announced publicly that they admired and confirmed his outstanding leadership ability (Fortune, 10/30/1995:74). Indeed Acer turned around and has become a leading microcomputer manufacturer worldwide. This paralleled a gesture of Samsung's CEO, who promised to donate up to 80% of his worth to a fund for his factory workers if his program on quality improvement was successful (London Times, 10/23/1994). Flexibility is entrenched in the polarized philosophy of Confucianism and Taoism, which co-exist in both the Korean and the Taiwanese societies. Confucianism emphasizes tradition, ceremony, discipline, and material success whereas Taoism adopts a post-materialistic view such as abandoning conventional social obligation and distancing oneself from worldly affairs. Sun Tzu commented on the importance of flexibility in his famous book The Art of Warfare: "Just as water shapes itself according to the ground, an army should manage its victory in accordance with the situation of the enemy. Just as water has no constant shape, so in warfare there are no fixed rules and regulations. Therefore, do not repeat the tactics that won you a victory, but vary them according to the circumstances" (Wang 1976: 201). The core value of flexibility of the Taiwanese and the Koreans can be seen in their attitude towards social class. Though social class systems exist in both Korea and Taiwan, the Koreans and the Taiwanese openly accept upward mobility as a result of one's talent and capacity. In addition, the flexibility of Taiwanese MNEs can be illustrated by their participation in the microcomputer component industry. Indeed the early participants were driven into the production of Apple II compatible components after the government banned video game manufacturing in the late 1970s. On the other hand, Korean MNEs flexibility was also noted. One of Hitachi's managers explained the flexibility of the Koreans as: "Korea is a land of division, so the people are willing to listen and not get their feet stuck in concrete" (Business Week, 1/19/1987).

Finally, the dual cultural values of Korean MNEs' are derived from the country's interwoven history with China and Japan. From 109BC to 735AD Korea was a tribute state of China; Confucianism and Taoism were introduced and subsequently evolved into part of Korean thinking and behaviour. Chang and Chang (1994:9) explained that Korea "has been under the influence of the Chinese culture for more than a thousand years ... All aspects of the Chinese culture spread into Korea through Chinese books, traditions, customs, and the value system." In fact, the Korean national flag carries the symbol of yin-yang, which is utilized by the Taoists to denote a dichotomous perception of life. Japan's influence upon Korea began in 1904 during the Russo-Japanese War and was formalized after its annexation of Korea in 1910. Its impact upon Korea was highly visible; Korean built schools, highways, railroads, ports, and communications facilities; they also created a government bureaucracy. Consequently, Korean MNEs exhibited the characteristics of seniority and familism that are associated with the Chinese philosophy of Confucianism. Their consensus decision making system, the pumui system, however, came from Japan and is similar to the ringi system.

HYPOTHESIS TESTING

In this section, I will be examining the validity of the Cultural Selectivity Hypothesis with data collected from the microcomputer component industry and its participants. Both primary and secondary data were utilized; the former were based upon information collected from interviews while the latter were gathered in the public domain from newspapers, the Internet, and industry publications. I will first incorporate primary data of seven MNEs that originated from societies endowed with the core cultural values of individualism, continuous improvement, and flexibility in order to examine the hypothesis. I will then analyze the content of the secondary data with reference to the 4-nation MNEs in selected areas of concept-intensive, capital-intensive, and labor-intensive components.

Case Study Analysis

This section will discuss seven case studies that were derived mainly from person-to-person interviews with senior managers of the 4-nation microcomputer component MNEs between September 1995 and September 1998. These interviews were undertaken on the MNEs' premises within Europe; the interviews comprise one US MNE exhibiting the core cultural value of individualism (Company A) and six Far East MNEs originating from societies characterized by the values of continuous improvement (Company B, Company C, Company D, and Company E) as well as flexibility (Company F and Company G). On each case, I will examine whether the core cultural value enhanced the MNE's competitive position in the corresponding component areas.

Company A, which was headquartered in the USA, was renowned for its product design capabilities and had considerable market shares in its patented microcomputer component products. Company A took part in the development of Reduced Instruction Set Computer (RISC) microprocessors and had a diverse interest in software, which included system software such as operating systems and application software such as graphics packages. Though it had fostered a strong corporate culture, Company A always emphasized the importance of individualism. It was keen to establish its products as industry standards; indeed, its network architecture had become the basis for data communications work. Moreover, it assigned two or more groups to work on similar projects to generate competition among the groups and challenged its employees to be the best in their fields; it then rewarded the best people with individual incentive schemes. The practice of its sales executives were also synonymous with individualism i.e. they were the single points of contact with customers and were allocated sufficient authority to make relevant decisions that were critical to satisfying customers' needs. Company A publicly depicted its philosophy towards employees as being intended to bring out their energies, skills, talents, and creativity and hence restated its respect for the individual. Nevertheless, Company A had participated in various capital-intensive component production for internal consumption and merchant market sales; it had even established a market presence with items such as hard disk drives. In addition, Company A also used to undertake some labor-intensive components prior to the mid-1990s.

