Great Strategy or Great Strategy Implementation — Two Ways of Competing in Global Markets

On 19 April 1775, British troops (Redcoats) marched toward Concord, Massachusetts, to destroy military stores that had been collected by the American revolutionaries (Minutemen). At Lexington Green, a large, flat, open area, the Redcoats met the Minutemen in the first battle of the day. Both sides employed a similar battle strategy, firing at each other in the open from closed ranks. It was the dominant battle strategy of the day. The Redcoats quickly prevailed, and the Minutemen dispersed after a few volleys and a number of casualties.

Later in the day, as the Redcoats returned to Boston, a second battle developed. At various points along the road, Minutemen fired upon the Redcoat formations from inside houses and behind stone fences. When the Redcoats charged these positions, the Minutemen withdrew into the countryside and reappeared farther down the road. The Minutemen’s skirmish tactics took a heavy toll on the massed and extremely vulnerable Redcoats, who could do little but set fire to the buildings the Minutemen had already abandoned.

The two battles between the same opponents on the same day produced two very different results. It seems reasonable to suggest that between the two battles the Minutemen made a fundamental shift in their battle strategy, and this new strategy, rather than luck or chance, produced the significant difference in results.

Like the Minutemen, competitors frequently don’t compete the way one expects them to. This is one of the key difficulties for managers trying to understand, prepare for, and manage global competition. Western firms, and U.S. firms in particular, generally try to compete through some kind of strategic advantage. That is, they often try to develop a unique business strategy that will allow them to outmaneuver competitors. Yet many global Asian firms appear to compete successfully without much attempt to develop distinctive business strategies. Instead, they try to implement not-so-unique business strategies better than competitors and thereby to gain competitive advantage.

In this paper, I will distinguish between these two modes of competing, discuss some of their implications, and illustrate my analysis with research conducted in U.S. and Japanese semiconductor companies.1 I will then consider what happens when these two modes confront each other in a global market and discuss some of the implications for U.S. competitiveness in global markets.

In order to clarify the difference between competing through strategy and competing through strategy implementation, let’s return to the Redcoats and the Minutemen and analyze the differences between the two battles more conceptually. Figure 1 shows that in the first battle the two sides followed a similar battle strategy and, therefore, differed only in how well they implemented it. The Redcoats clearly had the edge with their superior capability and greater experience in executing the conventional battle strategy.

Figure 2 shows what happened in the second battle. The Redcoats continued to follow conventional battle strategy, which called for retaining relatively closed ranks and centralized direction. Their implementation capability was presumably similar to what it had been earlier in the day at Lexington Green, thus their strategy and implementation capabilities in the second battle (R₂) were similar to what they had been in the first battle (R₁). The important difference in the second battle was that the Minutemen shifted up the vertical axis and adopted a different — and superior — battle strategy (M₂) They changed to a skirmish strategy, which was ideally suited to the winding road and broken countryside where the second battle took place. Although not shown in Figure 2, it is probable that their implementation capability also improved (shifted to the right). It was the different and superior battle strategy, however, that allowed the Minutemen to overcome the Redcoats in the second battle.

The Minutemen had two basic options for trying to defeat the Redcoats. First, they could have done what most armies of the eighteenth century would have done: stuck with the conventional strategy but tried to execute it better. This might have required increasing the number of troops or making available more weapons and ammunition (essentially committing more resources). Or it might have called for more training and practice so that the troops could reload quicker, shoot more accurately, and hold their ranks under fire. Officers might also have been trained to better direct and coordinate battle tactics (essentially getting the most out of available resources).

The Minutemen chose a second approach: to select a different and superior battle strategy. A strategy is superior to the extent that it better fits the competitive environment and the organization’s capabilities to implement it. In this case, the skirmish strategy was ideally suited to the terrain (winding road, broken countryside) and to allowing the Minutemen to exploit an important enemy weakness (the absence of cavalry). It also capitalized on the Minutemen’s superior knowledge of the locale and downplayed their lack of training and experience with conventional battle tactics.

Competition in the Global Semiconductor Industry

Now let’s examine a more recent competitive situation — the battle for supremacy in the global semiconductor industry — and discuss it in terms of the strategies and strategy implementation capabilities of the two major competitor groups — U.S. and Japanese semiconductor firms. This examination is based on a field study of eight U.S. and eight Japanese semiconductor firms. Data was collected through a series of in-depth structured and semistructured interviews with each firm’s general manager, research and development (R&D) manager, manufacturing manager, and marketing manager. Interviews focused on the business and functional strategies of each company’s semiconductor business and the organizational arrangements and managerial practices used to implement them. The appendix describes the methodology and the sample companies more fully.

