Informal Modes of Knowledge Acquisition
Though existing research has emphasized the role of formal modes of external learning such as alliances (Contractor and Lorange, 2004; Inkpen, 2002; Mowery, et al., 1996) and acquisitions (Ahuja and Katila, 2001; Ranft and Lord, 2002; Schweizer, 2005), there is also significant evidence that knowledge is transferred across firms by informal means that may often not be easily identifiable and may not be linked to any formal organization level relationship or contract. From Porter’s (1998) description of inter-firm knowledge flows in the Italian ceramic tile industry, to Saxenian’s (1991) rich ethnography of semiconductor engineers sharing knowledge in Silicon Valley, and Liebeskind et al.’s (1996) description of communities of practice in biotechnology, informal mechanisms of knowledge transfer seem to play an important role in facilitating organizational learning. In her seminal work discussing the ‘invisible college of scientists,’ Diana Crane (1972) recognizes the fact that communities exist across organizations and these communities facilitate knowledge flows between their members. These communities have a strong social dimension (common language and norms) that governs the flow of knowledge between individuals often across firms, universities, and research institutions. These links act as informal bridges across firm and geographic boundaries (Allen and Cohen, 1969). Though the importance of these communities to inter-firm knowledge flows has been commonly acknowledged, there is need for further conceptual and empirical research that assesses the implications for organizational innovation and learning. In the next section we review the main modes of informal knowledge flows across firms facilitated by social communities (including those linked to individual mobility, ethnicity, geography, board inter-locks, and communities of practice) and subsequently evaluate the implications of these knowledge flows for innovation and learning. The informal social communities are not mutually exclusive—indeed they often overlap and build on each other—but each has a significant stream of research associated with them.
Mobility, hiring, and knowledge
An important mechanism for the transfer of knowledge across firm boundaries is through the mobility of people. Several primarily descriptive studies suggest that people are an important conduit of inter-firm knowledge transfer (Malecki, 1991). Most early research suggests only a connection between mobility and knowledge flows, offering, at best, indirect evidence. For instance, Markusen et al. (1986) find that regions with high concentrations of technical workers attract new high technology investment. In technology intensive industries as well, there are numerous descriptive studies of people carrying knowledge across firms (Hanson, 1982). In the semiconductor industry, interviews with engineers reveal many anecdotes of inter-firm knowledge flows associated with the mobility of engineers (Saxenian, 1991).
The most direct evidence linking mobility of engineers to inter-firm knowledge transfers may be accomplished through patent records. Almeida (1996) shows that after a semiconductor firm hired a new engineer, there was a significantly greater tendency for the hiring firm to cite the prior patents of the newly employed engineer than would be expected given its technology profile. In addition, Song et al. (2001) demonstrate that during the early stage of development of Korean semiconductor firms, the practice of bringing US-educated and US-employed nationals back home leads to similar patenting practices. An important mechanism for the transfer of knowledge across firm boundaries is through the mobility of people. Song et al. (2003) examine the conditions under which learning-by-hiring is more likely to be successful. Using patent applications from software engineers who moved from US firms to non-US firms, the authors demonstrate that mobility is more likely to result in inter-firm knowledge transfers if the hiring firm is less path dependent, if the hired engineers possess technological expertise distinct from that found in the hiring firm, and if the hired engineers work in non-core technological areas in their new firm. They also demonstrate that domestic mobility and international mobility are similarly beneficial to learning-by-hiring.
How does the knowledge obtained from mobility help firm innovation? Rosenkopf and Almeida (2003) demonstrate that the effectiveness of mobility increases with technological distance and is particularly helpful to overcome the challenge of learning from geographical distances. Tzabbar (2009) shows that the recruitment of experts from other organizations is positively associated with technological repositioning particularly in the case of hiring scientists who are experts in areas in which the firm does not have existing competence. Internal factors like breadth and the existence of star researchers moderate this relationship. In a recent study, Corredoira and Rosenkopf (2010) find that hiring can result in bi-directional knowledge flows and learning. Not only does the hiring firm gain knowledge but the firm that lost employees also appears to learn from the hiring firm. This suggests that mobility facilitates the formation of social relationships across firms that then permit the two-way flow of knowledge. Furthermore, the authors find that outbound mobility is a particularly relevant knowledge channel between geographically distant firms, but its importance decreases for geographically proximate firms since other knowledge channels exist within regions. Similarly, looking at the international dimension of inventor related knowledge flows, Oettl and Agrawal (2008) find that the inventor’s new country and firm gains from the inflow of inventors and that there is also a reverse flow of knowledge. These studies suggest that hiring across organizations helps creates conduits for knowledge transfer across them, and perhaps even leads to the formation of inter-organizational and inter-country social communities that facilitate the flow of informal knowledge.