The Japanese MNEs Company B, Company C, Company D, and Company E engaged in arrays of capital-intensive components. All four MNEs implemented continuous improvement within their global structure, and were credited with their emphasis on product quality. Company B was one of the five leading printer engine manufacturers worldwide and had established itself with technological innovation and superior quality; it extended its quality control program to the finance and the personnel functions. It should also be noted that Company B was involved in the writing of software that supported its printers, even though it did not originate from a culture associated with the core value of individualism. Though most of its software development was focused in Japan, it had also selected the UK as a center to undertake research activities that aimed to maximize printing performance. Moreover, Company B had initiated collaboration with Microsoft to work on the Windows interface. Contrary to most Japanese MNEs, its management encouraged individualism. Company B was highly decentralized, whereas subsidiaries maintained control over major strategic issues. In terms of product strategy, the UK subsidiary incorporated products that it perceived as suitable for the local market rather than those dictated by the headquarters. Its Marketing Manager mentioned that the UK subsidiary had even furthered the notion of decentralization to the promotion of a British national for the position of General Manager, which was considered untypical among Japanese MNEs.

Company C was one of the top ten D-rams manufacturers in the world and had invested in 64-megabytes, D-ram production. Company C had been widely regarded as having continuous improvement thinking. It had, in fact, diversified in the 1960s from telecommunications into computing and semiconductor production and had gradually improved its products to the current high standard. Its early semiconductors were directed towards the Japanese market and were below international quality and product configurations, Nevertheless, it initiated product-reliability campaigns as well as research and development activities to continuously improve the design and the quality of semiconductors. Company C had publicly stated that it admired IBM's expertise in marketing, and it had endeavored to nurture such capability. By the late 1980s, it was acknowledged for its sales prowess within Japan.

Company D was a younger company that had endeavored to continuously improve its component technology; it had participated in information technology products in the 1970s. Company D had increased its market shares in laser printer engines significantly in the mid-1990s; it manufactured approximately 36 million engines that were incorporated into its own brand of laser printers or in original equipment manufacturing contracts. Though equipped with only a mechanical engineering background, it proceeded step by step to acquire the component production embedded in printers. In an early stage of laser printer production, it constructed the systems around the engines, controllers and software that it procured from established suppliers. Then it acquired the manufacturing skill and obtained a license for some elements of the engine technology from an established market leader. To assure the quality of its product, Company D worked closely with its suppliers on a long-term basis. It even assigned two employees permanently to one of its UK suppliers' factories. This meant that when problems concerning component quality arose, it would not end abruptly its relationships with its suppliers. Instead, it would utilize its resources to solve the problems jointly with them. Company D found that such practice not only facilitated its quality assurance procedure, but also enabled it to negotiate reasonable profit margins on procurement.

Company E's corporate history was interwoven with the development of Japan's telecommunication infrastructure. Though its founder had experienced an early technological setback in telecommunication products, he persisted in their production as he believed, rightly, that telecommunication was a critical business area in the 20th century. Company E participated in Japan's computer industry in the 1960s; it then diversified into microcomputer products as well as the production of its components and peripherals. Company E was also involved in D-ram manufacturing. It implemented continuous improvement and its manufacturing subsidiary in the UK had been nominated for its well-established organizational practices. Despite not being associated with the cultural value of flexibility, Company E was involved in the captive production of printed circuit boards for its microcomputer systems. Actually, its General Manager in the UK mentioned the inflexibility of the operation. He pointed out that the management team in Japan tended to impose decisions on the subsidiaries, and he considered that these decisions were based on the expectations of headquarters personnel, and sometimes highlighted the cultural difference between the East and the West.

The two MNEs from Taiwan with people management capabilities were reputed in labor-intensive components. Company F and to a lesser extent Company G had considerable global market shares in monitors and keyboards. The two MNEs also forward-integrated into the final assembly of microcomputers, specializing in own-brand as well as original equipment manufacturing. Company F had always stressed that one of its strengths was human resources, and it believed that the utilization of local management expertise was indispensable for its expansion into Europe. It appointed local nationals as General Managers in its subsidiaries and granted them full autonomy in decision making. Company F had demonstrated flexible attitudes in its management operations within Europe. For instance, Company F allowed the use of a trilingual communication mode in its manufacturing plant situated in a predominantly English speaking country to lower communication costs. This contrasted with US MNEs' tendency to persist in the use of English to the extent that it encouraged its European staff to adopt anglicized names. Besides, Company F transformed one of its warehouse facilities into a manufacturing plant prior to the completion of a facility built for that purpose.

Company G had grown from a 40 people operation to approximately 1,000 employees worldwide within a decade. It emphasized the importance of trust towards its staff and as a result was able to illicit good faith from them. Its UK personnel, on one occasion, worked overtime to undertake repair work on a portable personal computer sold by its US subsidiary under an original equipment manufacturing contract. Company G also incorporated flexibility in its European operations, and expected that its employees combine flexibility with self-discipline in handling all company issues. For example, it did not perceive a need to provide clear definitions concerning the job responsibilities of its senior managers, and it did not have precise rulings on the replacement of company cars for its sales staff.

When asked about the scope of authority with its UK sales manager, Company G's General Manager replied "he can have all the authority he wants if what he does is good for the company." He also explained that there were various factors affecting the replacement of company cars for its UK sales team, e.g. total mileage, model, company financial position at the time, etc. There would not be rigid guidelines, then, for the replacement of company cars as with its US counterparts that tended to adhere to a set of standardized corporate policies.

