ABSTRACT: 

How can a start-up firm with a novel technology capture global market share from large, established firms with installed capacity in an entrenched technology? One theoretical perspective suggests that under certain circumstances, firms can achieve higher innovative performance by sharing their technological knowledge with competitors than by keeping that knowledge secret. This paper provides a qualitative case study to explore three hypotheses stemming from such a theoretical argument. 

In 1992 a French entrepreneur named Jean Luc Grand-Clement founded a small start up company to develop and commercialize a new kind of flat panel display (FPD) called a field emission display (FED). Grand Clement's display had several characteristics that made it potentially superior to the FPDs that dominated the market, including lower power consumption, wider viewing angle, greater screen durability, and potentially lower manufacturing costs (OTA, 1995). 

Grand Clement's firm, PixTech, was not alone in pursuing FED technology. Several small start-ups and a few large, multiproduct firms began developing their own FEDs at about the same time. A Japanese firm, Futaba, and several US multiproduct firms including Raytheon, Motorola, and Texas Instruments (TI), saw FEDs as a strategic investment that complemented their existing competencies and product lines. And several smaller, US-based companies including Candescent Technologies, Micron Display, and FED Corporation invested in FEDs in hopes of supplying displays for use in a diverse set of applications. 

From the day of its founding PixTech stood at the frontier of scientific advance on the FED technological trajectory. Grand-Clement recognized, however, that a strong patent position in an exciting technology, by itself, would not be sufficient to propel his firm to competitive success. All FED firms faced an uphill battle to establish their technology against an entrenched product design. PixTech faced additional obstacles because its home country of France lay on the periphery of the FPD industry and lacked the domestic resources and institutions that could support a strong domestic firm. 

Firms pursuing alternative FPD technologies had spent the nineteen-eighties gaining market acceptance for the specific attributes of their own product designs, traveling down a learning curve, and building large production facilities that benefited from economies of scale. Existing FPD firms also enjoyed a strong industry infrastructure that supplied production equipment and specialized materials for the manufacture of their existing FPD technologies. 

PixTech enjoyed neither the market acceptance nor the production experience of its established rivals. FEDs were yet unproven at large sizes and neither PixTech nor other FED firms had ramped up large scale manufacturing lines. In order to attract customers and suppliers, FED start-ups had to prove not that FEDs were on par with existing FPD technologies, but that they were substantially better, substantially cheaper, or both. And other technologies continued to improve. 

The industry infrastructure for FEDs consisted of a handful of nascent equipment and materials suppliers developing their own prototypes. Infrastructural companies had shown reluctance to develop products for a small number of FED firms with no proven product and enormous competition from an entrenched technology. The lack of strong infrastructure hindered all FED firms' speed to market. 

PixTech's evaluation of its technological and market environments motivates the research question: 

How can an entrepreneurial start-up with a novel technology capture global market share from large, established firms with installed capacity in an entrenched technology? 

Elsewhere, we have argued that under some circumstances, firms that share their technological knowledge with global competitors will achieve higher innovative performance than firms that keep their technological knowledge secret. In this paper, we will use an embedded case study approach to empirically investigate that argument in the context of high technology start-ups. In the next section we summarize the logic of the argument and the conditions under which this international knowledge-sharing strategy will likely be most effective. After we summarize the theoretical argument and outline three hypotheses, we will provide a qualitative case study of firms' strategies on the FED technological trajectory. 

Competition Between Product Designs in High Technology Industries 

Management scholars have long recognized that the nature of competition changes over the course of new product innovation. At certain times in the emergence of a new technology, rivalry centers on differences in the fundamental design of firms' products (Teece, 1987). During these times of preparadigmatic competition, the path dependent nature of technological progress ensures that competition between product designs is particularly intense. Firms hold diverse beliefs concerning the commercial and technical feasibility of various technological approaches (Nelson and Winter, 1982). After a firm devotes resources to one technology, its competencies become specialized (Arthur, 1988). In this way, each firm holds an enormous interest in seeing its own technological trajectory "win" the competition between product designs. Even within one trajectory, firms' technologies compete in terms of more specific product design attributes. 