Geographically mediated communities and knowledge
Research points to the importance of geographic proximity in facilitating knowledge flows through the formation of spatially clustered social networks (Rogers and Larsen, 1984). Localized knowledge sharing has been a well observed phenomenon through history. Allen (1983) points to the importance of knowledge sharing across organizations as a force in maintaining the supremacy of the nineteenth century steel industry in England. Case studies of regional clusters in Italy (Piore and Sabel, 1984) and Baden-Wuerttemberg in Germany (Herrigel, 1993) indicate extensive knowledge flows through networks in these regions. Why does co-location matter to the transfer of knowledge? Common to Alfred Marshall’s (1920) ‘industrial districts’ and Porter’s (1998) localized industry ‘clusters’ is the idea that industry-specific knowledge develops in geographically concentrated locations. This phenomenon is true not only of traditional, craft-based industries, but also of high-technology industries. This in turn leads to greater knowledge transfers between firms, due to the similarity in their knowledge bases and to the extensive linkages that develop within a region. In a recent article, Tappeiner et al. (2008) suggest that the geographical clustering of innovation may be best explained not so much by the direct flows of knowledge across neighboring organizations but by the underlying social networks and the commonality in the shared cultural and institutional setting that organizations and their employees share. Hence, some of the advantages of geography to knowledge acquisition and innovation may be indirect.
A question worth investigating is whether these geographically-mediated knowledge flows will grow less important in the face of globalization. For instance, Ponds et al. (2010) find that knowledge spillovers between universities and firms do take place locally but are not restricted to inter-regional flows. The authors do not separate formal and informal knowledge flow across organizations. In an extensive study of patents, Sonn and Storper (2008) find that the localization of inter-organizational knowledge flows has increased across time. Perhaps as more information becomes available to the individual and organization, local communities embedded in geography act as filters to highlight what knowledge is viewed as salient and therefore relevant to innovation and learning. In fact, Alcácer and Chung (2007) suggest that this phenomenon of geographically mediated informal knowledge flows is used strategically by multinational firms when deciding the location of activities to maximize knowledge inflows and minimize outflows.
So why do localized knowledge flows exist? Though linkages between firms could develop across geographic distances, proximity enhances the development of complex networks (Almeida and Kogut, 1999). Locational proximity reduces the cost and increases the frequency of personal contacts which serve to build social relations between players in a network (Saxenian, 1991; Zucker et al., 1998) that can be appropriated for learning purposes. Further, proximity builds common institutional and professional ties that help construct a context for knowledge transfers (Saxenian, 1991). In fact, Saxenian (1991) relates the dynamism and the vitality of Silicon Valley to the extensive networking both at the firm level (between firms and universities, buyers and suppliers, venture capitalists, etc.) and between individuals within the region. Porter’s (1998) description of the localized Italian ceramic tile industry points to close and repeated interactions between the various small businesses in the region.
Ethnic communities and knowledge
Aldrich and Waldinger (1990) describe an ethnic community as having members of common culture and origin who are aware of their membership in a group. Ethnic social communities confer the benefits of social interaction, common value systems, and trust based relationships that facilitate social cohesion that can enhance the economic success of its members (Iyer and Shapiro, 1999; Tsai and Ghoshal, 1998). One well-researched area in economic sociology is the entrepreneurial role played by ethnic groups (Greene and Butler, 2004). In technology intensive industries, scholars point to the role of ethnic communities in facilitating not just entrepreneurship, but also innovation (Saxenian and Hsu, 2001). Saxenian (2002) believes that ethnic communities offer a flexible mechanism for transferring knowledge between participants even across distant regional regions. Similar to arguments made by Light (2002), she posits that immigrants often view themselves as outsiders to the mainstream community and consequently foreign-born engineers and scientists forge social relationships based on their national identity that enable the exchange of information and know-how. This would suggest that membership in ethnic communities should enhance an individual’s and, therefore, his or her organization’s innovativeness.