The evidence obtained from the case studies of microcomputer component MNEs demonstrated the role of national culture towards MNEs' international competitiveness. Company A excelled in concept-intensive components where competitive advantages were consistent with their product design capabilities deriving from the core American value of individualism. Furthermore, the four Japanese MNEs' superior performance in capital-intensive components hinged on manufacturing process capabilities that in turn were based on their core values of continuous improvement. Company F and Company G's competitive advantages in labor-intensive components were based on their people management capabilities built primarily on flexibility. However, case studies also highlighted that the individualistic Company A had invested in both capital-intensive and labor-intensive components. Though it had withdrawn from labor-intensive component production, it still had market presence in capital-intensive components. In addition, Company B and Company E, which were from societies not characterized by individualism and flexibility, engaged in innovative software development and in printed circuit board production respectively. Hence, there seems to be a countervailing pattern that argues against the hypothesis. Nevertheless, Company A's expertise in capital-intensive component was in product development rather than mass production process. It should also be pointed out that Company B had been criticized by industry participants for weaknesses in its printer software as compared to that developed by US MNEs, i.e., Company B was unable to write sufficiently good software to allow for the interface with laser printers. As Company B had made considerable investments in software development, its limited achievement may possibly imply that without the foundation of the cultural value of individualism, it will always lag behind US MNEs in concept-intensive software. In the case of Company E, the counter argument is that it restructured the work process of printed circuit board manufacturing and reduced the labor content of the operation. In that sense, the greater level of mechanization may render invalid the classification of their printed circuit board production as a labor-intensive process. Alternatively, its manufacturing activity may be more appropriately classified under capital-intensive component production. Given the divergence of the component focus within Company A's capital-intensive activity and Company E's labor-intensive activity, and given Company B's weakness in concept-intensive component, it can be concluded that these counter arguments do not present a convincing case against the notion of cultural selectivity. In other words, the hypothesized relationship between national culture and national competitiveness is, to a great extent, substantiated.

Country-based Review

I will now triangulate the above findings with an analysis of secondary industry data. I will examine the competitiveness of individualistic US MNEs in the domain of concept-intensive design of microprocessors and software, the competitiveness of continuously improved Japanese and Korean MNEs in the area of capital-intensive components of D-rams and printer engines, as well as the competitiveness of flexible Taiwanese and Korean MNEs in relation to labor-intensive keyboards and monitor production.

The product design capabilities of individualistic US concept-intensive component MNEs are well established and have enabled the USA to shape the path of development in both the microprocessor and software industries. The Complex Instruction Set Computer (CISC) microprocessor architecture evolved around Intel's X86 and Pentium products. Besides, Intel was credited with inventing the first CISC microprocessor, the 4004, in 1971.(5) Other important CISC microprocessor designers are US MNEs, Advanced Micro Devices and Cyrix. Advanced Micro Devices introduced the Intel compatible 386 microprocessor approximately five years after Intel's product launch and lagged behind Intel's 486 for 3 years and Intel's Pentium for about 2 years. Nevertheless, it has, over the years, shortened its product design gap with Intel and has aimed at approaching simultaneous release of microprocessors with Intel in future. The RISC microprocessor market, which developed in the mid-1980s, was again associated with the product design capabilities of US MNEs. The RISC technology could be traced back to the work of IBM's 801 team in the T. J. Watson Research center during the 1970s; its subsequent development has continued at IBM, as well as within Sun, SGI, DEC, Motorola, and Apple.

The product design capabilities in system software and application software among US concept-intensive software MNEs have been demonstrated by their long lists of important innovations. A 1995 review indeed regarded 20 US software products which included CP/M 2.0, VisiCalc, WordStar, dbase II, AutoCAD, Lotus 1-2-3, DOS 2.0, Novell Netware, Unix System V, Mac OS and System 7, PageMaker, LANtastic, Adobe Type, Lotus Notes 3.0 ... as the most important software products in history (BYTE, 9/1995:68-70). Moreover, US MNEs are at the forefront of software development areas such as the internet software which had been estimated to reach total sales of US$8.5 billion by 1999; Internet server software made up to more than 50% of the overall Internet software in the mid-1990s and has been dominated by American providers such as Microsoft, Netscape, NCSA, and Apache (The Economist, 5/25/1996:28; Information Week, 6/2/1997:92). Finally, America's product design capabilities within concept-intensive components can further be seen in the total number of US patents obtained by its MNEs. For example, in 1997 IBM was awarded as many as 550 patents in software whereas Microsoft had 199 patents. Intel, on the other hand, obtained 407 patents for its semiconductor technology (Wall Street Journal Europe, 1/12/1998:5).

The product design capabilities of US concept-intensive component MNEs have not only resulted in generations of highly specific and highly competitive products but also enabled the designers to capture monopoly rents. The international competitiveness of the US software industry can be seen in the fact that it represented, in 1995, 82% of the revenues in fifty largest software companies (Datamation, 6/15/1996:5). More significantly, Table 2 shows that the software revenues for the five leading US MNEs in 1995 were US$93,014 million as compared to the US$9,303 million of their Japanese counterparts. In addition, the net profit margins and return on equity of the software industry, as illustrated by the Fortune 500 in 1998 which is presented in Table 3, were well above the median of other industries. The net profit margins of leading MNEs, Microsoft, Oracle and Computer Associates,' ranged from 9% to 30% as contrasted with the Fortune 500 median of 4.9%. The aggregate market shares of US software MNEs were equally overwhelming: in 1991 they shared approximately 80% of the US$54 billion packaged software market containing application software such as word processing, spreadsheets, and graphics (The Economist, 11/12/1994:102). American MNEs also dominated the system software markets. For example, US products DOS, Windows, Macintosh OS, and UNIX are the operating system software standards for microcomputers whereas Oracle, Informix, Sybase, Microsoft, and IBM are widely adopted as database management software.

