Outcomes and Antecedents

A great number of studies have examined the antecedents and especially outcomes of absorptive capacity. A wide variety of insights have been gained in the past two decades, but such insights have also been obfuscated by the different ways in which absorptive capacity has been operationalized and by the different levels of analysis at which absorptive capacity’s antecedents and outcomes have been examined.

Outcomes

Research on the various outcomes of absorptive capacity is extensive (see Table 13.3). One of the main reasons for Cohen and Levinthal (1990) to introduce absorptive capacity was to explain the side effects of research and development and their relation to innovation. In line with their argument, other research has adopted research and development-based measures as proxies for absorptive capacity to understand its effect on innovation. The evidence gained so far is mixed. Current insights lean towards a positive effect (e.g. Tsai, 2001), but studies have surfaced reporting negative (e.g. Ernst, 1998) and zero (e.g. Singh, 2008) effects of these measures on innovation. Also more complex relationships have been found. For example, Stock et al. (2001) found that absorptive capacity contributes to new product development up to a certain point but then its effect starts to decrease and becomes negative. Since research and development-based measures are antecedent to absorptive capacity and capture more than absorptive capacity alone, it remains to be understood whether such curvilinear effect is due to research and development investments as such or to absorptive capacity.

Table 13.3 Outcomes of absorptive capacity

The evidence of studies relying on experience-based measures as a gauge for absorptive capacity is also mixed at best with some studies finding positive relationships (e.g. Kusunoki et al., 1998; Macher and Boerner, 2006) and others finding negative and insignificant relationships (e.g. Haas, 2006). In contrast, with few exceptions (e.g. Singh, 2008), research relying on patent stocks to measure absorptive capacity tends to be evident of a positive relationship with innovation (e.g. Cattani, 2005; Yayavaram and Ahuja, 2008). Similarly, studies using scales (e.g. Lichtenthaler, 2009) and personnel measures (e.g. White and Liu, 1998) as proxies for absorptive capacity generally found positive effects on innovation.

The variety in earlier findings of the effect of absorptive capacity on innovation is due in part to the different ways in which studies have applied measures of absorptive capacity. Different measures allow researchers to emphasize distinct aspects of a firm’s knowledge base. As mentioned, patent-based measures have been used not only to measure the volume and richness of a firm’s knowledge base by counting patents but also to gauge its diversity by differentiating between patent classes that characterize a firm’s patent stock (Almeida and Phene, 2004; Zhang et al., 2007b). This distinction is especially salient because richness and depth of knowledge enable a firm to master technological advances, while diversity and breadth of knowledge render a firm’s capability to combine knowledge in new and novel ways (Katila and Ahuja, 2002; Subramaniam and Youndt, 2005). Absorptive capacity is a construct to explain why and how firms differ in their ability to acquire knowledge across organizational boundaries. Because it rests on the presence of a relevant knowledge base and its development is path dependent, firms tend to accumulate deep knowledge within a technological domain as this enables firms to exploit technological knowledge (Cohen and Levinthal, 1990, 1994). A firm seeking to innovate beyond its current technological trajectory needs to span technological boundaries in addition to spanning organizational boundaries, and use knowledge and technologies from domains beyond its current product offerings (Rosenkopf and Nerkar, 2001).

Both mastery of a single technology and the combination of different technologies may lead to innovation, albeit to different kinds of innovation. As innovation sheds light on the mixed evidence of the effect of absorptive capacity on innovation, studies have begun to make a distinction between its different dimensions and manifestations. A common distinction that is reflective of the difference between mastering and combining technologies is between exploitative and exploratory innovations. Exploitative innovations build on current knowledge and incrementally refine and extend existing products and competences, whereas exploratory innovations depart from existing knowledge and aim to develop new alternatives (March, 1991). Consistent with Cohen and Levinthal’s (1990) argument, Bierly et al. (2009) found that technological relatedness, a measure of relative absorptive capacity, inhibits exploration. The more the technologies of partnering firms relate the more efficient knowledge acquisition will become and fewer opportunities exist for making new linkages. Similarly, Bergh and Lim (2008) differentiate absorptive capacity from improvisation and creativity. However, Van den Bosch et al. (1999) contend that knowledge absorption may vary in its efficiency, scope, and flexibility, and have found that efficiency facilitates exploitation whereas scope and flexibility contribute to exploration. Lane et al. (2006) take another perspective and argue that absorptive capacity caters for both exploration and exploitation. As part of their definition of absorptive capacity, they submit that the recognition and understanding of potentially valuable knowledge constitute the processes through which firms explore new knowledge, while using it is part of the exploitative learning process.

Seeking to further understand the effect of absorptive capacity on a firm’s competitive position, studies have also examined its effect on performance. A variety of studies at the inter-firm (Lane et al., 2001), firm (Bogner and Bansal, 2007; Dushnitsky and Lenox, 2005; Rothaermel and Hill, 2005), and unit (Cohen and Levinthal, 1990; Tsai, 2001) level found a positive effect of absorptive capacity on performance. The effects found are, however, again dependent on the operationalization chosen by the investigators. A strong case in point is evident in the study of Steensma and Corley (2000) who used research and development intensity, size, and technological relatedness as proxies for absorptive capacity and only found relatedness to influence a firm’s return on investment. In his study on 300 medium-sized and large German firms, Lichtenthaler (2009) differentiates between the three learning processes of absorptive capacity suggested by Lane et al. (2006) and found that especially the exploitative learning process influences firm performance. Likewise, Lane et al. (2001) found strong evidence that the application of external knowledge leads to performance increments.

