Collaborative Production Strategies for Technological Innovation and Leadership in Network Industries: The Case of Bluetooth John Rice PhD Student School of Management Queensland University of Technology GPO Box 2434 Brisbane Queensland 4001 Australia Ph: 61 (0)7 3864 1345 Fax: 61 (0)7 3864 1766 Email:
[email protected] Abstract This paper notes the creative tension created by network industries in high technology production. The tension emerges due to firms’ needs to both focus on developing strong and defensible competencies and also the need to share knowledge as part of wider development networks. The paper looks at the development of the Bluetooth wireless standard within this context. We find that Bluetooth’s promoters made the standard open to create support within a competitive technical environment. Downstream profits will flow to the technology’s developers both through the creation of patentable technologies relating to the standard’s use, and also through the ability of the technology to extend the role and importance of the promoter firms’ core products.
2
Introduction As products and services in the information technology and telecommunications industries have become more interconnected and complex, it has become increasingly important for all participants to develop positions of sustainable competitive advantage based upon some degree of knowledge leadership. Peteraf (1993) posited that such sustainable competitive advantage emerges from an accumulation of knowledge resources that are controlled through some means by the firm (through either tacit or formal procedures). These knowledge resources, in turn, have a value determined by competitive markets that value knowledge according to a variety of criteria – determined in part by its necessity within broader systems, its inimitability and its imperfect mobility. Control of knowledge that is a necessary resource in broader productive activities provides an evident avenue for a defensible position of competitive advantage. Robertson (1996), Rumelt (1991), Amit and Schoemaker (1993) and many others have noted that the rent-seeking and achieving status of technology based firms can be derived from successful organizational differentiation based upon technological competencies. While the imperfect mobility of knowledge emerges from both tacit and non-tacit factors, the control of non-tacit knowledge through patenting and IPR protection (with their concomitant royalty payments) tends to be favored by management. While not underestimating the importance of tacit knowledge, in network based industries with stable technological bases, the fluidity of knowledge transfer relates primarily to the formalisation of the control and ownership of knowledge through patent based controls of intellectual property rights (IPRs). It is these formalised knowledge assets that form the very basis of what has become described as “intellectual capitalism” (Granstrand, 1999, 318) by providing the initial impetus and support for research and development of knowledge assets. These knowledge assets, like all others, require the investment of time, energy and capital. The strategic factor markets literature developed by Barney (1986) and others sought to better understand the real value of inimitable knowledge assets or competencies. For many managers, a business strategy based upon free dissemination of valuable knowledge resources seems to be counterintuitive. The spread of technological standards, especially within information technology and telecommunications, can formalize and regularize the necessity of certain IPRs and knowledge use. This in turn provides further impetus for rent seeking behaviour by firms seeking to increase the value of their competence stock. The need for technological compatibility has provided knowledge driven firms like Microsoft with the ability to control both the technological trajectory of their product and also the ability to protect their technology from the imitation evident in industrial product markets. This desirable strategic position (at least from the point of the owner of the knowledge) tends to not last forever. As Microsoft have found, its dominant position within its corner of the information technology industries has not escaped the interest of regulators.
2
3
Further, while reducing the potential for imitation, formal knowledge controls within technical standards do not negate the potential for paradigmatic change that can reduce the value of entire sets of competencies. Indeed, tight knowledge controls tend to exacerbate the desire among competitors for technological foment and discontinuous technological change through standards change. Such processes, as Barney (1986) has illustrated, can have significant deleterious effects of the value of firms’ knowledge stocks as technological foment occurs. These observations are consistent with much of the work emanating from the resource-based view of the firm, a view that presupposes the continual recombination of knowledge resources within and between competing firms and the regular creation of new knowledge resources within industries. It is logical for dominant firms to seek to control the processes of knowledge innovation through whatever means (formal and informal) are available to them, while it is logical for innovative firms to seek to overcome these controls wherever possible. Technological change that may be unrelated to the competitive dynamics within and between standards and technological paradigms may also to be important. Collis (1994) noted that firms that are too reliant on a certain set of competencies tend to find themselves in a precarious game of catch-up during periods of discontinuous technological change where they do not have organizational competencies in emerging technological standards. These technological discontinuities are becoming more frequent as technological change drives increasing cycles of innovation and redundancy in information technology and telecommunications. Such a state of flux in knowledge markets makes the strategic planning related to knowledge management inherently difficult. Managers have to deal with continuous and discontinuous technological change, much beyond their control. Organizational survival and success tends to be reliant on an organization’s ability to adapt to changing circumstances and utilise changing technologies in an efficient and effective manner. Teece, Pisano and Shuen (1997) emphasised the need for dynamism in the strategic management of organizational competencies. By recognising the inevitable diminution of knowledge value over time, these authors placed at the centre of the strategic management of knowledge two key imperatives. First, management must be focused on renewing competencies where it is possible and necessary to ensure their continued value is protected and second, management must be cognizant of the emerging opportunities and threats made available by a dynamically changing environment. Care should be taken with regards to the nature and degree of environmental engagement. As has been noted, firms that seek to take active control of both their technological assets (competencies) and their technological environment (the technological standards of their industry, for example) may find themselves accentuating paradigmatic or standards-based technological competition.