After intense periods of competition, the marketplace generally chooses one design as the technological paradigm for any application. A technological paradigm defines the single product architecture that has won out for any particular application and sets the pattern for subsequent technological progress in the industry (Dosi, 1982; Sahal, 1981). In many industries, a single dominant design emerges for most or all end applications (Abernathy and Utterback, 1978). In other cases, a few competing technologies may survive for distinct applications when the criteria for evaluating the product design vary considerably across end product markets. Technological paradigms, like most institutional arrangements, generally persist over extended periods of time. However, they are not immutable. For instance, Anderson and Tushman (1991) found that a new dominant design arose three times over 25 years in the minicomputer industry. 

When one technological paradigm emerges for any end application, inter-firm rivalry shifts away from rivalry between product designs and toward competition based primarily on price and superficial differences in features (Teece, 1987). 

Selection of a Technological Paradigm 

Management researchers have noted that many technologies that achieve tremendous commercial success and persist over long periods of time fall short of the technical performance offered by alternative product designs (Arthur, 1988; David, 1985). Several researchers explained this paradox by arguing that a firm's innovative performance depends not only on its product's technical viability, but also on the institutional environment in which the firm operates (Bijker, Hughes and Pinch, 1987; Constant, 1980; Usher, 1954). The institutional environment consists of both technological and evaluation standards. Technological standards dictate the set of technical interfaces that a new product uses to interact with complementary products. Evaluation standards specify the criteria that are used to judge the innovation. Empirical evidence suggests that technological and evaluation standards emerge endogenously with the development of the technology. The institutional environment co-evolves with technological advances of innovating firms through repeated interactions among innovators, suppliers, end-users, and regulators (Garud and Rappa, 1994; Ruttan and Hayami, 1984; Van de Ven and Garud, 1993; 1989; Van de Ven, 1993). 

Firms' Strategies to Shape the Institutional Environment 

If technological and evaluation standards do, in fact, emerge endogenously, then every innovating firm can influence their character (Das, 1994). By directing researchers' dialogue about scientific advances in the industry, a high technology firm can shape the institutional environment in favor of its own technological trajectory. We have argued elsewhere that a firm can best influence this scientific dialogue by sharing its technological knowledge with potential and existing competitors. Knowledge sharing will help a firm influence the institutional environment through two mechanisms. 

First, the firm can attract competitors to its own technological trajectory. In doing so, it will form a critical mass of firms with a vested interest in the success of the new technology. A technological trajectory holding a critical mass of well-respected firms has a greater probability of winning the competition between product designs than does a trajectory with few firms. By sharing its technological knowledge, a firm can attract competitors to the firm's own trajectory. Since each firm's innovative performance depends on its ability to win the competition between product designs, firms that share knowledge with the innovation system will attain higher innovative performance than firms that do not share their technological knowledge. 

Second, firms that are able to guide emerging technical and evaluation standards in favor of their own technologies have a greater probability of winning the competition between product designs than do firms that do not actively influence these institutions. A firm can promote the emergence of favorable standards if it can influence scientists' perceptions concerning the most critical technical obstacles to overcome before commercialization and the appropriate criteria to judge the merits of the innovation. Previous research has shown that a firm can influence other researchers' perceptions about these technical priorities by publicizing its own research activities and making the limitations of the technology known to all members of the innovation system (Zuckerman, 1978; Merton, 1938; Barber, 1990). 

The primary objective of a knowledge-sharing strategy lies in increasing the probability that the firm's technology will become dominant in large commercial markets. In pursuing a knowledge-sharing strategy, the firm will likely strengthen its competitors and sacrifice a portion of its own market share in that technological trajectory. That is, by sharing its knowledge, the firm will likely increase the total size of its technology's potential market, but decrease its own share of that market. Therefore, a firm that shares knowledge with competitors will achieve higher innovative performance, but smaller total market share, than it would have by keeping its knowledge secret. 

Influencing the Global Institutional Environment 

Kobrin (1991, 1994) argued that in a growing number of high-technology industries, national markets no longer encompass sufficient geographic space to serve as minimally efficient markets. 

We have summarized the theoretical arguments of this paper in three testable hypotheses. 