TABLE 2
Knowledge-based Capabilities among leading US and Far East MNEs in

                   Software Suppliers

Nationality   Company     Software Revenues     Total IT
                          in 1995 (US$ m)     Revenues in
                                              1995 (US$ m)

American      IBM         12,949.2            71,940.0
              Microsoft    7,418.0             7,418.0
              CA           2,460.9             3,196.0
              Novell       1,897.2             2,040.0
              Oracle       1.526.9             2,707.3
                                   93,014.2

Japanese      Fujitsu     6,431.5             26,798.0
              NEC         2,322.0             19,350.0
              Hitachi     1,296.7             16,208.1
              Ricoh         334.1              1,965.5
              Oki           215.0              3,071.2
                                    9,302.6

  Software Suppliers                 Patented Products

                            Total
                          Offerings
Nationality   Company     in the UK            Examples

American      IBM          86         IBM PC DOS 7, IBM Visual
              Microsoft    49         Age..
              CA          271         Windows NT Workstation...
              Novell       15         CA-Acc PAC for Windows...
              Oracle       16         DR-DOS...
                                      Oracle 7, Oracle Financials...

Japanese      Fujitsu     17          TeamWARE, OfficePower...
              NEC          0          SOCKS, Universal Digital
              Hitachi      0          Library...
              Ricoh        0          Application software
              Oki          0          Application software
                                      Print Manager, Japanese-
                                      English Machine Translation
                                      System

Source: Datamation June 15 1996.
TABLE 3
Performance of Leading US Software MNEs

                      Net Profit Margins   Return on equity

Microsoft                          30.0%        32.0%
Oracle                             14.0%        35.0%
Computer Associates                 9.0%        24.0%

Fortune 500 Median                  4.9%        13.9%

Source: Fortune, April 27 1998.

Table 4 highlights the competitiveness of American designed microprocessors; US designs as listed in the figure were standard configurations incorporated into microcomputer products and were highly regarded in the industrial market. The leading microprocessor designer and manufacturer, Intel had a market value of US$125,741 million on 18 March 1998; its sales revenue and net profits after taxation in 1997 were US$25,070 million and US$6,945 million respectively. Intel achieved net profit margins of 27.7% and was ranked as the second most profitable company in the Fortune 500. Other than designing and manufacturing microprocessors, Intel is involved in the licensing of its patented designs to other manufacturers. Consequently, among the 76 million units CISC microprocessors incorporated into personal computers in 1996, 92% were of Intel design while Advanced Micro Devices and Cyrix's designs shared the remaining 8% of the market (The Wall Street Journal Europe, 5/30/1997:B4). As for the 9.83 million units of RISC microprocessors that were predominantly used in workstations or laser printers in 1994, the combined market shares of US MNEs--Intel, SGI, IBM, Motorola, Apple, AMD and Sun's designs-- were 81% (1CEC, 1996a: 4-28).

TABLE 4
Concept-intensive Component Capabilities among US and
Far Eastern MNEs

                                   Examples of Patented
          Designers               Microprocessor Designs

 Company         Total                RISC            CISC
            Revenue in 1995
                (US$ m)

Intel        3,240            i960                    P6
AMD          2,441            AM29000                 K5
Cyrix        2,379            --                      M1
NexGen                        --                      Nx686
SGI          2,541            MIPS R10000
IBM         71,940            RS600, Power PC620
Apple       11,378            Power PC620
Sun          6,500            UltraSparc-II
HP          26,073            PA-8000
Motorola     2,974            Power PC620, Coldfire
DEC         14,440            Alpha 21164A, Strong
                              ARM

Hitachi     16,208            SH-II
NEC         19,350            V831                    V30

Source: Various company websites (1998).

Japanese and Korean capital-intensive microcomputer component MNEs that are headquartered in societies endowed with the core cultural values of continuous improvement, on the other hand, obtained capabilities on manufacturing processes that enhanced their cost competitiveness and enabled them to capture market shares in capital-intensive components at the expense of early US leaders.(6) The manufacturing process capabilities of Japanese and Korean MNEs have been well documented. One of IBM's scientists acknowledged that Japan's "greatest technological strength" vis-a-vis the United States was "the speed with which developments are translated into improved products and processes" (Clark, Hayes & Lorenz, 1985:139). Rosenberg and Steinmueller (1988:233) also recognized that: "innovations in the IC industry have been strongly influenced by the incremental improvement of process technology.... Recent successes of Japanese IC firms in international competition have been dependent upon success at manufacturing improvement. In IC production, the proportion of workable devices emerging from the production process, production yield, is the most important manufacturing cost factor.... "This corresponded with early observations by industry participants in relation to Japanese MNEs' capabilities on quality control regarding the production process of wafer fabrication. Early US buyers of Japan's 16K D-rams, notably Hewlett Packard, reported that Japanese products had a lower frequency of defects than domestic ones. It concluded in 1978 that the rejection rate among 150,000 memory ICs shipped by three US manufacturers over a one year period ranged from 0.11-0.19% whereas the same quantity of ICs shipped from three Japanese manufacturers were all accepted (Anderson, 1980:18).