3
4
The firm seems to have a number of strategic choices with regards to the type of knowledge strategy employed vis-à-vis its external environment. A degree of control can be exercised within a particular technological paradigm, or a firm can seek to influence the way in which new paradigms emerge. The benefits of the former approach flow from the accumulated benefits from the ownership of knowledge, and tends to support the exploitation of Ricardian rents emanating from patents and IPR controls. Questions emerge as to the potential benefits available to a firm who adopt the alternative approach. Potentially, the adoption of a leadership position within the process of an open and non-proprietary standard’s emergence is facilitated by opening rather than controlling knowledge resources. The trade-offs continue regarding the choice of knowledge control of they impact knowledge diffusion: removing the potential for royalty payments is contradictory to many managers’ views of gaining maximum benefit from their value, though strong protection of knowledge through legal or other forms limits its diffusion across firms. Potential Benefits from Collaborative Knowledge Relationships Alliances tend to be driven by potential complementarities between some types of characteristics held by the firms involved. Stuart (2000) notes that the potential benefits for firms differ – for small firms the reputational benefits of involvement with a large, innovative firm tend to be significant. For larger firms, access to resources and know-how tend to be key drivers. Where alliances require the development and maintenance of physical assets, they tend to be more concrete in their form and durability. Where the resources involved tend to be intangible (like reputation) or knowledge based, more innovative structures tend to be required that can take into account the asymmetric nature of the knowledge development process. Given the increasing rate of technological change, a degree of dynamism is an important aspect of any proposed technology and knowledge strategy adopted by firms. Foss (1997, 311) notes that the major function of all firms is to deal with the challenges presented by “changing, partial, tacit, complex, asymmetrical etc. knowledge and its useful application”. The growing importance of knowledge as an asset is also challenging old assumptions regarding resource allocation within and between firms. As complexity in information technology and telecommunications has increased, a degree of vertical and horizontal disintegration has occurred in these industries (Lamming, 1993), requiring firms to cooperate within both design and supply chains to develop products and services (Lynskey, 1999). Processes of disintegration have created both the opportunity and need for network development between firms in developing complex products or services that are compatible with complex broader systems. Research by Larson (1992) and Larson and Starr (1993) emphasises the importance of stable, multidimensional and complex inter-organizational exchange relationships in the development of new products, with Larson (1992) specifically investigating the importance of these relationships for entrepreneurial, though resource-poor, firms. Eisenhardt and Schoonhoven (1996) found that when industries
4
5
were characterised by emergent technologies, alliances were beneficial due to the potential for the emergence of synergistic knowledge benefits and risk sharing. Drawing a nexus with the previous discussion relating to organizational competencies, the question arises as to whether the presence of contractual or mutually beneficial alliances can assist in the development of sustainable competitive advantage through competencies created and fostered within organizational networks. As knowledge development becomes more important as the basis of strategic alliances between firms, there is an opportunity and indeed a necessity to ensure a degree of fluidity and dynamism in its allocation and development. Thus a strategic dilemma of sorts has emerged for many industry participants in high technology endeavors. A degree of knowledge sharing is necessary to ensure a role in a broader complex system, while a degree of exclusivity is also necessary to ensure differentiation from other industry participants. This dynamic tension is at the core of the planning for and implementation of network alliance relationships among firms and creates operational challenges in a number of areas. Implications: Knowledge Management Accepting the need for some interconnectedness in the knowledge generation process, the nature and degree of this knowledge sharing process then emerges as an important area of investigation. In the business environment of information technology today, industrial power in network-based industries has tended to flow from control of knowledge through patents, intellectual property rights and research and development leadership. As has been mentioned, IPRs by their very nature hinder the flow of knowledge between firms and applications. The process of creating new technological knowledge thus features a dynamic and creative tension between knowledge controls and knowledge flows and a structural-industrial tension between innovative knowledge creators and knowledge owners. For dominant firms, the patent process tends to externalise their non-tacit knowledge, creating opportunities for competitors. Furthermore, the innovation processes in today’s information technology industry facilitate and require rapid iterations of knowledge sharing and progression. Thus the degree to which the knowledge underlying products and service structures is protected versus made accessible has a strong determinative impact on the type, level and location of innovation observed across a product category. This situation drives the emergence of opportunity costs created by knowledge control which must be weighed against the potential diminished strategic value of the knowledge should it be incorporated into competitors’ products. Implication: The Need for Consonance between Organizational Knowledge and Technical Standards The opportunity costs of knowledge control are exacerbated by technological competition among competing standards. A firm may have ownership and control of an excellent technological solution to a market driven opportunity, though if that
5
6
solution lags others to market due to long development times or other factors, the firm will not benefit. Building installed base is a key objective, and the complexity of systems based upon information technology ensures that standards to ensure interoperability are central for product or service success. Such observations are not new. Seminal papers from the microeconomics field (Farrell and Saloner, 1985; Katz and Shapiro, 1985) stressed the strong network effects (positive network externalities) that can support the diffusion of a new technological standard within the competitive context. Katz and Shapiro (1985, 424) for example noted that “there are numerous products for which the utility of a user derives from consumption increases with the number of other users consuming the good”. In the information technology industry a critical dimension of organizational success is the extent to which a technical standard can be broadly adopted, and this in turn is influenced by the degree of openness and competition allowed. An open and nonproprietary approach to the knowledge of the introductory technology can and does facilitate its uptake by vendors, users and innovators. From an industrial perspective, when a standard is developed within a knowledge architecture that features explicit and widely accessible knowledge, a number of desirable outcomes are facilitated to assist this end, namely: • • • •
reduced barriers to entry into the industry, accelerated diffusion of technical standards, more regular incremental and modular innovations and enhanced learning opportunities.