The case of the semiconductor industry is a familiar one, but a brief review may be helpful.2 U.S. companies dominated in the 1950s and 1960s, as they developed most of the new product and process technology. During the 1960s and 1970s, there was an active transfer of product and process technology from the United States to Japan. In the late 1970s and early 1980s, Japanese firms cooperating under government-sponsored research successfully developed the technology needed to produce very large scale integrated (VLSI) chips. Subsequent large scale investments in plant and equipment and concerted efforts within firms to improve process technology led to a Japanese advantage in terms of quality and productivity over most U.S. firms by the mid-1980s. However, this advantage was confined to standardized, high-volume products, primarily memory chips. U.S. firms continued to develop most product innovations and to dominate the newer, more innovative product markets. By the late 1980s, Japan had taken sufficient market share away from U.S. firms to become the industry leader. Serious trade friction had developed between the two countries, and Americans feared that the United States might be unable to compete successfully against Japanese semiconductor firms.3 It was during this period that the study was conducted.

Distinct Competitive Styles

Of the various aspects of strategy measured in the study, five characteristics emerged that, taken together, suggest that U.S. and Japanese firms have two different styles for competing in the semiconductor industry. Table 1 summarizes these characteristics for the two groups. Although no generalization holds across all the sample firms in each group, the patterns were remarkably distinctive, and secondary data available on other U.S. and Japanese semiconductor firms tends to be consistent.

Unique Product-Market Advantage

U.S. semiconductor firms generally seek to differentiate their products from those of competitors through innovative product design. As a product line matures and becomes more difficult to differentiate from similar competing products, a U.S. firm attempts to bring out the next generation of the product with improved performance characteristics. The classic example of this has been Intel’s sequential development and launching of the 286, 386, and 486 lines of microprocessors. Another common U.S. mechanism for product differentiation is to develop families of chips that work together to satisfy systems applications. Then, by regularly developing peripheral chips that extend the family and its applications capability, a company can continually maintain some uniqueness for the product family. It is therefore not surprising that the semicustom design techniques that permit the economical design and manufacture of application specific integrated circuits (ASICs) were primarily developed by U.S. firms. These techniques figure prominently in the strategies of nearly all firms.

The degree to which U.S. firms realize product differentiation varies. Firms with more innovative product research, such as Intel, realize it to a greater degree than firms such as National Semiconductor or Texas Instruments, which have more incremental product development capabilities. Yet nearly all firms try for some kind of uniqueness, even if it’s only a relatively minor performance characteristic that might make the product more attractive for certain applications. Thus, the dominant strategy by far among U.S. firms has been product differentiation, and very few firms have made low cost their primary competitive strategy.4

Japanese firms, on the other hand, place considerably less emphasis on developing differentiated products. The major producers all attempt to offer broad product lines that concentrate on covering the higher-volume market segments. Smaller producers have more of a niche strategy, usually emphasizing products that impart some special advantage to the company’s systems products and, therefore, that tend to be more unique in the marketplace. Even these companies, however, produce large quantities of basic products, such as memory chips and standard logic, in direct competition with the broad product line firms. The rationale is that these basic product volumes are required for manufacturing economies of scale and, in the case of DRAMs (dynamic random access memories), for staying near the forefront of process technology. Thus, Japanese strategies can be generally characterized as either: (1) a broad product line strategy heavily oriented toward basic products with low differentiation from competitors’ products (e.g., NEC, Toshiba, and Hitachi) or (2) a niche strategy coupled with a limited line of basic products with low differentiation (e.g., Sony, Sharp, and Sanyo). As a result, the dominant competitive strategy for Japanese firms has been based on high quality and low price, with much less emphasis on product differentiation.

Withdrawal Decisions

U.S. firms tend to withdraw relatively quickly from products and market segments where they cannot gain or maintain sufficient differentiation. An example was the widespread withdrawal of U.S. firms from the DRAM segment of the memory market during the latter half of the 1980s. Despite the fact that many technical managers believe that a viable presence in this product area is critical to staying at the forefront of process technology, all of the U.S. merchant semiconductor firms except Texas Instruments and Micron Technology strategically repositioned themselves out of this highly competitive segment, in which the only available strategy was low price.

In contrast, Japanese companies seldom withdraw from a significant product line or market segment even if it is losing money. If a company views the product or segment as important, it will focus on actions that will restore profitability, even if this requires a very long time. The study found that Japanese firms tend to consider a product or segment important if (1) it supports the wider company strategy of vertical integration, (2) it is large or will grow rapidly, and (3) a reasonable number of internal or external customers have come to depend on the company for it. As a result, there are substantially higher barriers to making withdrawal decisions in Japanese companies than in U.S. firms. Strategically repositioning oneself by deliberate withdrawal from unprofitable products or markets is an option much more available to U.S. managements than it is to Japanese managements.

Role of Process Technology

The U.S. firms vary considerably with regard to the importance they attach to process technology (the technology that transforms product designs into products). Because semiconductor manufacturing processes are complex, capital intensive, and still evolving at a rapid pace, there is considerable room for difference in firm capabilities. The strategies of some U.S. firms, such as Texas Instruments and IBM, call for staying at the forefront of process technology as an important part of competitive advantage. These tend to be the larger firms. For firms that rely most heavily on product-market differentiation, the role of process technology is less important, and process technology can be as much as two or three years behind the leading edge.

