Document View

Immersed in a global industry consolidation process, corporate managers are witnessing, in recent years, the proliferation of inter-organizational collaborative agreements, which aim to develop, manufacture and commercialize knowledge intensive products. The decision within a knowledge management (KM) framework to collaborate in knowledge sharing networks becomes a complicated issue, since such a decision needs to be made often under conditions of uncertainty and irreversibility. This study deals with questions such as why, how, and when to be a member of a knowledge network and provides some empirical evidence about the formation of inter-organizational networks in knowledge intensive industries.

Importance of knowledge networks

The formation of knowledge networks becomes an important research subject in today's turbulent environment where announcements of technological partnerships between knowledge-intensive companies abound. In order to avoid the risk of falling into technological obsolescence and knowledge erosion, companies have no other choice than to invest large sums of their budgets to remain competitive in the market.

In order to survive and, what is more challenging, to enhance competitive advantage, firms must possess a knowledge base and capabilities which add value to the firm; resources which are inimitable, nonsubstitutable and scarce. Some authors argue that profit heterogeneity comes precisely from the diversity in firm assets and skills, since each company is unique and is the result of an idiosyncratic historical trajectory. Our current competitive landscape and the trend towards globalization require new forms of organizational structure and interrelationships between different stakeholders. Few firms are endowed with a sufficiently sound knowledge base to continuously innovate and guide the course of new technologies. The present competitive environment asks for new interorganizational arrangements which must respond to an increasingly sophisticated market demand (Von Krog and Roos, 1996). In addition to strengthening internal capabilities, corporate managers need to develop partnership skills in an industry environment where interdependencies among organizations emerge naturally.

About a decade ago, Pisano (1991) pointed out that "knowing when to integrate vertically, when to collaborate, and when to license is likely to be a critical managerial skill required for both new and established firms to succeed in biotechnology". After a period full of interesting partnership experiences in this sector, we can observe that Pisano's assertion was quite accurate.

An integrated knowledge management approach, IKM, should mix together firm internal core competencies with interorganizational extensions to absorb and transfer knowledge beyond the boundaries of a firm. Companies are no longer selfsufficient for the creation, development and commercial exploitation of their knowledge base and, consequently, seek interorganizational networks in order to succeed in their respective technological fields. In the so-called new economy, examples of knowledge networks in the biotechnology, telecommunication, and other high-tech industries are numerous. To mention just a few, Motorola, Novartis, and Toyota have relied on knowledge sharing networks, to market their products successfully and, more importantly, to strengthen their core competencies and knowledge bases over time.

In this flurry of strategic alliances, a recent interesting phenomenon has emerged, namely, the formation of coalitions and expansion towards specialized conglomerates. Competition has become not only an issue of rivalry among firms, but also an issue of rivalry among inter-organizational networks and contending knowledge platforms (or technological fronts). Corporate managers should be aware that if they decide not to participate in knowledge networks, they are restricting access to relevant information shared by other companies, and what may be worse, they could risk losing their current market position.

In such a turbulent industry environment, knowledge-intensive firm executives must make critical business decisions. They need to decide whether or not to become a member of a particular knowledge network (or family of companies sharing the same knowledge base; usually explicit technical knowledge such as patents, manufacturing processes, data banks, ... , which may determine the future technological path of an industry) by taking into account the potential for high profits and the presence of high market uncertainty. Furthermore, if corporate managers decide to become members of a network, other no less important questions follow: which knowledge network must be chosen, when, how, and why? This article is an attempt to address these interesting questions related to knowledge networks as part of an integrated KM framework and to show some empirical evidence from knowledge intensive industries.

The paper contains the following sections. The next section deals with a brief description of relevant findings in the literature regarding the formation of inter-organizational networks. An integrated KM model is proposed in the following section, a model which integrates both internal and external knowledge development elements. The penultimate section presents some empirical evidence from inter-organizational networks developed in knowledge intensive industries. Finally, main conclusions and implications are highlighted.

Building knowledge networks

Many knowledge intensive industries are characterized by numerous and complex inter-organizational relationships which result from spot market failures and transactional difficulties. As it is well known, the transfer of knowledge entails serious difficulties. Transaction cost economics is a research stream, which draws on organization theory and takes the transaction, rather than the firm or market, as the basic unit of analysis. Within this conceptual framework, the governance form embedded in a specific knowledge network may range from markets (i.e. transfer of knowledge via products) to internal organization (i.e. full ownership of network members). According to Williamson (1975), factors such as, opportunism, uncertainty, asset specificity and product complexity will determine the optimum governance form to consummate an inter-firm transaction. In some instances, firm managers refuse to share proprietary knowledge and prefer not to participate actively in a knowledge network. Often, all the rights and obligations outlined in contracts to transfer, develop and utilize knowledge by network members are not enough to eliminate the potential opportunistic behavior of partners.