Examples from information technology are plentiful, and include: •
The IBM personal computer was conceived as an open system. Through an open architecture, the personal computer (PC) by IBM (and clones) has advanced very rapidly. Even though many believe the Macintosh to be a superior system, the PC has become the dominant design because of complementary products such as software and the rapid development of the product through having multiple suppliers involved. Apple succeeded on their second attempt with the Macintosh (following the Lisa) due in part to the fact that it allowed third parties to become involved at the software development level through having an architecture that was not quite as closed as previous hardware releases (Morris and Ferguson, 1993).
•
The case of the rise of Sun Microsystems and the fall of Apollo show some of the advantages associated with open standards based products. Sun started from a weaker position, but rapidly overtook Apollo by utilising the potential of their open architecture approach and allowing other suppliers to become involved in manufacturing using the Sun architecture (Garud and Kumaraswamy, 1993). This case simultaneously demonstrates the problems of a completely open architecture, as Sun is now embroiled in a battle with a range of other competitors that have
6
7
entered the workstation market, using the same basic architecture (Baldwin and Clark, 1997). •
The GSM mobile telecommunications standard has emerged as a dominant standard among digital offerings, outstretching its competing North American and Japanese alternatives in takeup. This has in turn proved to be a strong driver of success for European firms, and most especially the Nordic duo of Ericsson and Nokia who could leverage their TDMA competencies in the development of new GSM equipment. GSM tended to be far less proprietary with regards to technology diffusion and royalty requirements than the competing CDMA solutions from the US and the proprietary NTT solutions from Japan.
Open Architectures – Potential and Problems Products with open architectures are constructed using information structures that can be accessed (directly or indirectly) by others within the industry. This generally occurs when the developing firm does not try to protect the information structure or even actively encourages the assimilation of it into the industry. In contrast, closed product architectures utilise information structures that are tightly controlled by the developing firms. The accessibility and transferability of the various elements of the information structure that underlie products are the key determining factors as to whether the product’s architecture is open or closed. Beyond the lowering of barriers to entry within an industry and the opportunities for a geographically dispersed industry to develop, an open architecture provides greater opportunities for a product to become the dominant design. Products normally go through two stages in terms of evolutionary development; the pre-paradigmatic stage when multiple architectures are available, and the paradigmatic stage, where there is acceptance of a particular ‘standard’ (Teece, 1986). To achieve a dominant design position, a product must be floated in the market such that acceptance of the product can be gained. There is evident risk of leakage of technological competencies at this point for lead developers. In the case of a closed architecture these risks are limited through legal protection mechanisms. Problems can and do emerge in closed architectures through information leakage via component suppliers and the potential for reverse engineering by competitors of products with similar functionality. An open architecture allows relatively easy duplication by competitors due to the nature of the underlying information structure and the type of knowledge within it. However, an open architecture also allows for competitors and suppliers to become involved in the further development of the product such that the technological advances made through this joint effort can lead to a dominant standard very quickly. This is particularly important in the case of a system of products as the number, quality and availability of complementary products often determine their success. The strategic objective of releasing an open and closed standards based product is thus quite different. Teece (1986) suggests the achievement of a dominant design is one of the three building blocks to reaping profits from an innovation. Whilst profiting from an innovation can occasionally occur because of the higher margins that can be
7
8
achieved, more commonly, success comes from capturing a large share of the market. Certainly a dominant design on the basis of an open architecture is unlikely to allow for higher margins to be achieved, but it does allow for rapid development of the product or system, rapid growth and possibly financial success through a larger market share. Growth can and does occur at the periphery of the product, where the supplier of the open standard product may develop complementary and compatible products and services. As information technology has moved towards a solutions focus, the sales of software and hardware have become a central though early stage aspect of market strength for vendors. Later sales of complementary solutions and services are emerging as important areas of market activity. An open standard, though, may not necessarily draw in firms unless this is facilitated through cooperative means. Open standards may facilitate the rapid development of the product if other firms become involved in the design process. The attractiveness of entry for other firms increases if an open standard emerges as a dominant design. Open Standards, Open Partnerships? Once a decision has been made to freely disseminate information regarding a standard, it would seem to logically flow that firms might wish to collaborate cooperatively in the processes of research and development. Such a focus accelerates product introduction and tends to ensure the creation of a variety of products early in the competitive product cycle. This will tend to build installed based of standard compatible equipment of both core and complementary products by consumers. Within projects developing new technologies, such rapidity is a prerequisite for success. The use of concurrent engineering, where a number of firms and research groups are involved early in the research, development and commercialisation process, has been shown to assist in the development of new technologies. Within the context of high technology industry, the use of concurrent engineering by US technology firms has successfully brought forward key strategic decisions regarding the technology that may have otherwise hindered its diffusion (Zirger and Maidique, 1990). Doz (1986) noted the benefits of open alliances in developing technologies. He found that open projects (where partnership between firms are facilitated), technological innovation tended to cycle through iterations of learning, re-evaluation, and readjustment more rapidly than in closed projects. The pace of development of the extant technology is clearly in strong consideration in an environment of competitive standards setting (as is the case examined here). The potential for an open partnership system seems to be consistent with many of the aims of open standards, namely creating vendor commitment and rapidly getting products to the market. Further, for the lead firms who have made the choice to make the core technology freely available, it seems to offer the strong advantage of accelerating standard take-up, with limited risks.