Table 5 illustrates Japanese and Korean MNEs' competitive positions in terms of worldwide market shares of D-rams in 1994. Manufacturing process capabilities have enabled Korean MNEs Samsung, Hyundai, and Goldstar to capture 27% of the US$23,420 million D-ram market while the Japanese MNEs, Toshiba, NEC, Hitachi, and Mitsubishi, had 43% of the market. TI, Micron and IBM, on the other hand, were the only US MNEs that managed to compete in the D-ram market by 1994; Figure 5 shows that TI and Micron had 12% of the D-ram market. IBM at the time was focused on captive production of semiconductors; its external sales of semiconductor devices amounted to US$970 million as against US$3,450 million of internal sales (ICEC, 1996b: 1-115). Japanese and Korean MNEs manufacturing process capabilities had also resulted in their subsequent technological leadership in D-rams. Japanese MNEs were the first to deliver the 64-kilobytes D-rams in 1978, the 256-kilobytes D-rams in 1982, the 1-megabyte D-rams in 1985 and the 4-megabytes D-rams in 1989 whereas the Korean MNE Samsung was the first to launch the 64-megabytes D-rams in 1992. Furthermore, Samsung was also the first company to complete the prototype circuit design for the 256-megabytes D-rams in 1994 as well as the 1-gigabyte D-rams in 1995. It successfully developed the 1-gigabyte D-ram in 1997 and then moved on to the next generation of 4-gigabytes product. Though the D-ram sector has, since 1998, gone through a period of consolidation, Japanese and Korean MNEs had gained significantly from their operations. For instance, Samsung increased its sales revenues by 25% on average per year from the mid-1980s through the mid-1990s. It obtained US$13 billion sales revenues in 1982 and US$70 billion in 1996. Though it has been suggested that Samsung's astounding performance can be attributed to factors including government support, cheap capital, and a protected domestic market, it cannot be denied that US MNEs did possess first mover advantage in terms of market access and technological accumulation at the early stage of Samsung's market involvement. Hence, it is difficult to account for Samsung or other Korean and Japanese MNEs' abilities to surpass US MNEs; the most convincing argument is that Japanese and Korean MNEs out-performed the early US entrants as a consequence of their culturally derived manufacturing process capabilities.

Table 5
Manufacturing Process Capabilities among Leading D-ram
Manufacturers

Nationality        Company      1994 DRAMs Sales (US$ m)

Korean             Samsung            US$3,325
                   Hyundai            US$1,525
                   Goldstar           US$1,335
                                               6,085

Japanese           Toshiba            US$2,845
                   NEC                US$2,650
                   Hitachi            US$2,325
                   Mitsubishi         US$1,355
                   Fujitsu            US   745
                                              10,100

USA                Micron             US$1,315
                   TI                 US$1,295
                                               2,610

                                     total US$23,420

Source: ICEC 1996b 2-22.

Manufacturing process capabilities have also enabled Japanese MNEs to dominate the market for laser printer engines. Table 6 shows that Japanese MNEs, Canon, TEC, and Konica, were among the widely adopted laser printer engine manufacturers. In addition, other established Japanese or Korean laser printer MNEs included Samsung, NEC, Konica, Brother and Minolta. Despite the fact that US MNEs were more experienced in the technology, due to their early exposure to the technology in conjunction with mainframe or minicomputer laser printers, Japanese and Korean MNEs have undertaken recent product improvements. Hitachi was indeed the first company to introduce and deliver the color laser printer engine for desktop printing (Edge: Group Computing Report, 11/24/1997). In addition, Japanese MNE Minolta has also continuously refined the printing mechanism one the past decade. Minolta's recent laser printer engines that were built into its PageWorks Series contained the patented conductive flexible sleeve which "eliminates toner scatter, smoothes edge and produces more solid blacks and finer edges than typical laser printer engines" (Seybold Report on Internet Publishing, 6/1997:26). It should be noted that the Japanese MNE Canon alone had approximately 70% of world market shares of laser printer engines (Financial Times, 2/16/1996:31).

Table 6
Leading Laser Printer Engine MNEs

Engine Producers                   Models

Canon (Japan)                    Canon LBP-1260,
                                 HP Laser Jet 4L,
                                 Oce 6470,
                                 Digital DEClaser 1800

TEC (Japan)                      Mannesmann T9008,
                                 Olivetti PG304

Konica (Japan)                   HP Color LaserJet,
                                 Mannesmann MT911

IBM (USA)                        Lexmark Winwriter,
                                 Memorex Telex 5032

Xerox (USA)                      Star LS-5, Apple
                                 Laserwriter Select360,
                                 Epson EPL-9000 PS

Source: What PC Buyers Guide September 1995 p16.