By contrast, both the large broad product line and smaller limited product line Japanese firms pursue strategies that call for staying at or near the forefront of process technology, both in terms of technology development and capital investment. The cost of these investments would be beyond most niche strategy or limited-line U.S. companies. The cost is borne more easily in Japan because the limited-line producers tend to be divisions of very large electronics systems companies that can absorb the research expense and provide the capital. In addition, all of the limited-line companies have entered and maintained a significant presence in a number of high-volume basic product markets. These products provide both the high volumes of production necessary to drive down the technology learning curve and the economies of scale necessary for optimum production facility size.

Sources of Process Technology

U.S. and Japanese firms frequently have different strategies regarding the source and process for acquiring new process technology. Many U.S. firms depend to a considerable extent on external sources, concentrating their internal R&D efforts on circuit design in order to support strategies that emphasize unique or differentiated products. Equipment manufacturers are an important source of new process technology for large and small firms. Many smaller firms depend almost entirely on such suppliers for process technology innovations. Larger firms often collaborate closely with an equipment manufacturer during the development stage of new equipment. Several major industry groups (Microelectronics and Computer Technology Corporation, Semiconductor Research Association, and Sematech) also facilitate collaboration on process technology research. Typically, this work is several years from the commercialization stage.

Another external source available to U.S. firms is purchasing or trading process technology with other semiconductor firms. This is often done in conjunction with a transfer of product technology. Firms will license other firms to produce certain products (e.g., Intel has licensed numerous firms to produce its famous sixteen-bit microprocessor). When product designs are transferred to the licensee, some process technology may also have to be transferred in order for the licensee to properly manufacture the product. Sometimes licensees will convert the designs to run on newer technologies (for example, from an NMOS to a CMOS technology) and, as part of the agreement, assist the licensor in also converting the product to the newer technology. Such technology transfer, however, tends to operate behind the forefront of new process technology.

The relative importance of external sources for new process technology varies widely across U.S. firms. Some large firms, such as IBM and Texas Instruments, have extensive internal R&D operations and even equipment-building operations committed to keeping the company at the forefront of new process technology. Other relatively large firms, such as Intel, pursue strategies that emphasize product technology innovations and that rely on external sources for most new process technology. The many small U.S. firms also pursue this latter strategy. Thus, although there is considerable variance, most U.S. firms are quite dependent on external sources for process technology improvements.

In the past, Japanese firms depended heavily on external sources for new process technology (primarily U.S. semiconductor firms and equipment manufacturers), but today they rely most heavily on internal development. Among the eight Japanese sample companies, only two of the smaller firms indicated an interest in collaborating with other semiconductor firms in developing new process technology. Since the study, significant increases in the cost of developing next-generation CMOS technology appear to have forced even the largest Japanese firms into collaboration. Agreements now exist between Hitachi and Texas Instruments, Toshiba and Motorola, and Toshiba and Siemens. Generally, Japanese firms consider leading-edge process technology the key strength of their strategies, and, consequently, they have little incentive to collaborate with other firms pursuing similar strategies (that is, other Japanese firms).

The importance of government-sponsored research projects, which earlier played an important role in improving the process technology of Japanese firms, has declined considerably. The rapid buildup in company-sponsored research has greatly reduced its relative importance. Although nearly all companies indicated an interest in continuing to participate in long-range, government-sponsored projects as a source of technology stimulation, several raised questions about the value of such programs in the future. Not only has company-sponsored research eclipsed government-sponsored research in terms of magnitude, but in some cases, the time horizons of company-sponsored research have started to exceed those of the government-sponsored projects. Thus Japanese semiconductor companies seem to be much more self-sufficient than U.S. firms when it comes to new process technology.

Importance of Vertical Integration

Semiconductor products are the basic components for all electronics systems products, from radios to computers. Virtually all significant Japanese semiconductor producers are subunits of larger electronics systems products companies. Original entry into semiconductor products was generally motivated by the desire to gain more control over the components that made up the company’s final products. This vertical integration strategy is still very strong in Japanese firms, even though internal sales typically account for only 20 percent to 25 percent of total semiconductor sales. Japanese firms still tend to manufacture many of the key semiconductor components that go into their final products, even when such components are readily available in the marketplace at prices competitive with the cost of internal manufacture.5

Many U.S. semiconductor firms are independent companies, dependent on the merchant market for all of their sales. Such firms are obviously not motivated to make products because they further a strategy of vertical integration and reduced external dependency. Instead, they are driven by strategies of low cost or product differentiation. Even where U.S. semiconductor producers belong to larger companies (such as Texas Instruments and NCR) that use semiconductors in their end products, the vertical integration strategy is somewhat different than in Japanese companies. In U.S. firms, these vertical integration strategies are largely confined to supplying user divisions with semiconductor products that contribute to the uniqueness of their electronics systems products. Excluding the few semiconductor firms that are solely captive suppliers (such as IBM), it is unlikely that the semiconductor division will manufacture and supply standardized, basic products to the systems products divisions unless it is already making these products for the external market. Thus, in U.S. firms, vertical integration strategies focus largely on creating systems uniqueness, with less emphasis on increasing self-sufficiency and reducing external dependency.