Hypothesis 1: Under conditions of preparadigmatic competition, some high technology start-ups deliberately share their technological knowledge with foreign and domestic competitors in order to legitimize their technology and build a global industry infrastructure. 

Hypothesis 2: Under conditions of preparadigmatic competition, start-up firms' strategies to share technological knowledge with competitors will increase the legitimacy of their technology and the size of the global industry infrastructure. 

Hypothesis 3: Under conditions of preparadigmatic competition, a start-up firm's strategy to share technological knowledge with competitors will increase the firm's chances of survival and the firm's innovative performance, but decrease the size of the firm's market share. 

The remainder of this paper provides an embedded case study detailing the strategies of the start-up firms on the FED technological trajectory of the flat panel display industry. The qualitative evidence used to construct the case study stems from four years of study of the FPD industry. We sought to increase construct validity within the case study design by using multiple sources of data, including both direct interviews of industry participants and documented archival sources (Yin,1994). We conducted and documented formal interviews of nineteen firms in the FPD and supporting industries in the U.S., Europe, Japan, and Korea. Additionally, we conducted and documented formal interviews of five firms in the FPD and supporting industries. WE conducted and transcribed interviews of representatives of government agencies that dealt with FPD firms in the U.S., Japan, and Korea and had multiple discussions with several of the leading industry market researchers. The primary data was backed up with extensive archival sources. We used keyword searches of the lexis-nexis retrieval service of newspaper and magazine articles to develop a detailed database of over 3,000 discrete FPD industry events and quotations from industry participants. 

The case study begins by briefly describing the history of the global flat panel display industry and the FED technological trajectory. It then details two distinct strategies used by FED start-ups. First, PixTech designed a strategic alliance to share its technological knowledge with several large, multiproduct firms. The case provides a description of PixTech's corporate history and its knowledge-sharing alliance. The case also briefly accounts for the strategies of those firms that joined PixTech's alliance. Second, other entrepreneurial firms such as Candescent Technologies and FED Corporation designed strategies to keep their technological knowledge secret from competitors. The case describes the strategies and motivations of each of these other FED firms. Finally, the case summarizes the performance of the FED start-ups and the FED trajectory over the first five years of commercial activity. 

The Flat Panel Display Industry 

The term "flat panel display" (FPD) refers to flat information display screens, usually less than one inch in depth, formed through one of several distinct technological approaches. The earliest scientific advances in FPDs occurred in American university and industrial laboratories during the 1960s. For the next 30 years, several distinct technologies competed for selection in the largest application for FPDs, laptop computers. During the 1970s and 1980s computer makers introduced laptops containing screens from at least three different FPD trajectories. 

Several attributes emerged as important criteria for evaluating the FPD technologies. Consumers wanted screens that were clear, easy to read under various lighting conditions, readable from several different viewing angles, efficient enough to be used for hours at a time on batteries, and, of course, affordable. Because each technology contained a different set of favorable and unfavorable attributes and no consensus had yet emerged concerning the relative importance of each attribute, market analysts' opinions regarding the preferred technology for laptop computer applications differed. Additionally, because technological advances ensured that the state of the art for each technology was in flux, managers found it extremely difficult to predict which technology held the greatest market potential. 

In approximately 1989, one technology called a liquid crystal display (LCD) emerged as the clear favorite for laptop computer applications. By 1995, LCDs comprised approximately 87% of total FPD sales (Businessweek, 8/28/95). 

Emergence of FED as a Challenger 

Researchers at Stanford University and private firms such as RCA made many initial breakthroughs leading to the development of field emission display (FED) technology in the 1960s (Businessweek, 1995). However, technical obstacles prohibited any firm from commercializing FED products until very recently. RCA abandoned its FED efforts in 1976 (New Technology Week, 8/21/95) and no firm was able to demonstrate a commercially viable FED screen until the 1990s. 

Field emission displays use a technology similar to that used in traditional cathode ray tube television sets. FEDs are comprised of a thin sandwich. The back is a sheet of glass or silicon that contains millions of tiny field emitters. The front is a sheet of glass that is coated by phosphor dots. When voltage is applied to the field emitters, they bombard the phosphor dots with electrons, causing them to glow. 