Finally, the core cultural values of flexibility among Taiwanese and Korean labor-intensive component MNEs enhance their people management capabilities and facilitate their competing intentionally with lower cost structures. Taiwanese and Korean MNEs' people management capabilities as illustrated by the ratio of management expenses to total sales are shown in Table 7. Management expenses are composed of selling expenses, general expenses, and administrative expenses and therefore provide an approximation of the efficiency in managing the labor inputs for every dollar of sales generated. The figure shows that the ratios for the Taiwanese peripheral MNEs are as low as 4.5% while the ratios for the diversified Korean MNEs are also relatively lower than their Japanese or US counterparts. Indeed, the people management capabilities for Taiwanese monitor and keyboard MNEs were particularly impressive since they are much smaller than Japanese and US rivals and were subjected to a lesser extent to economies of scale. Since they possess a greater level of expertise to maneuver various culturally, economically and politically related human resources issues, Taiwanese and Korean labor-intensive MNEs are more willing to explore the cost advantages of developing areas in Southeast Asia and Eastern Europe.

Table 7
People Management Capabilities of Leading Labor-intensive
Components

Monitor                   Worldwide              Management
Producers                 Ranking of            Expenses as a
                        Monitor                 % to Sales(b)
                        Sales
                          by value(a)
TAIWANESE
MNEs                            6th                     6.5%
  ADI                           7th                     4.5%
  Lite-On                       8th                    10.5%
  Tatung                        9th                     5.5%
  Acer Peripherals                              Average 6.8%

KOREAN MNEs
  Samsung                       1st                    11.4%
  Goldstar                      3rd                    15.3%
                                               Average 13.4%
JAPANESE
MNEs                            2nd                    30.0%
  NEC                           5th                    29.0%
  Sony                                         Average 29.5%

Leading                 Management
keyboard                Expenses as
producers               a % of Sales(b)

Taiwanese
MNE                         5.5%
  Acer

Korean MNE                 11.4%
  Samsung

Japanese MNE               14.8%
  Alps

US MNE
  Key Tronic               12.6%

Notes:

(a.) Based on Computer Reseller News. Jun 2 1997 p130.

(b.) Management expenses referred to the selling and general
administrative expenses reported in the Company Annual Reports.
Management expenses as a % to sales were calculated mainly from
the 1997 Annual Reports. However, the ratio for Lite-On
was based on 1994 Annual Reports while the ratios for ADI,
Acer, Alps, Samsung and Tatung were 1996 figures.

(c.) The ratio for Goldstar's figure confined to its
electronics business.

Source: Various company annual reports.

The competitiveness of Taiwanese and Korean labor-intensive component industries as based on their people management capabilities are reflected in their worldwide market shares. Table 7 shows that seven of the top ten of the globe's leading monitor producers in terms of the value of production in 1997 originated either from Taiwan or from Korea. In fact, the Market Intelligence Center from Taiwan reported that Korea manufactured in 1995 18.4 million units of monitors, while Taiwan produced 37.7 million units. In addition, Japan who specialized in high quality large screen models supplied approximately 6.2 million units (CeBIT, 1996). It has also been estimated that Taiwanese MNEs supplied as many as 32.8 million units to the worldwide keyboard market in 1995, with its largest producer, Acer Peripherals, contributing 4 million units to Taiwan's aggregate output ([CHINESE TEXT NOT REPRODUCIBLE IN ASCII], 7/19/1996:6).

The above evidence, in conjunction with the 4-nation microcomputer component MNEs, corroborates the Cultural Selectivity Hypothesis. However, it has also been shown that MNEs lacking the core cultural values of individualism, continuous improvement, and flexibility also engaged in concept-intensive components, capital-intensive components and labor-intensive components activities. Table 2 demonstrates that Japanese MNEs Fujitsu, NEC, and Hitachi were involved in concept-intensive software development and attained revenues from their activities. In addition, NEC was known for its microprocessor design activity in Japan. In the area of capital-intensive components, Table 4 and Table 5 also demonstrate US MNEs' involvement in D-rams and laser printer engines, which contradicts with the prediction of the Cultural Selectivity Hypothesis. Finally, US and Japanese MNEs' engagement in the production of monitors and keyboards as displayed in Table 7 also serve to reinforce the arguments against the hypothesis. In light of these counter-arguments, it seems difficult in its entirety to accept the evidence in favor of the Cultural Selectivity Hypothesis. Despite these counter-arguments, however, the following points should be considered. First, though Japanese MNEs diverted noticeable resources towards software design, their achievement has been limited. Furthermore, NEC's patented microprocessor design was mainly applied to its microcomputers. In fact, NEC abandoned its proprietary design and switched to US design microprocessors in the 1990s. Second, in response to the recent collapse of D-ram prices, US MNEs TI announced its withdrawal from the market and sold its semiconductor business to Micron in 1998 (Financial Times, 6/23/1998:35). The more efficient Far East MNEs, on the other hand, rescheduled their production targets and remained in businesses. This therefore illustrates the relative competitiveness of the US and the Far East D-rams producers. Third, Japanese MNEs, which are not equipped with the core cultural value of flexibility compete in a different segment of the labor-intensive component. The Japanese monitor producers, NEC and Sony, specialized in larger screen and higher resolution monitors. These additional points therefore provoke reinterpretation of the results where MNEs which lack the core cultural values of individualism, continuous improvement, and flexibility participated in the respective concept-intensive, capital-intensive and labor-intensive component areas.