To summarize, U.S. firms have strategies that tend to emphasize product differentiation and that seek to capture some unique product-market advantage. They place relatively less importance on process technology and, consequently, avoid competing on low cost. This kind of business strategy has contributed to an unusually high degree of strategic diversity in the U.S. industry, with most firms pursuing niche strategies and even the broad product line firms varying considerably in their product strengths and important market segments. Firms tend to withdraw relatively quickly from products and markets with too much competition or when they miss a technological change. As a result, price competition tends to be relatively moderate across much of a company’s product range.

Japanese strategies, in contrast, emphasize forefront process technology, with only a secondary and much lower emphasis on product differentiation and unique product-market advantages. Low cost is the most important basis for competing. This kind of business strategy has led to a relatively low degree of strategic diversity among Japanese firms. Even the companies with niche strategies tend to compete with the broad product line producers on price for much of their sales volume. Because there is little tendency to withdraw from products or markets, competition over price in the standardized, basic product areas is intense.

These differences translate into differences in competitive behavior. U.S. firms generally attempt to manage competition through strategic repositioning or by creating a more unique strategy. Japanese firms tend to compete against each other using very similar strategies. They do not withdraw or try to avoid or reduce direct competition through strategic repositioning, as U.S. firms do. This low strategic diversity (or high strategic similarity) means that firms cannot readily compete with each other by having a superior strategy (of more attractive products or markets). As a result, Japanese semiconductor firms compete not so much at the level of strategy as at the level of strategy implementation.6 They are the Redcoats and U.S. firms are the Minutemen from our earlier example.

Other comparative frameworks might also be used to explain these differences. For example, these differences might be viewed as a battle between innovation and operations management or between product technology and process technology. It is undoubtedly true that many superior business strategies in the semiconductor industry are currently based on innovation or some unique product technology. But they could also be based on other differential advantages. For example, brand image, distribution channels, and access to low-cost raw materials or other inputs are differential advantages that firms in other industries frequently use to build superior business strategies. As the rate of technological change in semiconductors declines, one would expect such factors to increasingly replace technological innovation and unique product technology as important differential advantages in firms’ business strategies. In any case, striving for technological innovation and unique product technology is simply one way of attempting to compete with a differentiated and superior business strategy. Similarly, incrementally improving process technology and operations management is one way of seeking to compete with superior strategy implementation. The concept of a distinction between superior strategy and superior strategy implementation is broader than the alternative comparative frameworks and thus more useful from a strategic management perspective.

This analysis provides a somewhat different perspective than has previously been used to understand competition between U.S. and Japanese semiconductor firms.7 Most research studies and articles have attributed the loss of U.S. competitive position in semiconductors to the following factors: (1) Japanese trade barriers, (2) dumping (selling below cost) by Japanese firms in export markets, (3) unfair targeting of the U.S. semiconductor market by Japanese firms supported by the Japanese Ministry for International Trade and Industry (MITI), (4) inadequate spending on research and capital investment in the United States, and (5) the lack of U.S. industrial or technology policy.8 Although such industry-level economic and political factors are important contributors to the competitiveness problem, my study suggests that there may be another dimension to this problem — the way firms themselves attempt to compete.

Advantages and Disadvantages of the Competitive Styles

Each competitive style has advantages and disadvantages, both for an industry’s firms and for its customers, suppliers, and regulators (see Table 2 for a summary). The first three conditions largely describe how industry structure might be expected to vary depending on which competitive style is most prevalent. The last three conditions outline the relative advantages and disadvantages firms face when they select either of the two competitive styles.

Strategic Variety

When firms in an industry attempt to compete by possessing strategies that are superior to those of their potential competitors, the result is a variety of different strategies. These differing strategies produce a wide variety of products and services and numerous strategic groups. When firms in an industry compete through superior strategy implementation, a high level of strategic similarity occurs. These firms provide products and services that are already successful in a market, but they make them better (cheaper or of better quality). Thus, industries dominated by firms competing through superior strategy will provide customers with a wider range of products than industries dominated by firms concentrating on strategy implementation. The latter, however, will provide more direct competition with lower prices and higher quality.

Industry Concentration

The dominant competitive style is likely to influence the degree of industry concentration. Competing through superior strategy creates a richer competitive environment full of differentiated niches, thus making room for many successful competitors. Firms don’t need to be large or efficient to survive, just sufficiently different. When competition occurs largely around strategy implementation, however, there is no place to hide from direct competition. Even relatively small differences in implementation capability can lead to long-run cost and quality advantages that will tend to drive less capable competitors from an industry once competition sets in (and once supply exceeds demand).

Survival Patterns

Survival patterns should also look quite different between the two cases. Competing through superior strategy favors innovative firms that can pick and even create growing niches. Entrepreneurial firms are likely to dominate such industries. But just as innovation and strategic maneuvering create new niches, they also destroy old ones. This contributes to a more dynamic industry in which firms can rise and fall rapidly and industry leadership is subject to considerable change. Competing through strategy implementation favors the efficient as opposed to the innovative competitor. Because efficiency within firms is likely to be more predictable and stable than innovation, the relative positions of firms in such an industry will change more slowly. As a result, industry leadership should be more stable.