8
9
Mowery et al. (1996) found that open partnerships in the development of technological products created specialisation among the member firms. It could be expected that the specialisation process would support the larger firms dominating core products and smaller firms tending to specialise in the development of peripheral or complementary products or services. Lead firms, who tend to be the founders of the basic technology, will be leaders in the development of products that meet with massmarket demand. All firms, though especially the leading firms, need to be mindful that within a cooperative alliance there is a need to manage the distribution of common and private benefits among and between firms. Khanna et al. (1998) noted from an empirical study that where the ratio of private to common benefit is high, competitive pressures negate much of the potential for shared learning. If a key objective of the open standard model is to accelerate development beyond that achieved by rival standards, this sharing of potential rewards needs to be carefully considered. Wireless Networking The cables connecting the variety of PCs, servers and peripherals that occupy a typical workspace are one of the banes of modern office environments. As wireless technology has developed, its potential to remove much of this confusion has become evident. Data services utilised within existing, broadly deployed technologies like GSM, for example, were developed to support voice based services and have tended to be very slow (14.4 kbps) in the delivery of data and thus not suited to the routing of large amounts of information able to be handled by cables. Relatively recent technological developments have created a quantum leap in the amount of data able to be transferred in a wireless mode. The use of digital devices such as data packeting and frequency modulation (time based and code based) has seen data rates of 1 mbps as the norm in many wireless applications, and 54 mbps (approx. 1000 times the data rate that is best achieved by a standard 56 kbps modem) achievable under controlled situations (Johnsson, 2000). Thus a strong imperative has emerged for the development of both standards and products to deliver wireless solutions to office, industrial and home environments. The Bluetooth Standard The Bluetooth communications standard is an interesting technology for many reasons. Bluetooth is a standard for short distance wireless communications, connecting electronic devices such as computers, peripherals and mobile and fixedline telecommunications devices at speeds up to 1 mbps and distances up to 10 meters. Bluetooth is a technology that exemplifies the potential of the converged information technology and telecommunications industries in delivering innovative and practical network-based solutions for business and personal users. The technology in its current iteration seeks to remove the need for cabling in office environments, an innovation that will have large benefits for commercial and home users of information technology equipment. Bluetooth in effect creates a piconet
9
10
around a given user, connecting up to eight compatible appliances. It is vertical in its architectural design in that it employs a master-slave relationship among appliances. Such an architectural design was a key attraction to the standard’s early backer LM Ericsson. By making the mobile telephone the central device in the Bluetooth architecture, (the mobile telephone holds key user details and facilitates access to the PSTN and existing wireless network), value addition could be developed within the core mobile telephony business of that firm. Strong internal competencies in both Bluetooth and mobile telephony will place Ericsson in a strong position to be a lead innovator as 3G systems emerge for mobile telecommunications, where data transfer rates will be far higher than currently experienced and interoperability between devices will be necessary for commercial success. As a technology that seeks to facilitate wireless connections between computers, peripherals and other high technology devices, Bluetooth is far from unique. The 2.45GHz frequency band has been reserved internationally1 for unlicensed use for industrial, scientific and medical (ISM) purposes. This frequency of the radio band is well suited for high rate data transmission within a local environment and a number of technologies have emerged that provide peripheral or direct competition for Bluetooth. Bluetooth’s technological nature, though similar to the competing wireless local area network (W-LAN) standards described below, has important differences in its potential uses and market. Bluetooth can best be described as cable replacement technology or Personal Area Network rather than a W-LAN technology. Many of the devices that Bluetooth seeks to interconnect may have very limited computing resources (digital cameras and mobile telephones, for example). This having been said, the W-LAN competitive environment was an important consideration for the Bluetooth consortium throughout the period of its development for a number of reasons. As Bluetooth’s acceptance as a proxy standard develops, similar products will need to ensure that at the physical layer at least (frequency allocation, data transmission rate) that there is a degree of interoperability between future standards and Bluetooth. This will in turn accentuate the value of the component expertise developed by Bluetooth consortium firms relating to aspects of the Bluetooth physical layer. It will potentially also facilitate commercial opportunities for proprietary and complementary additions to Bluetooth and other W-LAN standard that consortium members may have been developing. Standards Competition Within this competitive context, other developments in W-LAN technologies are noteworthy. While Bluetooth was never conceived as being able to provide a universal wireless solution for all data exchange needs within an office or commercial 1
The sole international exception is provided by the French military, which utilises the technology for certain local applications. Negotiations are being undertaken to remedy this problem within European fora.