However, membership in an ethnic community could be a double-edged sword. After all, being embedded in any social community provides both opportunities and constraints (Uzzi and Lancaster, 2003). Porter (1998) suggests that the sense of altruism is especially strong in ethnic communities and helps bind community members together but this could lead to over-embeddedness. Karra et al. (2006), in their case study of Balkan immigrant communities, find that the strong sense of solidarity in the community does lead to over-embeddedness. Individuals are tied so strongly to the expectations of others in the community that their relationships with other non-ethnics are constrained (Bowles and Gintis, 2004) and they do not break away from these constraints due to the solidarity norms that bind the community. Porter (1998) cites the example of the narrow lines of business practiced by San Francisco’s ethnic Chinese community to suggest that in many cases ethnic communities force solidarity on their members and the current practices and ways of thinking stifle the availability of new knowledge. This solidarity could take place to the extent to which an individual’s ability to innovate or seek new ways of doing things is suppressed. Another way ethnic communities may stifle innovation is that, due to specialization and local learning, there could exist a lack of diversity of knowledge and expertise within the community, and therefore knowledge, resources, and competences obtained from the community could have limited usefulness.
Similar to the idea of path dependence in evolutionary economics, sociologists since Becker and Granovetter have referred to the idea of cumulative causation whereby historical decisions and actions determine future possibilities. Of course community norms and expectations can play a role in forming and directing the actions of individuals along particular paths. Waldinger (1994) attributes the dominance of Egyptian and Indian engineers in the New York City bureaucracy to cumulative causation where historical actions by early community members lead others to view possibilities and opportunities through a narrow historical lens. This leads to a continued reliance on community knowledge and ideas even when opportunities elsewhere may exist. The case study by Karra et al. (2006) suggests that this can lead to lock-in. They observed that individuals influenced by habit, social expectations, and limited world views continued to be a part of the ethnic community long after they played a constructive role or after it was useful. In a recent study, Almeida and Phene (2010) analyzing innovation influences of immigrant Indian inventors in the US find that this informal knowledge can have both positive and negative effects on innovativeness. While at low levels of interaction, ethnic communities can enhance innovation, as their influence increases innovation can actually be suppressed.
Communities of practice and knowledge
Lave and Wenger (1991) gave birth to the concept of communities of practice while examining apprenticeship as a learning model. The concept captured the idea of a community within an organization that collectively acts as a teacher to a new apprentice. Since community based learning is not limited to novices, and applies to numerous contexts and situations, the concept was quickly applied to more general learning and knowledge exchange situations. Brown and Duguid (2001), who worked mainly on the theoretical foundation of communities of practice, linked the concept to collective and social learning theories. They observed that learning and knowledge exchange is social and comes largely from experience with, and the relationship to, others. Communities of practice are institutions whose participants are informally bound by common activities and by what they have learned through their mutual involvement in these endeavors (Wenger, 1998). They can be seen as ‘informal, spontaneous, self-organized groups of individuals who share knowledge, solve common problems and exchange insights and frustrations based on the similar work roles and a common context’ (Lesser and Prusak, 2000). Since communities are not limited to organizations, they can be extremely important mechanisms for facilitating the flow of knowledge across organizations.
Scientific and technological communities can be seen as examples of communities of practice. Researchers have long studied how scientists interact with each other and under which socially constructed rules and norms they work. For example, Crane (1972) investigated the importance of individuals in creating knowledge bridges across organizations and the ‘invisible college’ of scientists that helps to diffuse knowledge within scientific communities. Her work is similar to related studies by Tushman (1977) and Allen and Cohen (1969) who argue about the positive effects of boundary spanning activities by certain individuals who are well-connected internally and externally. Boundary spanning scientists can use their social ties to develop links to experts in other firms, universities, and research institutions, and thereby act as informal bridges across organizational boundaries.
More recent research builds on this theme and tends to examine the effect of knowledge exchange by scientists on innovation. Looking directly at the benefits of firm–university collaboration, Cassiman et al. (2008) analyze the effect of science linkages to innovation performance at the patent level. The authors demonstrate that citations in scientific publications are not the main driver to explain forward citations, but they are positively related to their generality and geographical dispersion. Moreover, they illustrate that science linkages, at the firm level, matter more for forward citations with the exception of emerging technologies. Particularly, non-science related patents which have no scientific linkages are less frequently and less easily cited than comparable patents of firms with science linkages. When looking at the impact of high level scientific output on patents, Gittelman and Kogut (2003) find that publications, collaborations, and science intensity are associated to patented innovations; however, important scientific papers are negatively associated with high-impact innovations. The authors conclude that scientific and marketable innovations follow a different underlying logic and that the direct move from science to patent is more difficult and complex than previously assumed (Gittelman and Kogut, 2003). George et al. (2002) examine the effect of science linkages on patent variables and show that firms (in the Biotech sector) with university linkages have lower research and development expenses though they have higher levels of innovative output. However, they do not find support for the proposition that companies with university linkages show greater financial performance than similar firms without such linkages (George et al., 2002). An explanation for the, at best, mixed findings regarding the performance implication of firm–university collaborations is provided by Murray (2002). She argues that the underlying social structures are very different between ‘science’ and ‘technology,’ but that they co-evolve. Furthermore, she shows empirically that neither co-publishing nor citations as predicted from current literature drives performance, it is rather the co-mingling through founding, licensing, and consulting. Therefore, both theoretical and empirical studies suggest that knowledge exchange between scientists and engineers can have positive effects on firm innovation.