These three conditions describe how competitive style can influence an industry’s structure. The advantages and disadvantages of industry structures are of particular interest to customers, suppliers, regulators, and potential competitors. The next three conditions focus more on the advantages and disadvantages of competitive style for competing firms.

Stage of Product Life Cycle

Competing through superior strategy works best in the early stages of a product life cycle, when technological and environmental change tend to be rapid and a wide variety of feasible strategies exists. In later stages of a product life cycle, technological and environmental change are slowing and a limited number of dominant product designs have evolved. Firms have less opportunity to come up with a new design that the market will support and, hence, less opportunity to develop a strategy that differs from those already in place. These circumstances imply that more firms will be competing with similar strategies and that difference in strategy implementation rather than difference in strategy will tend to determine the winners.

Technological and Environmental Change Risk

The kind of business risk a firm faces also varies with the type of competitive style it employs. Firms attempting to compete through superior strategy are most vulnerable to losing their strategic uniqueness. Their risk is greatest when the rate of technological and environmental change level off, making it more difficult to create new bases for uniqueness. They are especially vulnerable when efficient imitators enter their strategic domains and change the basis of competition to one of strategy implementation.9 This is what happened in the DRAM (memory) market of the U.S. semiconductor industry.

The creation of innovative technology in the semiconductor industry often creates only a temporary differential advantage. One reason is that new technologies associated with standard products and standard manufacturing processes are often quite “leaky.” Ouchi and Bolton distinguish between private intellectual property (technology) and leaky intellectual property.10 Private intellectual property is property that a private party can practically or legally appropriate and retain or transfer at will. Examples in the semiconductor industry are patentable product designs that cannot be readily bypassed or engineered around. Leaky intellectual property can be effectively appropriated by a private party but only for a short time. Soon others will either duplicate or bypass the new technology and eliminate the creator’s ability to earn a return on it. Leaky technologies are primarily associated with standard products, such as designs for the next generation of DRAMs, and standard manufacturing processes, such as moving process technology from a one-micron technology to a sub-micron technology.

It is apparent that leaky technology must be frequently replenished or recreated in order to remain a differential advantage. In highly competitive industries, this is often difficult, and it becomes increasingly difficult and costly when the technology curve levels off. Even when technology is not leaky, the licensing of proprietary technology in the semiconductor industry is widespread for a number of reasons: it is a way of establishing a product as an industry standard, it satisfies customer demands for second sourcing, and it avoids inadvertent patent infringements and legal suits. As a result, new technology development in the semiconductor industry often provides only a temporary differential advantage.

Conversely, firms used to competing with superior strategy implementation face the greatest risk when the rate of technological and environmental change increases. Such conditions create opportunities for technological breakthroughs and other changes that can become the bases for new strategies. These new technologies are often sufficiently different to be more proprietary and less leaky than the mature technology they replace. This hinders the strategy implementer from adopting the new strategy and might eliminate the advantage the implementer presently realizes from its superior ability to implement the old strategy (much as the Minutemen’s new strategy eliminated the superior implementation capability of the Redcoats in the second battle).

For example, many believed that improvements in ASIC technologies might allow U.S. firms to outflank Japanese dominance in standard products. This would make Japan’s superior strategy implementation capability less relevant. This belief has waned, however, as growth in ASICs has been slower than anticipated (they still account for only 20 percent to 25 percent of the market).

Profit Margins

Finally, firms able to compete with a superior strategy should encounter less direct competition and higher profit margins than firms competing with similar strategies and attempting to excel through strategy implementation.

Thus competitive style has a number of important implications for both firms and industries. This comparison reveals that the two modes of competing tend to work best under different conditions and tend to provide different kinds of advantages.

Managerial Requirements for the Competitive Styles

In this section, I will consider the different managerial requirements that the two competitive styles demand (see Table 3 for a summary).

Leadership Style

Leadership in a firm that competes through superior strategy should be different from leadership in a firm that largely competes through strategy implementation. In the former, the leader needs to be the great strategist — brilliant, capable of seeing what others do not see, and possessing considerable insight into environmental threats and opportunities. The strategist’s job is to plot the best course through this environment. By comparison, the leader in a firm that competes through strategy implementation needs to be an “organization man” —a great organizer and motivator of people in their day-to-day activities. Here the job is to implement rather than plot strategy, and the leader does this by focusing human, physical, and financial resources on the right activities and managing them with unusual efficiency.

Management Focus

In firms competing through superior strategy, management’s focus needs to be more external than internal. Insight into environmental threats and opportunities and a deep understanding of the industry tend to be more important than great insight into a firm’s internal characteristics and ways of operating. Although both are needed, it is in the former area that the firm primarily attempts to gain advantage over its competitors. In contrast, in firms that attempt to compete through strategy implementation, management’s focus needs to be more internal. A deep understanding of how to fine-tune an organization through better organizational design, management of organizational processes, and motivation of employees lies at the core of this way of competing.