10
11
environment, potential uses of W-LAN and Bluetooth do intersect. Bluetooth’s less ambitious technical requirements have created a more rapid development cycle than has been evident in competing technologies. This rapid cycle has assisted Bluetooth in becoming widely accepted, locking-in vendor commitment and extending the potential for proprietary modular innovations for peripheral uses. From a consumer perspective, Bluetooth has a number of potential benefits not shared by the more complex W-LAN technologies. Bluetooth’s chipsets will be cheap and relatively simple. Mass production of the devices, it has been estimated by Ericsson, will drive the cost to around USD 5.00 within five years. As such it will be feasible for Bluetooth technology to be incorporated into all electronic products, building significant positive network effects unavailable for the more expensive and complex alternative options detailed below. Bluetooth will also be easy to use. Most user interfaces will occur automatically, with Bluetooth enabled devices automatically recognising and configuring with one another when they come into range. IEEE 802.11 Among Bluetooth’s competitors in the higher end of the local data transmission market (connectivity between PCs and servers, for example) the IEEE 802.11 is the most notable and best established. It was first conceived in 1990, with final drafts of specifications released in 1997. Like all others in the area including Bluetooth, IEEE 802.11 technology uses packet switching to achieve data transmission rates of around 1 Mb/s. As technology has developed, IEEE engineers have developed later iterations of the standard, though at the cost of backward compatibility with earlier versions and appliances. IEEE 802.11 has suffered most from a lack of market focus. The second version of the standard (IEEE 802.11a) will not be available for use in Japan and Europe due to frequency reservations problems in those regions. While these problems were rectified in the latest technological iteration, such a fundamental market and technological mistake would never have been made by the international Bluetooth consortium. IEEE 802.11’s most serious problem has been lack of strong direction from and backing by commercial vendors. Conversely, this has and continues to be an area of strength for Bluetooth. While IEEE 802.11 has tried to develop solutions within the standard, Bluetooth’s rather more parsimonious approach to what is possible technologically in the early stages of release has created the potential for modular innovations later on. This has been attractive to a range of small, medium and large hardware and software vendors. A variant of the IEEE standard, IEEE 802.15 Personal Area Network is also in development and work is underway by this IEEE group to ensure interoperability between it and Bluetooth. This illustrates the emerging role of Bluetooth as a proxy wireless interconnectivity standard, at least at the physical layer (Palenchar, 2000).