Organizational integration is viewed as a method for overcoming some of these problems, namely transaction costs problems, associated with imperfect longterm contracts (Shane, 1994). A possible outcome of a knowledge network could be the full acquisition of the rest of network members by a firm. The main objective advocated by the transaction cost rationale is to overcome the problem of market failure and to find among different interorganizational arrangements the one that minimizes transaction costs and opportunistic behavior.

Many scholars agree that uncertainty is a major impediment to collaboration in knowledge networks. One can classify uncertainty sources in two main groups: technological uncertainty and resource uncertainty. Technological uncertainty comes from the lack of knowledge about the viability of a shared technological project. Firm managers do not know ex ante whether or not a technology can be developed, and even less, whether or not the product derived from that project will reach the market. Technological uncertainty will be reduced once the project is undertaken and significant advances are made. The perception of uncertainty is subjective. Network members may have the same information but not necessarily the same knowledge about the progress and potential validity of a shared technological project. Consequently, each network member will reduce technological uncertainty at his own pace and, obviously, will benefit distinctly from that network.

Resource uncertainty comes from the conflict arisen due to asymmetric information about the adequacy of the network partners' pool of resources and knowledge base. One way to reduce resource uncertainty is by developing a long-term relationship with a partner where investment in the technological project is completed gradually by following an option approach (Kogut and Kulatilaka, 1994). As common knowledge about each partner's capabilities and skills is gained over time, a member is in a better position to commit him/herself and collaborate further in the network.

Transaction cost theory seems to be an important, but yet an incomplete, framework to examine knowledge networks. An integrated KM approach also needs to include the strategic motives for the formation of knowledge networks. In recent years, firms operating in knowledge intensive industries (i.e. biotechnology, telecommunications, life science, etc.) have become involved in a myriad of inter-firm networks. Instead of exclusively reducing transaction costs, knowledge networks can be understood as cooperative agreements carried out among firms with the aim of strategically gaining and preserving a leadership position in the marketplace. In fact, transaction cost savings may not be as critical as the acquisition of knowledge, capabilities, and technology allies from the knowledge network (Osborn and Hagedoorn, 1997).

Market leaders may incorporate a company in their network simply to prevent other industry rivals from having access to the knowledge base of the incoming company to the network. Securing network partners' assets and making them unavailable to competitors would be expected to enhance a firm's competitive position. This can be particularly true in concentrated industries, where leading firms are clearly more motivated to preserve market power and better endowed to follow this type of strategy. In some instances, knowledge networks developed by successful firms may determine the future of certain industries (i.e. push effect), whereas in other instances, inter-organizational agreements might become a sine qua non condition to survive and thrive in the industry (i.e. pull effect).

Corporate managers may wish to form networks with other companies in the same industry sector (i.e. horizontal interorganizational network) either to reduce costs through synergy advantages or to improve bargaining power through increasing industry concentration. A horizontal knowledge network could be a powerful platform to intensify a new technology in a particular industry (i.e. a technology derived from the knowledge that network members have developed together) and an effective scheme to appropriate technology rents in the long run. Furthermore, the trend towards a higher industry concentration in several knowledge intensive sectors may drive firms to pursue inter-organizational horizontal networks in order to remain competitive and survive in the sector.

On the other hand, vertical agreements seem to be common when corporate managers look for technological complementarity, that is, a rapid access to lacking knowledge resources. As Kogut (1988) pointed out in his study, joint ventures are an appropriate form of agreement for the transfer of organizationally embedded knowledge, which cannot be easily blueprinted through licensing or market transactions. In these instances, the partnership usually takes a more complex organizational governance form than just a long-term contractual arrangement. In contrast, non-equity or simpler organizational arrangements are expected to provide a better environment for discoveries, but they do not seem to promote the absorption of knowledge by the initiator of the alliance as effectively as more hierarchical administrative forms appear to. A vertical inter-organizational agreement, in either of its up-stream or down-stream forms, becomes an interesting alternative for the extension of the knowledge base of a company by efficiently acquiring critical functions, capabilities and knowledge from network partners that such a firm may lack.

Timing to enter a knowledge network is another important aspect for strategic decision making. A first mover strategy may bring monopoly rents to a company for being the first in introducing an innovative knowledge intensive product in the market. Companies operating under this strategy are often eager to participate in knowledge networks and devote abundant resources to developing innovative technologies faster than competitors. As Prusak (1996) points out on this matter, "the only thing that gives an organization a competitive edge - the only thing that is sustainable - is what it knows, how it uses what it knows and how fast it can know something new".