Basis of Excellence

The primary burden of achieving competitive advantage rests on different groups of people under the two styles. Competing through superior strategy places most of the burden on top management and a few other critical people. In semiconductor firms with unique strategies, for example, the people responsible for making the strategy unique tend to be the product designers. In such firms, this small strategic core of people can command very high compensation and status. In firms where competing through strategy implementation is stronger, the strategic core of people is usually much broader. In large Japanese semiconductor firms, for example, competitive advantage depends on the actions of thousands of people scattered across the product design, process development, and manufacturing subunits. No small elite strategic core can by itself establish and maintain competitive advantage in such an organization. Compensation and status tend to be much more equalized in such firms.

Careers

Finally, the two competitive modes tend to require and support different career patterns. Competing through superior strategy implies that company experience is usually less important than industry experience because the latter is required to uniquely position a firm in its competitive environment. One would expect such firms to hire more people from the outside and to have managers with multicompany careers. Competing through strategy implementation, in contrast, favors long careers within a company so that managers can acquire the knowledge and credibility to fine-tune coordination between subunits, motivate critical behavior, and develop a reinforcing company culture.

Data from the sixteen semiconductor companies in my sample tend to support the relationships hypothesized. The firms attempting to compete most strongly through superior strategy were the three U.S. companies with niche strategies based largely on developing and making unique products. In these firms, the general manager tended to see (and frequently referred to) the strategy as his own, and others in the firms also tended to refer to the strategy as “Bob’s” or “Jim’s” strategy. All of these general managers had rich multicompany careers, as did most of the key functional managers as well, and spent considerable time maintaining and using their extensive intercompany networks to help establish a new product as an industry standard, develop second-sourcing and licensing arrangements, and hire critical talent. The general manager and a few people, usually product designers, tended to be viewed as the strategic core that made the strategy viable. In one firm, with a new general manager, the strategy subsequently underwent significant change.

The firms in the sample competing most strongly through strategy implementation were the five broad product line Japanese firms. Their strategies were relatively similar in terms of breadth of product line and markets and where they sought to achieve competitive advantage. In these firms, managers presented strategies as stable and long term and did not identify them with a creator. Top managers seemed to spend most of their time coordinating, organizing, training, and motivating people rather than plotting strategy. For example, the president of one Japanese company was reported to spend three days a week at the company training and development center, interacting with and motivating the employees in the various training programs. As is typical in large Japanese firms, all senior managers had single-company careers and all general managers had spent time in divisions outside of semiconductors. They had largely been selected to be general managers because they were good “organization men,” not because they knew a lot about semiconductors. Their impact on company success was not direct but rather felt through their influence on the actions of many employees.

These two sets of companies were the extremes. Other companies lay between them, competing with a combination of superior strategy and strategy implementation (although generally leaning more toward one or the other). These in-between firms generally had in-between managerial characteristics. Thus the sample provides some preliminary support for my attempt to link differences in managerial requirements to differences in competitive style. If we look back at Table 3, it seems that the two sets of managerial requirements are relatively incompatible, which raises a question about how successful companies or business units can be when they attempt to embrace both competitive styles more or less equally. As already stated, the sample companies tended to lean toward one style or the other.

Confrontation between the Competitive Styles

With the internationalization of semiconductor markets (especially the U.S. market), the two competitive styles have come into direct competition. As one would predict, firms pursuing the typical U.S. strategy have withdrawn rapidly from products and market segments soon after they have been assaulted by Japanese producers attempting to compete largely through superior strategy implementation. This is illustrated by the substantial decline in U.S. market share and increase in Japanese market share for DRAMs, static random access memories (SRAMs), and gate arrays, all products the United States once dominated. The U.S. firms have generally attempted to reposition themselves into more innovative product areas such as electrically erasable programmable read-only memories (EEPROMs) and ASICs. This is the typical U.S. attempt to compete with product-market differentiation or strategic repositioning. The Japanese firms have rapidly gained large shares of the abandoned markets, only to find themselves facing charges of unfair competition and potential protectionist action by the U.S. government. The sentiment among many U.S. semiconductor executives is that the Japanese firms have not played the game the way it should be played.

The interviews revealed that Japanese executives also believed that the game was not being played properly —by the Americans. Several were puzzled that strong U.S. firms abandoned product areas so quickly, that they had allowed their process technology to slip so far behind, and that there was a clamor for government protection. This is not the way the game would have been played among Japanese companies, which have little expectation that competitors will retreat or leave the field when competition becomes intense. Direct confrontation between these two competitive styles seems to have led to a high degree of misunderstanding and a sense that the opposing side has not been playing the game fairly.

Recent Trends

The semiconductor industry is extremely dynamic, and firms often change competitive position rapidly. It is useful to consider how the industry has changed since the study was completed. During 1988 and 1989, U.S. firms continued to lose market share as Japanese firms gained share, primarily owing to the rapid growth of standardized memory products. The relative market shares of the United States and Japan essentially stabilized in 1990 and 1991, which may suggest that U.S. competitiveness has improved. Also during this period, there are reports that large U.S. firms such as Motorola, IBM, and Texas Instruments have instituted massive total quality management (TQM) programs in semiconductor manufacturing and have made major gains in improving quality and reducing unit costs. Such programs have also grown rapidly in other industries, stimulated by the Malcolm Baldrige Award and widespread interest in renewing U.S. manufacturing competitiveness. These programs are primarily aimed at improving strategy implementation capabilities rather than developing a new differentiated strategy that will provide some meaningful advantage. Yet few would argue that U.S. firms can out-implement their Japanese competitors in DRAMs or other standard products or that U.S. firms are pinning their hopes on undifferentiated product strategies and superior strategy implementation capabilities.