Discussion

The primary and secondary data has sustained the concept of cultural selectivity and advocated a cultural embeddedness perspective of microcomputer component capability. The cultural impact on competitive success is further exemplified in the difficulties encountered by selected MNEs' attempts to acquire the capabilities that did not correspond with the core cultural values of their parents. To reiterate this point, despite the attempt of Far Eastern MNEs' to attain product design capabilities by undertaking foreign direct investment in the Silicon Valley, US MNEs still have daunting leaderships in most of the software markets. NEC, which set up the NEC Systems Laboratory Inc in San Jose in 1972, though achieving considerable success in software, is still behind individualistic American MNEs in application software, communication software, and system software. Similarly, even though leading US keyboard MNEs such as Key Tronic or Cherry have manufacturing facilities in low cost countries, their early domination in keyboard in terms of the volume of production has been eroded by later Far Eastern entrants. Parallel to this, Japanese labor-intensive component MNEs, who were highly competitive in the 1970s, were also replaced by Korean and Taiwanese MNEs that were endowed with the core cultural values of flexibility.

Furthermore, the nation-specific property of knowledge-based microcomputer component capabilities can be seen in the relocation of microcomputer component MNEs. US software MNEs are in the process of migrating their design bases to countries such as India, Singapore, the Philippines, and Russia whereas Taiwanese keyboard and monitor MNEs had been successfully relocated to other low cost countries such as Thailand, Malaysia, Indonesia and China since the late 1980s. It was reported that 40% of Taiwan's keyboards and approximately 20% of its monitors were manufactured outside Taiwan in 1995. Specifically, the distribution of Taiwanese monitor MNEs' overseas production bases in the mid-1990s were 45% in Malaysia, 25% in Thailand and 22% in Indonesia ([CHINESE TEXT NOT REPRODUCIBLE IN ASCII], 7/19/1996:6 and MIC, 1996). The cultural embeddedness of microcomputer component capabilities is therefore sustained.

More interestingly, national culture not only affects the competitiveness of microcomputer component MNEs, but also serves to illuminate their strategic directions. The individualistic US MNEs, equipped with product design capabilities, are competing with innovative designs in labor-intensive products. For example, Microsoft, Cherry, and Key Tronic have been actively involved in the redesign of keyboards. Key Tronic, for instance, held approximately 40 patents for keyboard technology in 1996 (Keytronic Website, 1996). Key Tronic is also developing next-generation keyboards with fingerprint recognition and smart-card reader capability (Edge: WorkGroup Computing Report, V8, 7/28/1997:42). Another relatively young keyboard manufacturer, the San Francisco company, Darwin Keyboards, also launched the patented SmartBoard(7) in January 1998 to solve both the wrist and the finger problems inherent in conventional keyboards. The SmartBoard keyboards contain two position adjustable wrist levers as well as keys that match the natural fanning movements of fingers. Parallel to this development is the continuously improved Japanese companies' attempts to tackle software development with manufacturing process capabilities. Cusumano (1991:9) found that Japanese software company managers believed that "they could structure and improve software operations and achieve higher, or more predictable, levels of productivity, quality, and scheduling control. They also acted on this conviction, demonstrating a long-term commitment to process improvement...." Cusumano (ibid) also identified that the Japanese adopted formal policies to encourage programmers to reuse their colleagues' codes and designs. The process of recycling old codes in new designs, to a great extent, reflects their entrenched mentality of manufacturing process improvement.

US labor-intensive keyboard MNEs and Japanese concept-intensive software designers' strategic inclinations can be further analyzed in terms of their abilities in enhancing or impeding their international competitiveness. The trajectory of US MNEs' strategies has generated new opportunities and aspects for their competition in keyboards, which are increasingly being subjected to a stringent regulatory framework and hence demand greater levels of concept-intensity. The tendency of Japanese software MNEs to compete with process capabilities, however, is diametrically opposed to concept creation and hence is presented as a possible liability in their pursuit of product innovation.

Finally, the finding of this study must be reviewed in its single industry dimension. However, as the microcomputer component is a segment of the electronics industry, the utilization of the results in other areas of the electronics industry on the basis of their standardized assembly process and shared components can therefore be justified.

CONCLUSION

This paper has shown that the Cultural Selectivity Hypothesis provides a comprehensive support for the persisting variance of international competitiveness among US concept-intensive microcomputer component MNEs, Japanese and Korean capital-intensive microcomputer component MNEs as well as Korean and Taiwanese microcomputer component MNEs on the grounds that US product design capabilities are consistent with the core American value of individualism, Japanese and Korean manufacturing process capabilities are consistent with their core cultural values of continuous improvement and Taiwanese and Korean people management capabilities conform to their core cultural values of flexibility. The implication for microcomputer component MNEs is that they should learn carefully and selectively. As they can be proficient in the component technology where the required capabilities are compatible with their core cultural values at relatively lower costs, they should initiate product strategies that accentuate their strengths. It is also suggested that only those competitive advantages that are derived from the appropriate core cultural values can be sustained in the international market in the long run; hence MNEs should reorientate themselves to product areas that capitalize on their national cultures.

Future research can continue to explore the importance of the relationship between national culture and capability acquisition in other areas of technology industry. In particular, a time frame can be incorporated into the study of competitive advantages in the concept-intensive component, an area that is important in the information age and also carries significant profit implications. As capability in software and microprocessor design is built upon innovative capability associated with individualism, it is suggested that the Americanization of Far Eastern societies that subsequently erodes their traditional collective values may enrich Far Eastern MNEs' ability to compete in concept-intensive component areas.