Managers implementing a wait-and-see strategy show a different behavior. That is, managers may prefer to gain additional knowledge regarding new market opportunities by observing what rivals do and the success level achieved in their moves. Once uncertainty is reduced, corporate managers are in a better position to make a decision about collaborating or not in a particular knowledge network. Therefore, the decision about when and also the velocity (i.e. how rapidly to enter new networks) to form interorganizational agreements offer some hints about the types of strategies that corporate managers are implementing for knowledge acquisition, creation and commercialization.

A simplified integrated knowledge management framework

A knowledge network can be defined as an inter-organizational agreement to share knowledge among network members for the exploration (i.e. creation and development) or exploitation (i.e. product transformation and commercialization) of new technologies. An integrated knowledge management framework must comprehend both internal and external elements to create, develop, and exploit knowledge as it is shown in Figure 1. Like information, knowledge can be managed inside the organization, but it can also be shared with other organizations (i.e. bidirectional transmission of knowledge, "INOUT").

One can distinguish two main reasons to engage firmly in networks: first, to reduce transactional hazards and, second, to achieve a strategic market-technology position. From an internal perspective, the enhancement of firm resources and capabilities becomes a critical issue to sustain competitive advantage over immediate rivals. As described earlier, these resources and capabilities must be unique, inimitable, non-substitutable, scarce, and most importantly, must generate value to the firm over time. Both tacit and explicit knowledge embedded in individuals, divisions and organizations are valuable resources that may satisfy these properties. As stated by Bhatt (2001), "organizational knowledge is formed through unique patterns of interactions between technologies, techniques, and people, which cannot be easily imitated by other organizations, because these interactions are shaped by the organization's unique history and culture".

Enlarge 200% Enlarge 400%

Since the knowledge base possessed by most organizations does not seem to suffice to face current competitive challenges (i.e. rapid pace of product innovation, complexity of modern technologies, new customer needs, etc.), corporate managers usually seek the knowledge they lack from external sources. From an external viewpoint, interorganizational agreement decisions must include at least three aspects: who is the partner, what is the involvement of the firm in the knowledge network, and when is it appropriate to enter the network. The answer to these central issues will determine the strategy to explore new knowledge sources or to exploit a knowledge base with network partners (Grant, 1996).

In summary, an integrated knowledge management framework cannot be complete without any of the two perspectives (i.e. internal and external). The internal knowledge base of an organization must be complemented with partnering practices and the involvement of other knowledge network members. The uniqueness of each firm's networking configuration is expected to shape firm core competencies and, ultimately, to determine business profitability in the long run.

Evidence from knowledge-intensive industries

Major US and European companies from the life science and biotechnology industry sectors are examined in this study, industries in which R&D projects (i.e. technical knowledge based projects) and interorganizational networks have proliferated. The companies selected for this study develop, process or commercialize knowledge intensive agricultural products such as transgenic crops, nutraceuticals, genetically modified organisms, etc. These firms are public and have completed at least an inter-organizational agreement during the 1994-1997 period in order to acquiredevelop a knowledge base or to exploitcommercialize a knowledge intensive product. In total, a sample of 615 knowledge-based agreements is examined. The selection of this time-frame (i.e. 19941997) is appropriate as many of the most critical networks formed in these industry sectors date from this period and many biotechnology products started to reach the market during these years (Bio, 1998).

Particular attention is centered on those companies developing new technologies in biotechnology and life science markets, companies involved in R&D projects and alliance agreements with other firms during the period of study. This includes companies operating in different industry sectors (i.e. US three-digit SIC codes: 200, 286, 287, 519, 873). The aim is to find out how firms get involved in knowledge networks:

What is the extent of their commitment in the network? ("Commitment".)

With whom are they partnering? ("Partners".)

When do they enter the knowledge network? ("Timing".)

Industry evidence from knowledge intensive industries is summarized in Table I. In particular, results from this table provide useful insights regarding the commitment and partnership characteristics embedded in knowledge networks.

"Commitment" in the knowledge network

In this study, three commitment categories are distinguished for a knowledge network membership: non-equity agreement, majority-equity agreement (over 51 percent of control over the network partner) and acquisition of the network partner. Evidence from the sample suggests that acquisitions, 35 percent of all agreements of the sample, are common when the collaboration consists of exploiting a knowledge base (i.e. manufacturing process of an invention or commercializing a knowledge intensive product). Only 13 percent of acquisitions dealt with R&D projects. Uncertainty involved in knowledge-exploitation agreements is lower than in knowledgeexploration agreements where firms face higher technological uncertainty. During this period 1994-1997, hostile investment strategies of companies, like Agribiotech Inc., seemed to be mostly oriented towards the manufacturing and commercialization of their patented and trademarked products resulting from technological innovations created or developed in the past.