11
12
SWAP The SWAP (Shared Wireless Application Protocol) also targets W-LAN environments. Interestingly from the point of view of this research, the HomeRF consortium that are developing the technology have also developed a consortium approach to encourage diverse participation by hardware and software vendors. This was something of a late decision taken in March 1998 as the technology was becoming increasingly marginalised among its competitors. The consortium of companies includes Siemens, Compaq, IBM, NTT and Nortel, among others. Membership between Bluetooth and HomeRF is not mutually exclusive. While Nokia and Ericsson are not members of Home RF, Intel, an early Bluetooth consortium member is. The SWAP standard is targeted at home users rather than office environments and thus provides peripheral competition to Bluetooth. Products that have been released so far have reflected this focus, and indications seem to show that target users will be users of multiple PCs in a home environment who wish to access single cable modem and other fixed line modem points within domestic environments. The standard builds upon the European DECT cordless standard and it has been developed to allow up to six simultaneous wireless voice lines, a feature that will be of increasing attractiveness as the takeup of Voice over IP and ISDN expands. Early products released using the SWAP standard have been quite expensive, aimed at higher end users and achieving more complicated tasks than Bluetooth is aiming to satisfy. Other Standards There have been a number of other variations developed, though none have achieved the degree of vendor or market support evident for Bluetooth. The European Telecommunications Standards Institute (ETSI) are developing a number of standards called HiperLan, though progress has been slow (Johnsson, 2000). Bluetooth Processes The cooperative nature of W-LAN and Bluetooth technological development sits as an exception in the standards setting processes within the broader wireless context. Wireless standards development processes in 3G mobile systems, for example, have been fraught with turf wars and court cases challenging patent and IPR ownership. Notable here was a lengthy conflict between Ericsson and Qualcomm (of the United States) regarding IPRs relating to CDMA technology. From the point of view of technology development, the process of developing Bluetooth has relied on developing networks between often competing firms. Managing patent claims is always difficult – managing them within a cooperative forum like Bluetooth would be a Herculean task. A solution emerged early in the
12
13
technology’s development due to broad acceptance by all participants to the relative openness of the standard. Such an approach tends to reduce the costs of the overall device by negating the need for royalty payments. It thus also increases the amount of opportunities for peripheral innovators and other late entrants due to the fact that profits tend to emerge from the development of innovative new products and applications rather than from ownership of the relevant protocol IPRs. The choice of the Bluetooth name, (in memory of Viking King Harald Blåtand who united a few warring Scandinavian tribes, for a short time, at the beginning of the tenth century), illustrates that the development of cooperation among competitors was an early and important element of the standard’s development process2. The potential for royalties in Bluetooth were always going to be negligible after the product was conceived as a low costs addition to virtually all electronic devices. Early proponents of the technology like Ericsson and Intel viewed the technology as a way of adding value within their core businesses rather than having profit potential in the long-term. Building a critical mass of vendor support was a key, early challenge for the technology’s backers. To assist in this process, the consortium was organized as a cooperative forum. Initially begun with five members, the “Special Interest Group” (SIG) now comprises members from mobile telecommunications, IT hardware and software, media, medical and industrial sectors. All members of the SIG have access to necessary protocols to facilitate the development of hardware and software for Bluetooth applications. Broader control of central elements of the standard (for example, changes in the protocol) is exercised by a “Promoter Group” (PG), currently comprising 3Com, Ericsson, Intel, IBM, Lucent, Microsoft, Motorola, Nokia and Toshiba. The cooperative nature of the PG and SIG is the result of an early strategic decision by early developers of the technology – and most notably Ericsson. While this firm was a founder of the Bluetooth consortium with Nokia, IBM, Intel and Toshiba, it was Ericsson that adopted the early technological champion role, undertaking essential early R&D which saw Bluetooth emerge from a competitive environment as the preferred wireless connectivity standard. Ericsson’s view of Bluetooth has been to see its potential to add value to and complement its core businesses of mobile and fixed-line telecommunications infrastructure and terminals. Ownership of the core Bluetooth technology was thus seen to be of secondary importance to the broader objective of Bluetooth’s ability to increase demand for Ericsson’s core products. The creation of the PG also required a little magnanimity on the part of Ericsson as both Nokia and Intel had similar projects underway, albeit at an earlier stage of development. The decision by these firms to
2
Historians are divided with regards to the applicability and relevance of King Harald as an exemplar of peaceful coexistence. The warring Vikings didn’t stay peaceful for too long and soon resorted to their pillaging ways.
13
14
back the Ericsson proposal would not have been possible had an emphasis on royalty payments been made at the early stages of the technology’s development. The entire developmental strategy also had a neat symmetry for Ericsson and Intel, the Bluetooth founders. The cooperative development consortium model, and the decision to make the entire standard openly available to all participants, supported the aims of: • building installed base within the marketplace and forestalling the growth of a competing, proprietary standard, • locking in vendor commitment through developing installed base among customers, • building competencies in and value for complementary technologies within the core businesses of PCs (in the case of Intel) and the mobile phone (in the case of Ericsson), • developing organizational competencies in key elements of the technology that will facilitate market leadership where interactions occur with their existing products. While the wireless interactivity market may offer great potential for these firms in the long term, such potential development will make itself felt within the firms’ core markets in the first instance. For Ericsson, the opportunity of engineering a strong role for mobile terminals through the Bluetooth technical architecture was of central importance. The decision to invite Microsoft to join the PG in mid-1999 (and their decision to accept) can also be seen in this context. Not noted for its commitment to open-source software and open standards, Microsoft’s active cooperation within the PG was aimed at ensuring Bluetooth compatibility with their proposed Windows CE operating system rather than developing proprietary value within the Bluetooth consortium. For Microsoft, joining