FED technologies fall into two principal technology branches. Some technologies use a five-layer microtip device that requires 200 million field emitter tips to be laid down per square inch of screen space. Each microtip is smaller than one micrometer and must be deposited into a dense grid. Other technologies use a simpler, three layer display that creates an array of field emitter tips directly on the glass substrate by depositing a diamond-like film that is meshed with a phosphor-coated plate (New Technology Week, 8/21/95). Within each of the two basic approaches to FEDs, further variations can be made based on the process and materials used to construct the microtips and the use of low-voltage and high-voltage designs. Additionally, firms have used different methods to ensure the long-term purity of the display and have chosen different combinations of phosphors to display color. 

Firms showed interest in FEDs due primarily to a belief that they were superior to other FPD technologies for many important display applications. For instance, FEDs exceeded LCDs' capabilities in full motion video and viewing angle. They were more durable in challenging environments, such as temperature extremes in automobiles and aircraft. FED technology also received the stamp of approval from one major consumer of displays, the US military. Program managers in the Defense Advanced Research Projects Agency (DARPA) reported publicly that FED technology held great promise for both military and civilian markets (Hartney, 1995; Slobodin, 1995). 

PixTech, Incorporated. 

The French national laboratory, Laboratoire d' Electronique de Technologie et d'Instrumentation (LETI), began researching FED technology in the late 1980s (Business Wire, 4/14/94). By 1991, it had publicly demonstrated the world's first 6-inch black and white FED (Report from Japan, 8/26/91). In 1992, Jean Luc Grand-Clement bought an exclusive license to all of LETI's FED patents and launched a new company called Pixel International (Businessweek 8/28/95). The name was later changed to PixTech. PixTech led the forefront of the development frontier for FEDs throughout the emergence of the technological trajectory. According to Grand-Clement, the company always led other FED firms by 10 to 18 months (Grand Clement, 1996a). Objective evidence tends to support this claim. PixTech was the first company to show color displays in 1993, and consistently led the industry in the size of screens that it demonstrated. In fact, no other FED company publicly demonstrated an operating FED that did not use PixTech's own cathodes until 1996. PixTech was the first to show products off a manufacturing line. It was the first to sell its products commercially. In 1996, PixTech won a "Display of the Year" award from Information Display magazine for its 5.2-inch monochrome FED (Information Display 12/96) and also showed a 24-bit full-color, 10.5-inch FED prototype with full motion video. By nearly every measure, PixTech defined the frontier of FED technology from its founding through the 1990s. 

Although PixTech led of the pack of FED producers, Grand-Clement and other company managers realized that they faced an uphill battle for the acceptance of the new technology given a sea of alternatives and the dominance of LCDs. Grand-Clement also acknowledged that as the single producer of the technology, it would be extremely difficult for PixTech, alone, to attract the necessary infrastructure of equipment suppliers to the FED trajectory (Grand-Clement, 1996). 

In order to address the challenges inherent in gaining acceptance for a new technology with strong competing product designs available, PixTech designed a two pronged strategy. First, the company continued to improve its manufacturing technologies to sell its own displays commercially. To that end, it entered into a relationship with Unipac Optoelectronics Corporation, that allowed PixTech to begin planning commercial manufacturing in Unipac's Taiwan facilities. Second, Grand-Clement created a strategic alliance with a subset of other FED firms. The objective of the alliance was to share PixTech's technological knowledge with potential and existing competitors. 

The Grand Alliance 

In 1993, CEO Grand-Clement devised a unique strategic alliance with several large multiproduct firms from around the world. Under the terms of the FED Alliance, the firms agreed to cross-license all intellectual property (in the form of patents, know-how, and trade secrets) to all other alliance members. TI became the first alliance partner in June, 1993. Raytheon, Futaba, and Motorola joined the alliance in 1993, 1994, and 1995. Grand-Clement stated openly that he was willing to expand the alliance to include six companies in addition to PixTech (1995 Annual Report). By 1997, the alliance members were sharing over 300 patents (1996 10-K Report). Grand-Clement designed the FED Alliance to promote the sharing of tacit as well as explicit knowledge. According to Grand-Clement, "This is not soft cooperation. Companies work intimately together and keep no secrets from one another" (Grand Clement, 1996a). Members of the alliance were allowed to make regular visits of one another's facilities and to install researchers to work in another company's labs for months at a time (Grand-Clement, 1996). 