Enlarge 200% Enlarge 400%

It appears that knowledge and experience gained by firms in completing business acquisitions increase the likelihood of pursuing a new organizational arrangement of this type. Business acquisitions are more popular in firms with a larger number of acquisitions conducted in the past. Nevertheless, the financial condition of acquirer companies is not as sound compared to the companies that pursue less aggressive investment strategies. On average, the debt ratio of acquirer companies suggests that their networking strategy is funded with a relatively larger portion of external financial sources.

Companies pursuing equity-based networks, 8 percent of all the sample agreements, portray different firm attributes. They are larger in size, and in addition, they are better endowed with research and learning skills. Prior to 1994, these companies owned a larger number of patents relative to firms pursuing non-equity or acquisition agreements. As in acquisitions, 80 percent of the majorityequity agreements are formed with manufacturing and commercialization purposes. The majority of these agreements are adopted primarily by companies of the organic-chemical industry sector (i.e. 43 percent of controlling equity agreements). One may suspect that some of these agreements are motivated by an exploratory investment conduct by which chemical and biotechnology firms have some control of new technologies. In contrast to the high commitment strategy implemented by Monsanto via acquisitions, companies like Dow Chemical, Hoechst, Merck, Zeneca or DuPont are examples of networking strategies with lower commitment levels.

Non-equity knowledge networks, 56 percent of the sample, seem to be associated with smaller firms conducting mostly R&D agreements, with an inferior competitive advantage regarding research and technological learning skills, and almost no experience in pursuing business acquisitions. Almost 80 percent of the nonequity agreements for biotechnology firms are sought for R&D purposes. Apparently, these firms are still young and lack the pool of tangible and intangible assets owned by other firms. The risk to which they seem to be exposed appears to be higher relative to other firms (i.e. higher technological and resource uncertainty) and it is not surprising to observe a low commitment and dependence on the knowledge network. Clearly, R&D based partnerships are positively associated with a non-equity network membership.

"Partners" of the knowledge network

Regarding the form of the knowledge-based partnerships, two types of network partners are recognized: partners from the same industry sector (i.e. horizontal agreements) and partners from different industry sectors (i.e. vertical agreement). In the study sample, 56 percent of the agreements are vertical knowledge networks. The remaining horizontal inter-firm relationships seem to be characterized by manufacturing and commercialization types of agreements. Advantages derived from knowledge exploitation and synergy effects might serve to explain the agreements of this nature.

Interestingly, the formation of horizontal inter-organizational networks seems to be more likely in industries showing high concentration ratios (i.e. food manufacturing and wholesaling industry sectors). Certainly, follower firms will collaborate with other industry companies to reach the market position achieved by industry leaders. But leading firms may also collaborate with other industry companies to enhance their competitive position, to achieve scale economies, to strengthen their bargaining power or to preserve their leadership through the consolidation of a shared technology. Regardless the type (i.e. leaders or followers) and motive of companies (i.e. efficiency, market power, technology) to pursue horizontal agreements, what seems obvious is that firms still continue to respond to the consolidation process initiated in many knowledge intensive sectors at the end of the last decade.

Vertical agreements seem to be more likely when the dominant party operates in the initial stages of the value chain (i.e. knowledge creation). For instance, managers from biotechnology companies apparently prefer vertical agreements forwards (i.e. 87 percent of agreements from biotechnology firms) to develop their innovations with more modest R&D firms or manufacturing companies, since they usually lack the complementary resources (i.e. production facilities and operation management skills) needed to exploit their core R&D capabilities.

"Timing" for entering a network

Timing of being a member of a knowledge network is measured by the velocity of a firm in becoming part of different networks over time. Figure 2 shows that firms from the agbiotechnology, food manufacturing and agwholesaling industries are the ones that seek new partnerships at a faster pace. In fact, the average of months from a given network partnership to the next partnership is about six to seven months. Companies in the chemical industries seem to invest following a wait-and-see type of strategy, or perhaps, considering their well-endowed base of knowledge, resources and capabilities such firms are not pressed to pursue further collaborative agreements so intensely.