the consortium provided an opportunity to bolster the flagging Windows CE operating system, or at least an opportunity to limit the process of marginalisation of their technology that the introduction of competing operating systems had begun. The Bluetooth standard’s development process can thus be seen as illustrative of the potential of cooperative approaches to standards development. The converged IT&T industries will increasingly require such approaches to develop new knowledge across traditional industry boundaries, thus the success of the Bluetooth cooperative model provides salient lessons for industry and policy makers with interest in the converging IT&T hardware, content and service provision industries. The following graph is illustrative of early developments in Bluetooth hardware and software. It is intended to be illustrative rather than comprehensive in its presentation of new product introduction. Ericsson was the first firm to introduce test chipsets for developers. This indicated their lead role in developing the technology and their advanced position in understanding the issues related to its operation. In the chipset segment of the Bluetooth market (the basic componentry end), a number of competitors soon entered. Ericsson’s depth of competencies, though, gave them a lead position in the development of high value added products like their wireless 14
15
headsets and also placed them in a strong position to enter alliances with industry leaders like Microsoft and Qualcomm to further develop software and hardware for Bluetooth applications. The graph also illustrates the entrance of both large and small firms in central and peripheral areas of the market. Both Philips and National Semiconductor leveraged their earlier DECT competencies in the development of Bluetooth chipsets. Nokia has entered into an alliance with Japan’s Fuji to develop software for the transmission of digital photographs through wireless means. Handspring was an early adopter of Bluetooth in their Palm compatible PDAs. Acer, the Taiwanese hardware manufacturer, has introduced Bluetooth adaptors for cellular telephones. Lego, the Danish toymaker, have signalled their interest in using Bluetooth to replace the infrared technology used in some high value toys. (Mayer, 1999; Prophet, 2000; Schneiderman, 1999). A key element of Ericsson’s strategy has focused on the development of firm alliances to leverage their Bluetooth competencies in the development of new and defensible innovations. Illustrative of this have been projects undertaken by Ericsson with Microsoft and Qualcomm to develop software and hardware respectively. Such an approach may lead to patentable innovations within growing markets in the future. The single-headed arrows in the following graph illustrate the direction of competence leverage in the processes of product development and alliance formation. Silicon Wave is a relatively small firm based in San Diego. The decision by TDK of Japan to enter into numerous development alliances with them has been based upon the competencies that the small firm had generated in the development of Bluetooth chipsets for various applications early in the technology’s development cycle.
15
16
Product Introduction within the Bluetooth Standard EARLY STAGE
Ericsson Test Chipsets
National Semiconductor Chipsets
Handspring Visor PDA
Philips Chipsets
Ericsson Headsets
Silicon Wave Chipsets
Acer Cellular Adaptors
Fuji/Nokia Imaging Software
Microsoft/Ericsson Software Ericsson/Qualcomm CDMA Terminals
Silicon Wave/TDK Modems
Lego Toys
LATE STAGE MASS MARKET Software
SPECIALISED PRODUCTS
Hardware
Sources: Mayer, 1999; Prophet, 2000; Schneiderman, 1999. Conclusions Changing technologies are having far-reaching impacts upon not only the type of products and services that are produced, but also the manner in which production occurs. These changes extend beyond the nature of the productive process to the nature of the economy within which production occurs. Firms like Ericsson and Intel, Bluetooth’s original promoters, viewed the emerging W-LAN technologies as peripheral to their core activities. They did, though, view the technology as being important enough to develop a strategy that would add value in their core businesses through extending overall demand for their products (especially in the case of Ericsson). Both firms also viewed the emergence of a proprietary WLAN technology with trepidation, due to the potential for either incompatibilities to emerge with their existing products or though royalty payment requirements being created going-forward. By making the core specifications freely available, the Bluetooth consortium created a strong driver for industry acceptance and innovative-firm takeup. Core competencies that the firms developed in the early stages of the standard’s development were not
16
17
lost as Ericsson, for example, entered into strategic alliances with other major and minor firms to develop patentable technology. Ericsson also used its competency base to be first entrant into innovative product consumer markets (like wireless headsets for their mobile telephones). The strategy adopted by Ericsson in this case seems to illustrate an interesting development in technology strategy formulation. The value of paradigmatic knowledge within the Bluetooth standard, it seems, will be made relatively freely available at the non-tacit level. At the tacit level, lead developers will continue to exercise some degree of control, with this control and tacit knowledge accumulation accentuated within the promoter group of the Bluetooth consortium. The value of product specific knowledge will emerge from application patents and modular innovations undertaken by the firms of the consortium in the future. Early examples in the paper illustrate the potential for lead firms to garner a share of these innovations’ benefits in the future through a range of alliances. The strategy employed adds value at the paradigmatic level through the openness of the standard and the flow on effects this has regarding installed base, vendor commitment and positive network externalities. The cooperative approach, led by Ericsson and other promoter group firms, tends to place them in a strong position to leverage tacit and deep knowledge of the standard and organizational reputation into value alliance relationships, again with a view to creating patentable and defensible product niches. Both of these characteristics of the paradigmatic knowledge held by the promoter group firms (its tacit nature and depth) are hard to imitate by the smaller firms and thus is particularly attractive to potential collaborators. In traditional competitive analysis relating to firm competencies, companies that have been able to exercise some degree of control within knowledge growth (learning) within a technical standard have also been able to maintain a position of knowledge leadership (Khanna et al, 1998). This has tended to be at the expense of competition among standards, where firms tend to commit to a particular paradigm to their eventual benefit or detriment. The strategy adopted by Ericsson in this case reflects a strategic trade-off to forego some of the potential benefits of knowledge control within a standard with a view of better controlling the way in which standards change and develop. Patent strategies at the application level, rather than at the paradigmatic level, have facilitated a broad range of potential revenues possibilities in the future.