While all knowledge and all patents were held in common among alliance partners, Grand-Clement claimed that he built a "Chinese Wall" around the consortium, sharing little or nothing with the rest of the industry (Grand Clement, 1996a). 

Grand-Clement built important safeguards into the alliance agreement. First, he initially discouraged equity investment from partners into PixTech because he believed that the large partners could overwhelm PixTech's independent strategy. Second, he chose initial partners to minimize the likelihood that any two alliance members would initially enter the same niche of the FPD industry. Each alliance member targeted a specific niche of the FED market. Grand-Clement acknowledged, however, that the firms would likely become direct competitors in the future (Grand-Clement, 1996). Finally, the knowledge sharing agreement with each partner was scheduled to expire three years from its inception. 

Grand-Clement articulated several objectives for the FED Alliance. He hoped that cross-licensing agreements would provide the latest technology for PixTech's use, and that the license fees paid by alliance members would ensure early revenues for PixTech. PixTech earned licensing payments amounting to $5.6 million in 1994 and $9.9 million in 1995. In the same years, PixTech paid alliance members $1.7 million and $3.1 million for use of their intellectual property (1995 Annual Report; 1996 10-K). Beyond these financial incentives, Grand-Clement intended for the alliance to build a critical mass of firms in the industry and, thus, help establish the worldwide acceptance of FED technology. Grand-Clement (1995) listed four ways that the alliance would help build the FED industry. 

• The alliance would promote the creation of a critical mass of R&D on FED technologies.
• A unified effort would accelerate market acceptance of the new technology.
• The investment of several large firms would help generate an infrastructure of equipment suppliers.
• The pooling of knowledge between several FED firms would decrease the time to market for all firms in the alliance.

PixTech's strategy was not without risks. In effect, PixTech gave large competitors the knowledge and experience they needed to compete effectively. Other alliance members were well positioned to absorb PixTech's technology. Large firms such as Motorola, Raytheon, and TI had access to capital, production facilities, and skilled employees who could make good use of the knowledge provided by PixTech. 

Alliance Members 

Each of PixTech's alliance partners saw a strategic interest in identifying an FPD technology that matched their own resources, competencies and product lines. Futaba had long sold other types of displays for instrumentation products. TI utilized displays in computer-based applications. Motorola saw displays as essential in providing a critical video interface to complement the rest of its telecommunications business. Raytheon saw flat displays as a key segment that could replace its own CRT business. 

For these established, multiproduct firms, the FED Alliance presented itself as an inexpensive route into the industry. The alliance partners received permission to use PixTech's intellectual property and gained a conduit through which to access both tacit and explicit knowledge from experienced industrial researchers. 

Futaba 

The Japanese firm Futaba joined the FED Alliance primarily to access knowledge and intellectual property regarding FED technologies. The company had long used another display technology, Vacuum Florescent Display (VFD) to produce screens for automobile dashboards. Futaba saw investment in FEDs as a way to maintain market share in that changing market. According to Chris Lupek, director of sales and marketing for Futaba Corp. of America, 

Texas Instruments (TI) 

By the time that it considered FED technologies, Texas Instruments had made stabs at several other technological trajectories in the FPD industry. It had pulled off of those trajectories largely due to technical difficulties and questions about the competitive risks inherent in FPD markets. By the 1990s, TI was procuring screens from Japanese firms and saw investments in FEDs as one way of getting into the display business and lowering the costs of its own components. Thomas Petrovich, TI's marketing manager for flat-display products, stated that his company lost out on the FPD market when Japanese firms took over the industry in the mid-1980s. ''We're not going to let that happen again.'' Petrovich saw the alliance as a way of helping all FED firms make a mark on the FPD industry. Working separately, progress would be slow, and ''even the first guy would get to market too late'' (Businessweek, 8/28/95). 