NOTES

(1) Korean MNEs refer to those from South Korea.

(2) Microcomputers refer to personal computers, workstations and desktop laser primers.

(3) "International competitiveness" may refer to either static competitiveness or dynamic competitiveness. In this paper, static competitiveness as approximated by the relative market share of the 4-nation within a component product category in 1995 is utilized.

(4) Convergent thinking follows strict logic and is directed towards a single right answer to a problem while divergent thinking is intuitive, unconventional, and involves sudden ideas. Japanese business consultants such as Umeda have criticized the Japanese culture and convergent thinking and written that "Japanese management is incapable of articulating a clear vision or providing leadership in the style of a US chief executive ... and will have to give up its parochialism, its perverse egalitarianism, its in-group orientation and its tendency to suppress individuality and creativity" in order to achieve success in the computer industry (Financial Times, 8/19/1994:10).

(5) It should be noted that the US Patent Office has also granted a basic patent for the microprocessor to Gilbert Hyatt from California who "has spent 20 years fighting for the patent. The patent covers Hyatt's invention in a makeshift laboratory in his home in 1968" (Computing, 9/6/1990:2).

(6) Intel launched the first advanced 1-kilobyte D-rams; nevertheless, by 1983 it was driven out of the D-ram market, IBM and Xerox, on the other hand, were early producers of laser printer engines used in mainframes and minicomputers.

(7) US Patent 5,336,001 and D348, 877.

REFERENCES

Anderson, R. W. (1980). The Japanese success formula: Quality equals the competitive edge. Verbatim record of a seminar hem in Washington DC, March 25 1980, under the sponsorship of the Electronic Industries Association of Japan.

Bain, J. S. (1956). Barriers to new competition. Cambridge, MA: Harvard University Press.

Casson, M. (1995). Entrepreneurship and business culture. Aldershot: Edward Elgar.

CeBIT (1996). Speech presented by Lee, K. Y. Development of display industry in the Asia-Pacific, CeBIT Germany.

Chang, C. S., & Chang N. J. (1994). The Korean management system: Cultural, political, economic foundations. Westport, CONN: Quorum Books.

Clark, K. B., Hayes, R. H., & Lorenz, C. (1985). The uneasy alliance. Boston: Harvard Business School Press.

Cusumano, M. A. (1991). Japan's software factories: A challenge to US management. Oxford: Oxford University Press.

Ferguson, C., & Morris, C. (1993). Computer wars: The fall of IBM and the future of global technology. New York: Random House, Times Books.

Finan, W. F., & LaMond, A. M. (1985). Sustaining US competitiveness in microelectronics: The challenge to U.S. policy. In B. R. Scott and G. C. Lodge (Eds), US competitiveness in the world economy. Boston: Harvard Business School Press.

Garrahan, P., & Stewart, P. (1992). The Nissan Enigma. London: Mansell.

Hofstede, G. (1980). Culture's consequences: International differences in work-related values. London: Sage.

Hong, Y. S. (1995). Technology transfer: the Korean experience. Korea's Economy, 11, 43-51.

Hudson, L. (1966). Country imaginations. Harmondsworth: Penguin.

ICEC (1996a). Microprocessor 1996. Scottsdale, AZ: Integrated Circuit Engineering Corporation.

ICEC (1996b). Memory 1995. Scottsdale, AZ: Integrated Circuit Engineering Corporation.

Jones, J. C. (1970). Design methods: Seeds of human futures. London: Wiley-International.

Khera, D., Cressewell, M. W., Loren, W., Ramanathan, G., Buzzeo, J., & Nagarajan, A. (1994). Increasing profitability and improving semiconductor manufacturing throughput using expert systems. IEEE Transactions on Engineering Management, 41(2), 143-150.

Lieberman, M., & Montgomery, D. (1988). First-mover advantage. Strategic Management Journal, 9 (Special Issue), 41-58.

MIC (1996). [CHINESE TEXT NOT REPRODUCIBLE IN ASCII]. Taipei: Market Intelligence Center, Institute for Information Industry.

Ohlin, B. (1933). Intra-regional and international trade. Cambridge, MA: Harvard University Press.

Penrose, E. T. (1959). The theory of the growth of the firm. Oxford: Basil Blackwell.

Porter, M. (1990). The competitive advantage of nations. London: Macmillan.

Prahalad, C. K., & Hamel, G. (1990) The core competence of the corporation. Harvard Business Review, 6(3), 79-91.

Ricardo, D. (1817). Principles of political economy. Reprinted as vol. 1 of P. Sraffa (Ed). The works and correspondence of David Ricardo, 1951. London: Cambridge University Press.

Rosenberg, N., & Steinmueller, W. E. (1988). Why are Americans such poor imitators? American Economic Review Papers and Proceedings, 78(2), 229-234.

Vernon, R. (1966). International investment and international trade in the product cycle. Quarterly Journal of Economics, 80, 190-207.

Wenerfelt, B. (1984). A resource-based view of the firm. Strategic Management Journal, 5(2), 171-180.

Wang, K. T. (1976). A commentary on the art of warfare. Taipei: Chung Man.

Denise Tsang (PhD, University of Reading) is a lecturer in International Business at the Bristol Business School. Her current research interests include comparative management and business strategy.