As firms from this sample gain experience in completing majority-equity based investment agreements, such firms seem to conduct further investments at a faster pace than firms without this experience. An interpretation of this phenomenon may be that this investment behavior is the result of a deliberate corporate strategy. Agricultural biotechnology and food technology markets comprehend few specialized biotechnology companies, and outsider investor firms compete to have access to their R&D capabilities. Accelerating investment and gaining access to such R&D capabilities through majority-equity based investments would prevent rivals benefiting from the research skills brought by the biotechnology partner company (Pena, 1999).

Enlarge 200% Enlarge 400%

This investment strategy of continuous majority-equity investments could represent the perception of increased risk of being blocked by competitors to the highly valued knowledge base of a particular firm. Particularly, companies with a greater risk of technology obsolescence would accelerate the timing for entering knowledge networks. For instance, the survival and growth of biotechnology companies in particular depend on the ability of such firms to absorb and exploit new technological advancements. The necessity of keeping abreast of technological innovations creates an incentive to enter networks at a faster pace than other less knowledge intensive companies.

Moreover, biotechnology firms are highly motivated to produce and commercialize the product to be the first to reach the market, since this would allow such firms to recoup their R&D investment in a more profitable manner (i.e. first-moving advantage). Since few biotechnology companies own manufacturing and distribution assets to market their knowledge intensive products, most of them are pressed to form contractual agreements and/or alliances with other firms owning such assets. Therefore, due to peculiar industry characteristics (i.e. continuous and disruptive innovation, lack of vertically structured firms, undefined government regulation) biotechnology firms may find themselves forced to pursue a fasttrack investment strategy in order to complete the whole value chain corresponding to their main R&D activity.

Conclusion and implications

This study has been an attempt to assemble an integrated knowledge management framework, with special emphasis on knowledge networks. Empirical evidence from this study indicates that in addition to transaction-specific factors, firm and industry factors are relevant in determining the knowledge networks. Strategic thinking (i.e. how, with whom and when to enter knowledge networks) seem to be important for examining inter-organizational agreements in knowledge intensive industries

This study provides several implications about how corporate managers may participate in knowledge networks. The first implication for corporate managers is that the choice of an optimum networking strategy (i.e. commitment, partnership and timing for an inter-organizational agreement) does not depend exclusively on the goal of minimizing transactional costs, but also on strategic objectives (i.e. exploration of new technologies and partners; exploitation of technological rents through market power, preemption, retaliation; expansion towards new industry segments through business conglomerates etc.).

Regarding the commitment of a knowledge network, corporate managers should expect an acquisition-type of agreement to be more likely to occur when the exchange involves a manufacturing or commercialization agreement as opposed to a R&D agreement. The acquisition investment outcome is also more likely when the acquirer has been operating for a long period in the industry, and has experience and resources for acquiring companies. Managers should expect that industries where little extrinsic uncertainty prevails, like the mature food manufacturing sectors, are very appropriate to complete business acquisitions. Advantages from synergy effects or strategic positioning in a consolidating industry may be reasons to explain such an investment behavior. The exploitation of technological rents would be the main priority of this type of investment strategy when forming certain types of networks.

Vertical agreements seem to be common among biotechnology companies for sharing R&D and having access to manufacturing capabilities. These agreements are usually non equity-based agreements and are apparently motivated by exploitation reasons or, in some instances, by an exploratory reason (i.e. to become familiar with new technological advancements of other firms). Business investment decisionmakers in general appear to value high managerial flexibility; particularly in companies operating in uncertain industry environments.

As far as network timing is concerned, with the exception of chemical companies, the rest of firms seems to behave following a firstmover strategy. Biotechnology firms enter fast knowledge networks because they want to accelerate their technological innovation process. Manufacturing and wholesale firms are impatient to obtain knowledge intensive products and introduce them into the market. In contrast, chemical companies project their future over a much longer period and participate in knowledge networks following a slower and more selective partnership process.

This study also offers some implications for policy makers. Recent works in the literature suggest that not only inter-firm competition, but also cooperation seem to be critical for the development of new technologies in knowledge intensive industries (von Krog and Roos, 1996). Findings from this study suggest that regarding concentration and antitrust issues, government authorities should be less concerned with unstable industry sectors (i.e. industries with higher uncertainty such as agricultural biotechnology and chemical industries) and more concerned with more stable industries (i.e. like the agricultural wholesaling and manufacturing sectors). That is, knowledge-exploiting companies, rather than knowledge-exploring companies, are the ones which pursue a relatively larger number of business acquisitions and, therefore, contribute most to industry concentration.

Further research should address questions such as the way a knowledge base is formed and developed over time and the identification of determinants of knowledgeexploration or knowledge-exploitation behavior. Moreover, it will be interesting to find out whether participation in knowledge networks is related to business performance. Answers to these questions would shed some light on more efficient and productive knowledge management.