Raytheon 

Raytheon had supplied traditional displays for applications in ships, aircraft, and air traffic control systems for 40 years and recognized a need to consider alternative display technologies that could provide better brightness and sunlight readability (Department of Defense, 1994). In 1995, Jack R. Kelble, manager of Raytheon Electronic Systems command, control, communications, and components business area commented, "We're interested in FEDs because these displays promise to outperform ¼ liquid crystal displays, especially in picture brightness, image quality, power efficiency, and manufacturing yield," (PR Newswire, 9/11/95). FED production would leverage Raytheon's competencies in display design and semiconductor processing. The company began researching FEDs in the mid 1980s and entered the FED Alliance in 1993 to gain access to critical proprietary technology that was essential in bringing its product to market. 

Motorola 

Motorola began investing in FEDs to complement its other competencies in telecommunications. According to Peter Shinyeda, general manager of Motorola's Flat Panel Display division, FEDs would eventually be used in a wide range of products, including portable computers, cellular phones and electronic games (Edge, 7/24/95). Shinyeda, pointed out that Motorola was "kind of late in entering into the display market," and this was one reason the company made a bet on FEDs (New Technology Week, 8/21/95). Rather than trying to come from behind in the LCD market, the company hoped to leapfrog LCDs by using a more advanced technology. He summarized Motorola's perspective on the alliance by saying, 

In conclusion, the members of the FED Alliance appeared to view the partnership as a door into the industry. Because Futaba, TI, Raytheon, and Motorola did not view one another as direct competitors, they saw such a partnership as a reasonable way to gain technical competencies as well as build an industry infrastructure to benefit all firms. 

In many ways, the commercial success of PixTech would have to be both positively and negatively correlated with the success of its alliance partners. If all partners failed, then the FED industry would be awarded neither the legitimacy nor the industry infrastructure that would buoy PixTech's own profits. However, if FEDs became well accepted, PixTech would find that it had fostered direct competition in most markets for FED products. By sharing its knowledge, PixTech hoped to vastly increase the size of the market for FEDs. However, it was forced to acknowledge that in the end, it would probably capture a fairly small piece of that market. 

Independent Firms 

Several other firms, including Micron Display, Candescent Technologies, and FED Corporation, chose to develop FED technologies in relative isolation and viewed technical advances and scientific knowledge as critical sources of their own competitive advantage. Each firm built a technical foundation based on expired patents and its own proprietary research. Each firm shared relatively little knowledge about technological approaches and breakthroughs with other FED firms. In fact, each displayed substantial secrecy even in publicly showing its prototypes and discussing broad industry strategy. 

Micron Display designed a strategy of relative isolation. It chose not to belong to any sort of knowledge-sharing alliance. It also avoided membership in the U.S. Display Consortium, the primary industry trade association for U.S. FPD firms. Its managers rarely spoke publicly about the firm's strategy or even made many announcements concerning future production investments. 

Candescent Technologies proclaimed that its strategy was to beat all other firms to market with the best FED technology (Marshall, 1996). Although the firm's managers agreed that the FED trajectory would benefit from a critical mass of industry rivals, the company never went out of its way to strengthen those competitors. Instead, Candescent chose to pursue partnerships with potential end consumers of displays, such as HP and Compaq. By enlisting established computer makers to invest in Candescent, the company attempted to increase the legitimacy of the FED trajectory, and of its own particular product design. 

Candescent designed a technical strategy to reduce its dependence on a new industry infrastructure made specifically for the FED trajectory. It designed its product and processes to utilize as much off-the-shelf production equipment as possible. The company designed its systems around the objective of buying 80-85% of its required tools, components, and materials from existing, commercial vendors--often using tools developed for use of LCD makers. According to Candescent's strategy, only 15-20% of tools would be designed and produced specifically for its FED proprietary design. 

Candescent, then, formulated a strategy to avoid the problems that PixTech's FED Alliance was designed to resolve. First, it sought to build legitimacy, not by attracting a critical mass of firms to produce its technology, but by attracting end-users to publicly proclaim the utility of the company's specific design. Second, it lessened the importance of building a new industry infrastructure from scratch by reducing the number of novel tools that it employed. 

FED Corp. also designed a relatively independent strategy. Its president Gary Jones viewed all other FED firms primarily as competitors. Some of the firms would regularly butt heads with FED Corp. in the marketplace. However, he believed that even firms that never called on the same customers still competed with FED Corp. for the money of investors (Jones, 1996). This perspective motivated FED Corp. toward a stance of relative isolation. The company did not pursue strategic alliances with other FED firms or with many large end-users. It also did not make a strong effort to contribute its technological knowledge to the global scientific community. 