Board interlocks and knowledge
Board interlocks are another example of inter-organizational relationships that permit individual-level and informal knowledge exchange that potentially impacts corporate behavior and performance. The literature on board interlocks suggests that senior executives who sit on the boards of multiple firms form a unique social community and that the interaction within this community influences firm behavior. Even though board appointments are a formal and institutionalized function, they provide board members with a context in which to develop informal relationships resulting in knowledge exchange that extends beyond task specific information.
Since the 1980s research on board interlocks has produced numerous insights about their significance and usefulness and this stream of research has grown more prominent during the 1990s. Initial research conceptualized interlocks as a means of simultaneous cooperation and competition and as vehicles to reduce risk (Johnson et al.,1996; Murray, 2002). In further research, the learning and knowledge transfer aspects of board interlocks have also been highlighted. For instance, Useem’s (1984) research explicitly discussed the learning effects of board interlocks as opposed to the control advantages these inter-firm linkages provided. He argued that interlocks provide managers with an additional source of knowledge regarding the latest business practices and developments in the broader business environment.
The social networks created by common board memberships have some specific characteristics that differentiate them from other informal exchange mechanisms. Board appointments enable executives to acquire and discuss knowledge related to new business and environmental trends at the highest organizational level (Haunschild and Beckman, 1998). These peer based interactions are otherwise less common for top-level executives. This high level interaction can be useful for benchmarking, identifying the adoption of new or different practices, and helping to clarify and evaluate these new practices (Davis, 1991; Gulati and Westphal, 1999). It is worth noting that the direct interaction between board members ensures the transfer of tacit components of knowledge. Hence, board membership provides executives with a context for social interaction that helps define the ‘unwritten rules and traits of the game’ (Westphal et al., 1997).
Not only do interlocks expose managers to new knowledge, but the direct contact between high level executives within an institutional setting is likely to make this knowledge appear more trustworthy. For example, in many countries (including the US), interlock partners are prohibited from being direct competitors and this reduces potential conflicts of interest (Zajac, 1988). A high level of trust created in part by the institutional setting probably motivates executives to share reliable knowledge (Haunschild and Beckman, 1998). Empirical research on board membership has documented several cases in which the exchange of knowledge—understood as knowledge of certain business practices—has directly influenced firm behavior. For example, Davis (1991) shows that firms are more likely to adopt a ‘poison pill’ as a takeover defense mechanism when their managers had board interlocks with firms that had already adopted poison pills. Haunschild (1993) investigated the type of knowledge transmitted by acquirers to other potential acquirers within board settings. She found that private information was not shared—rather it was broader knowledge related to ‘how to’ do things and other normative information. In another study of acquisitions and board interlocks, Haunschild and Beckman (1998) analyzed the interaction of alternative and often informal knowledge channels including business roundtables, business press, consultants, and private contacts with board interlocks on acquisition related knowledge. The paper finds that certain alternative knowledge channels (e.g. business roundtables) reduce the impact of interlocks while others (e.g. business press coverage) reinforce it. The authors argue that business roundtables provide rather similar knowledge as board interlocks whereas business press coverage of acquisitions is a complementary knowledge source. In a related study, Gulati and Westphal (1999) investigate the effect of interlocks on alliance formation. A CEO–board relationship characterized by independent board control reduces the likelihood of alliance formation by prompting distrust between corporate leaders, while CEO–board cooperation in strategic decision making appears to promote alliance formation by enhancing trust. In a recent study on alliance formation, Rosenkopf and Schleicher (2008) illustrate that interlocks and the participation in cooperative technical organizations facilitate alliance formation. However, interlocks only influence alliance formation positively when the common director serves as an officer in one of the firms. Hence, board interlocks can be seen to provide useful knowledge related to strategic decision making and organizational and managerial innovation.