A closer examination indicates that market shares have stabilized, not because U.S. firms have been able to halt market share erosion in highly competitive, standardized products, but because the U.S. has had continued success in innovative, differentiated products. The growth in ASICs and microprocessors, which are innovative and differentiated product categories that tend to be dominated by U.S. firms, have served to offset losses in other market segments. In fact, the bulk of this success has been concentrated in one product category and one firm. The rapid growth in Intel’s microprocessor sales during recent years has been largely responsible for the stabilized U.S. position. Thus, although U.S. firms have begun major efforts to improve strategy implementation capability (through TQM programs, the creation of Sematech, and major agreements with Japanese firms to cooperate on developing product and process technology for future generations of DRAMs), these efforts appear to date to have had little effect on recent results or existing competitive trends. Instead, the traditional ability of U.S. semiconductor firms to compete with differentiated business strategies based on superior product innovation continues to explain the relative success and failure of individual U.S. firms and the U.S. industry as a whole.

At the same time, there is some evidence that the major Japanese firms may be attempting to compete with more differentiated business strategies. Japanese firms are increasingly investing in new product technology, working hard to improve their ASIC design and manufacturing capability (most recently with ASIC plants in the United States), and cooperating in joint technology development with innovative U.S. and European firms. In several cases, Japanese firms are assisting Western partners to improve process technology in exchange for access to design tools and standard cell libraries that will improve their capability to compete for the growing ASIC market. Yet, to date, Japanese firms have been able to dominate only the more standardized and competitive gate array segment of the ASIC market, while U.S. firms have continued to dominate the more differentiated standard cell and programmable logic device (PLD) segments. In other words, Japanese firms still rely heavily on relatively undifferentiated, highly competitive products, where superior strategy implementation is the primary basis for competition.

Although U.S. firms are improving their ability to compete in terms of strategy implementation and Japanese firms are making some efforts to develop more differentiated product strategies, it is questionable how far such convergence will proceed. To date, there is insufficient evidence to suggest that the fundamental pattern of competition described by the study is changing.

Implications for Global Competitiveness

In this section I will attempt to generalize from the specific findings and limited hypotheses developed above to some of the broader implications for global competition. The globalization of an increasing number of U.S. markets has brought many U.S. firms into direct competition with foreign firms. Many of these foreign firms, especially those based in the Pacific Rim countries, primarily compete through superior strategy implementation. At the same time, the maturing of many product categories has made it more difficult to compete with unique product-market strategies. In many cases, it has been impossible to consistently stay a step ahead of a determined competitor that possesses superior strategy implementation capabilities. As technology curves flatten out, time is on the side of the better implementer.

Attempting to gain strategic competitive advantage through frequent repositioning of one’s product lines and market segments seems to work best when one’s competitors also believe in this principle and are playing the same game. When one’s competitors are not playing this game, frequent strategic repositioning may not lead to long-run success. Compared to Japanese firms, U.S. firms have fairly flexible business strategies and are quite adept at realigning product lines and market segments to reflect changes in competitive advantage. This kind of continuous strategic maneuvering has worked well for those companies that can maneuver the best. Competitors generally understand the meaning of being outmaneuvered and back off quickly, either going into other areas or licensing the winner’s technology. A good example of this has been Intel, which has consistently focused its resources on developing the next product before others, quickly coalesced a group of customers and suppliers around the new product to make it an industry standard, and retreated from old products and markets as they have matured and become too price competitive.

But Japanese firms do not play this same game of frequently repositioning product lines and market segments. It is more difficult for a Japanese firm to alter strategy and very hard to withdraw from products or markets when one has been outmaneuvered or lost competitive advantage. A good example of this has been NEC’s persistence over the years in developing and establishing its own line of microprocessors, despite the dominance of the Intel and Motorola designs. The typical Japanese response is to refocus its efforts on catching up. Toshiba committed itself in 1982 to catching up in DRAMs and in 1988 became the largest and most successful producer of the one-megabit DRAM. Japanese firms do not generally attempt to leapfrog new developments in either product or process technology, the way many U.S. firms do. Thus, when a Japanese firm falls behind, it can generally catch up by simply working harder at implementing the existing strategy. Conversely, when a U.S. firm fails at implementing an overly ambitious strategy (such as leapfrogging a product improvement), it tends to fall so far behind that it frequently must exit this segment of the market.

Overall, the frequent strategic repositioning of U.S. firms seems to have a definite impact on the strategies of other U.S. firms but relatively little impact on the strategies of Japanese competitors. The Japanese firms are busy implementing product line and market segment strategies that are longer term, less flexible, and not subject to being judged on short-run profitability. When faced with a poor competitive situation, the Japanese firm is much more likely to alter the nature or intensity of strategy implementation than the strategy itself.