Performance of the FED trajectory 

It is impossible to provide hard conclusions for either the success or the failure of PixTech's knowledge-sharing alliance. The greatest optimists haven't expected to see FEDs in an extensive range of products before the beginning of the next century. 

However, five distinct points support the contention that the FED trajectory gained legitimacy after the formation of the FED Alliance in 1992. Industry insiders attribute at least some of this increased legitimacy to PixTech's strategy. 

Industry analysts began to predict that FEDs would capture some market share in the FPD industry. 

In the first years of the FED Alliance, the most prominent FPD market analysts rarely mentioned FEDs as a potential player in the industry. Through 1994, Stanford Resources, Inc., a market research firm specializing in the electronic display industry, rarely mentioned FED technologies when describing the future FPD industry. Their 1994 report projected future market shares for five different FPD technologies, but did not include FEDs. By 1995, Stanford Resources included explicit projections for FEDs. And by their 1996 report, the analysts projected FEDs as enjoying growing market shares from 1997 onward, holding 1.5% of the FPD market in 2001 and 1.9% of the market, by 2002. 

Existing LCD manufacturers joined the FED trajectory. 

Several very large firms, including many of the largest LCD manufacturers such as Toshiba, Fujitsu and Hitachi began to quietly employ research teams to investigate FEDs. These companies have filed over 100 patents on FED technologies in Japan. Although they have committed the bulk of their resources to maintaining a production edge in LCDs, their interest in FEDs demonstrates increasing legitimacy of the technology. These research investments suggest that firms on alternative technological trajectories see FEDs either as a potential threat to LCDs in the long term, or as an alternative technology in markets for which LCDs are not well suited. 

Several firms invested in FED manufacturing lines. 

Several FED makers have joined PixTech in installing manufacturing lines for FEDs. By 1996, Futaba was producing 5-inch displays with full color on a pilot manufacturing line in Japan. Micron installed production lines dedicated to very small FEDs for use in viewfinders and the firm made investments aimed at scaling its displays up for use in its own line of laptop computers. FED Corp. installed production capacity in upstate New York to manufacture customized displays in short production runs for specific customer needs. Candescent Technologies invested in a pilot line that could eventually turn out 100,000 screens per year. It also committed itself to a large-scale manufacturing line that will cost $400 million and produce over a million panels each year. Motorola dedicated well over a hundred million dollars to installing a pilot plant in Arizona. The pilot plant is scheduled to begin producing in 1998. According to Tom Credelle, marketing director at Motorola's FPD Division, 

FED firms attracted additional investors. 

Financial markets have continued to make investments in FED firms. For instance, Candescent Technologies raised $55 million in a private equity placement in 1996. (Clock, 1997) and in early 1997, PixTech announced that it had raised $24 million for its FED development. 

The industry infrastructure for the FED technological trajectory has grown. 

Several equipment and materials suppliers have emerged to support FED development and manufacturing. For instance, Display Technology Systems was founded in January, 1996 for the exclusive purpose of building FED process tools. Schott Corporation has developed competencies in vacuum sealing and reliability engineering to provide glass that is particularly well suited for use in FEDs. Accufab Corporation began developing customized tools for FED manufacturing. 

Industry insiders attribute at least some of the success of the FED trajectory to PixTech's strategy. David Mentley of Stanford Resources confirmed the potential utility of the FED Alliance. ''Historically, nothing ever happens until there's a critical mass of companies... For now, the rivalry is just what PixTech needs" (Businessweek, 8/28/95: 73). Additionally, a marketing manager at PixTech's rival, Candescent Technologies, acknowledged that there was no question that PixTech had contributed to the industry as a whole. Of course, he went on to point out the superiority of his own company's product (Sturiale, 1996). 

PixTech's Performance 

More ambiguous than the growing success of the FED trajectory is PixTech's own performance. PixTech and Futaba remain the only two firms to market FEDs commercially, and the company recently announced that it had received an order worth more than $10 million. The order of 50,000 displays is intended to be used in portable medical equipment. 