17
18
References Amit, R. and Schoemaker, P.J. (1993) “Strategic assets and organizational rent”, Strategic Management Journal, 14 (1): 33-46. Baldwin, C. Y. and K. B. Clark (1997). ‘Managing in an age of modularity’, Harvard Business Review, 75, 5, 84-93. Barney, J.B. (1986) Strategic factor markets: Expectations, luck, and business strategy. Management Science, 32(10): 1231-1241. Collis, D. (1994) “Resarch note: How valuable are organizational capabilities?”, Strategic Management Journal, 15 (2): 143-152. Doz, Y. (1996) The evolution of cooperation in strategic alliances: Initial conditions or learning processes? Strategic Management Journal, 17(1): 55-83. Eisenhardt, K. and Schoonhoven, C. (1996) “Resource-Based view of strategic alliance formation: strategic and social effects in entrepreneurial firms”, Organization Science, 7(2): 136-150. Farrell, J. and Saloner, G. (1985) “Standardization, compatability and innovation”, Rand Journal of Economics, 16(1): 70-83. Garud, R. and A. Kumaraswamy (1993). ‘Changing competitive dynamics in network industries: An exploration of Sun Microsystems’ open architecture’, Strategic Management Journal, 14, 351-369. Granstrand, O. (1999) The Economics and Management of Intellectual Property, Edward Elgar, Cheltenham. Hagedoorn, J. (1993) “Understanding the rationale of strategic technology partnering: Interorganizational modes of cooperation and sectoral differences”, Strategic Management Journal, 14 (5): 371-385. Johnsson, M (2000) “Hyping up the wireless LAN”, Telecommunications International Edition, 34(1): 63-66. Khanna, T., Gulati, R., and Nohria, N. (1998) The dynamics of learning alliances: Competition, cooperation, and relative scope. Strategic Management Journal, 19(3): 193-210. Katz, M. and Shapiro, C. (1985) “On the licensing of innovations”, Rand Journal of Economics, 16(4): 504-520. Khanna, T., Gulati, R. and Nohria, N. (1998) “The Dynamics of Learning Alliances: Competition, Cooperation, and Relative Scope”, Strategic Management Journal, 19: 193-210.
18
19
Mayer, J. (1999) “Wireless Networking Swims Toward the Mainstream”, Electronic Buyers’ News, 1184, 57-63. Morris, C. R. and C. H. Ferguson (1993). ‘How architecture wins technology wars’, Harvard Business Review, 71, 2, 86-96. Mowery, D., Oxley, J. and Silverman, B. (1996), “Strategic alliances and interfirm knowledge transfer”, Strategic Management Journal, 17 (1): 77-91. Nonaka, I. and Takeuchi, H. (1995) The Knowledge Creating Company. Oxford University Press, NY. Palenchar, J. “Bluetooth”, Twice, 15(17): 22-24. Peteraf, M.A. (1993) The cornerstones of competitive advantage: A resource-based view. Strategic Management Journal, 14 (3): 179-191. Prophet, G. (2000) “Living in a Wireless Wonderland”, EDN, 45(12): 79-94. Robertson, P. (1996) “Competences, Transaction Costs and Competitive Strategy”, in Foss, N. and Knudsen, C. (1996) Towards a Competence Theory of the Firm, Routledge, London and New York. Rumelt, R.P. (1991) “How Much Does Industry Matter?”, Strategic Management Journal, 12(3): 167-185. Sanchez, R. and Heene, A. (1997). ‘A competence perspective on strategic learning and knowledge management’, in R. Sanchez and A. Heene (Eds). Strategic Learning and Knowledge Management, Chichester, UK: John Wiley and Sons. Scheiderman, R. (1999) “Bluetooth projections push product development”, Microwaves and RF, 38(13): 9-14. Stuart, T.E. (2000) “Interorganizational Alliances and the Performance of Firms: A Study of Growth and Innovation Rates in a High-Technology Industry”, Strategic Management Journal, 21, 791-811. Teece, D.J. (1986). ‘Profiting from technological innovation: Implications for integration, collaboration, licensing and public policy’. Research Policy, 15, 285-305. Teece, D., Pisano, G. and Shuen, A. (1997) “Dynamic Capabilities and Strategic Management”, Strategic Management Journal, 18(7): 509-533. Zirger, B. and Maidique, M. (1990) “A model of new product development: An empirical test”, Management Science, 36(4): 867-883.
19