This kind of competitive behavior raises questions about the efficacy of strategic repositioning as a means of gaining competitive advantage. Most strategies based on frequent repositioning rely on competitors recognizing and backing away from firms in strategically more favorable positions. For awhile, at least, this leaves the fruits of the abandoned product line or market segment to the winner. If, however, the competitor fails to yield and insists on competing (even at a loss), the expectant winner will not realize the full benefits of its strategic advantage. In fact, if the competitor displays superior strategy implementation capabilities and has staying power, it is probable that the early winner will even lose much of the product line or market segment as the initial advantage wanes. This is because competitors with superior strategy implementation capabilities can usually catch up once they target a product or market segment in which a dominant product design or industry standard already exists. For example, U.S. firms were early leaders in SRAMs and once dominated this market. Now they only dominate the very fast SRAM segment, where technology development has been most rapid and unique designs abound. Japanese firms hold the bulk of the SRAM market and dominate those segments in which product technology has matured. If firms are not deterred by short-run losses and do not confront an unassailable mobility barrier, such as a patent or trade barrier, strategies based on implementation can probably negate the advantages of strategies built on frequent strategic repositioning.

It is probable that competing through superior strategy works only over a limited period of time in an industry’s life. Unless a firm can drive out all competitors while it possesses the superior strategy and raise entry barriers sufficient to keep them out, it must at some point change its competitive style to one stressing superior strategy implementation.

It is important to recognize that both modes of competing — creating a unique and superior business strategy and excelling at strategy implementation — constitute strategic behavior. It is not that one is strategic while the other is tactical or operational, although strategy implementation frequently involves getting tactical and operational behavior to strongly support key strategy elements. Both competitive modes can fundamentally influence or alter the firm’s strategic position in its competitive environment. The overall strategic management of a firm needs to be viewed as some combination of these two modes of competing, although it appears that most firms tend to emphasize one more than the other. While this tendency is undoubtedly influenced by industry and technological maturity, it also appears to be influenced by national culture. If this is true, it is a disturbing complexity for those who deal with global industries, and especially for those who seek to achieve “fairness” or a “level playing field” in global competition.

I conclude with three implications, as shown in Table 4. n the increasingly global competitive system, more U.S. firms will need to shift their way of competing from relying on superior strategy to developing superior strategy implementation capabilities. Superior strategy worked well for the Minutemen and for most industries during the early periods of product life cycles. Their experiences have helped to create a myth that this is the most successful and profitable way to compete. Unfortunately, flattening technology curves and global competition have made this myth out of date for many industries.

Education, reward systems, and cultural values need to change to facilitate this shift toward strategy implementation. U.S. business education has strongly touted the virtues of strategic maneuvering, breakthroughs and rapid success, outflanking rather than confronting competitors, and high early profits. The constant, incremental improvement of a company’s operations and performance is generally viewed as a boring way to compete; it is not the route to the cover of Business Week. Somehow, reward systems need to change.

If the United States cannot become more competitive in terms of strategy implementation capability, it will probably have to establish trade barriers to protect itself from foreign competitors that excel at this competitive mode. Trade will have to be more confined to regional and intracultural competitive systems, and it will have to be more carefully regulated across these regional and cultural differences. To some extent, this is the pattern that may be developing in Europe.

Porter argues that a competitive home environment is a prerequisite for global success.11 This article points out that national environments can be competitive in different ways. It suggests that the U.S. environment is not necessarily insufficiently competitive but that it is more used to a different kind of competition than the Japanese environment. Porter’s notion of competitive and noncompetitive home environments may be too simple to fully capture the reality of global competition when fundamentally different modes of competition are involved.

The framework and hypotheses in this article present a somewhat different perspective of global competition than previous articles. This perspective was suggested by an examination of the business strategies and competitive styles of firms in an industry that has recently become global. Most attempts to explain global competitiveness and international trade problems have employed industry and societal-level factors such as industry spending on R&D, differences in interest rates, and government policies. This new perspective needs to be seen as augmenting, not replacing, the existing perspectives. It suggests, however, that international competitiveness problems exist at multiple levels of analysis and are unlikely to be successfully addressed by simple, single-level solutions.

Appendix

This article is based on a major study of strategy and strategy implementation in twenty-three semiconductor firms. Eight were in the United States, eight were Japanese, and seven were European (the latter are not used here). I used information from public sources to choose companies that represented the strategic diversity of the industry in each geographical area.

I conducted structured interviews during 1986 and 1987 at each company’s headquarters and often at additional company sites. The interviews generally involved the general manager, R&D manager, manufacturing manager, and marketing manager, as well as others. In all but a few instances in which individual managers objected, I taped the interviews. The tapes facilitated the later quantification and coding of certain data and the transcription and analysis of qualitative data.

Although many of the variables and concepts I measured were suggested by prior work, the thrust of the study was exploratory. The central idea presented in this article evolved out of a comparative analysis of the firms’ various strategies.