PixTech has also been fairly successful at raising money. It recently received grants from French and European Union development contracts, completed a public offering on the Easdaq (European Association of Securities Dealers Automated Quotation) market, and attracted private placements with Motorola and United Microelectronics Corporation of Taiwan (PixTech Press Release 2/17/97). 

Even with these successes, however, it is clear that within a few years, PixTech will be one of several producers in the FED market. Its stock price has ridden a roller coaster, increasing and decreasing over the past three years between$2 7/8 to $9 1/8. And its share of the FED market will surely decline if Candescent Technologies and Motorola go through with plans to increase global capacity to well over a million panels per year. 

This paper introduced three hypotheses concerning entrepreneurial firms' strategies to share knowledge with the global scientific community. 
 

H1	Under conditions of preparadigmatic competition, some high technology start-ups deliberately share their technological knowledge with foreign and domestic competitors in order to legitimize their technology and build a global industry infrastructure.
H2	Under conditions of preparadigmatic competition, start-up firms' strategies to share technological knowledge with competitors will increase the legitimacy of their technology and the size of the global industry infrastructure.
H3	Under conditions of preparadigmatic competition, a start-up firm's strategy to share technological knowledge with competitors will increase the firm's chances of survival and the firm's innovative performance, but decrease the size of the firm's market share.

Hypothesis 1 received strong support. PixTech's president Jean Luc Grand-Clement drew a picture of an industry environment very similar to the one described more abstractly in the theory section. After evaluating the dynamics of this industry environment, Grand-Clement realized that one firm alone would not succeed in the long run in this high technology industry. In order to increase PixTech's chances of success, he sought to create a critical mass of firms on his own technological trajectory. 

Hypotheses 2 and 3 both received weaker support. The FED technological trajectory has clearly gained legitimacy over time. And industry participants suggest that some credit for this growing legitimacy belongs to PixTech for constructing the knowledge-sharing alliance. It is impossible, however, to provide clear evidence that the industry would have crumbled without PixTech's efforts. 

PixTech's future in the industry is more ambiguous, still. CEO Grand-Clement believes that PixTech would not have survived without a critical mass of competitors on the technological trajectory. The company is maintaining an acceptable level of performance. However, if other firms meet their objectives and bring new manufacturing facilities on line, PixTech will lose much of its market share. If this occurs, PixTech will quickly lose its position as a technological and market leader and become one of many FED producers. Grand Clement appears to have succeeded to contributing to the legitimacy of the FED technology. The success of his own firm is still uncertain. 

Three critical conclusions emerge from this case study. First, previous researchers have focused a great deal of attention on firms' need to safeguard proprietary knowledge from competitors. This paper, however, concludes that under conditions of preparadigmatic competition, firms may actually increase their overall performance by sharing their knowledge with potential and existing competitors. The paper has shown that some firms pursue deliberate strategies to share their technological knowledge with competitors. It has also provided support for the argument that knowledge sharing contributes to the overall strength of the firm's technological trajectory and, by extension, higher overall performance for the firm, itself. 

Second, the paper highlights the interdependence of all firms occupying a given trajectory of a high technology industry. Under conditions of preparadigmatic competition, an individual firm's success depends heavily on the performance of its technological trajectory and of other firms on that trajectory. For instance, PixTech's CEO concluded that their company's odds in the FPD industry would be heightened by the successes of its industry rivals. The company's managers believed that in emerging high technology trajectories, the success of one firm correlates positively with the performance of other firms pursuing the same technology. Their strategy to share knowledge with select firms attracted investors, helped build an industry infrastructure, and increased PixTech's chances for long term survival and profit. 

Finally, this paper supports the notion that firms' strategies must change course through different phases of the emergence of a new industry. A firm's decision to deliberately increase the number and strength of its direct competitors would be ludicrous in absence of preparadigmatic competition. We argue, however, that at certain times throughout the emergence of a new industry, such a strategy may improve a firm's performance by increasing the likelihood that its technological trajectory will win the preparadigmatic competition. If the marketplace selects the firm's design as the technological paradigm in an application with large end product demand, then all firms pursuing that technological path will achieve higher innovative performance than if they had been relegated to a 'losing" path.