In addition to providing evidence for the performance implications of absorptive capacity, the results of Lane et al. (2001) suggest that knowledge transfer mediates the relationship between absorptive capacity and performance. Differentiating between the three capabilities characterizing absorptive capacity, they found that the ability to understand and assimilate knowledge contributes strongly to the knowledge learned from foreign IJV parents. Using a variety of operationalizations of absorptive capacity, ranging from research and development-based measures to patents to psychometric scales, a growing number of studies indicate that absorptive capacity facilitates inter-organizational learning (Lyles and Salk, 1996; Mowery et al., 1996; Rosenkopf and Almeida, 2003). Similarly, the insights gained so far point out that absorptive capacity is a critical determinant of knowledge transfer across a firm’s business units (e.g. Minbaeva et al., 2003; Szulanski, 1996). In their meta-analytic study, Van Wijk et al. (2008) corroborate the direct effect of absorptive capacity on knowledge transfer both across units and across firms.

Remarkably, Gupta and Govindarajan (2000) found no significant effect of absorptive capacity on knowledge transfers across peer subsidiaries. Their study on MNC knowledge flows shows partial support for the role of absorptive capacity in that it is limited to knowledge that originates in the parent organization and is transferred to subsidiaries. However, this finding is likely the result of their operationalization of absorptive capacity as mode of entry. Although their argument that acquired subsidiaries are less likely to have overlapping knowledge with other subsidiaries than greenfield subsidiaries makes sense, entry mode encompasses much more than the ability to absorb knowledge. Parent organizations are more likely to provide information and knowledge when they set up a greenfield subsidiary and will continue to do so because such subsidiaries will be perceived as more relevant and requiring more attention. In their study on manufacturing facilities in the information and communications industry, Lenox and King (2004) found that the provision of information by corporate managers enhanced both the adoption of pollution prevention practices by organizational subunits as well as the effect of subunits’ absorptive capacity on such adoption.

Finally, studies have examined the effect of absorptive capacity on expectation formation (Cohen and Levinthal, 1990, 1994) and the formation of alliances (Rothaermel and Hill, 2005; Veugelers and Kesteloot, 1996; Zhang et al., 2007b). Absorptive capacity enables firms not only to acquire knowledge but, since it involves knowledge of why certain technological trajectories are valuable and of who knows what, also to assess what may be potential future technological advances and the value of new partners it can collaborate with. Moreover, effective absorptive capacity rests on both the ability to communicate internally and the ability to acquire knowledge externally (Rothaermel and Alexandre, 2009), as well as on both internal and external research and development activities (Cassiman and Veugelers, 2006; Nicholls-Nixon and Woo, 2003). External knowledge may be supplied by potential alliance partners; hence, firms with high absorptive capacity are more likely to enter into an alliance because their expectations of its benefits are more certain.

Antecedents

In contrast to the outcomes of absorptive capacity, the antecedents and determinants of absorptive capacity have received limited attention. Most insights on the antecedents of absorptive capacity have been gained from research centering on the variables that have been used often to operationalize absorptive capacity. An overview of current research illustrates that characteristics of the knowledge involved are important to consider as they influence the learning process. Since firm or unit absorptive capacity is not simply the sum of the absorptive capacities of its constituent individual members, however, Cohen and Levinthal (1990: 131) make a case for considering the aspects of absorptive capacity that are ‘distinctly organizational.’ Moreover, since absorptive capacity is dependent on the knowledge bases of all actors involved, also dyad and network characteristics have been identified. An overview of the antecedents of absorptive capacity is provided in Table 13.4.

Table 13.4 Antecedents of absorptive capacity

Knowledge characteristics

The basic tenet of Cohen and Levinthal’s (1990) argument is that absorptive capacity is dependent on the level of prior related knowledge. Studies abound underscoring this notion at the unit and firm level (Bogner and Bansal, 2007; Helfat, 1997; Lane et al., 2001; Shane, 2000; Van den Bosch et al., 1999). Firms and units endowed with knowledge and expertise within a domain will more easily learn new knowledge in that domain and closely related domains. In that sense, ‘accumulating absorptive capacity in one period will permit its more efficient accumulation in the next’ (Cohen and Levinthal, 1990: 136). The efficiency and performance benefits derived from the cumulative quality of absorptive capacity lead firms and units to build on existing knowledge in domains relevant to current operations. The caveat of being driven by prior related knowledge is that firms and units may come to search locally and neglect domains that are less relevant and familiar (Stuart and Podolny, 1996; Van den Bosch et al., 1999).

The notion that absorptive capacity is dependent on a relevant knowledge base highlights the importance of considering the richness and diversity of knowledge (cf. Almeida and Phene, 2004; Zahra and George, 2002; Zhang et al., 2007b). Knowledge richness represents the depth and the extent of knowledge in domains. Knowledge diversity reflects the breadth and number of domains in which a firm or unit has knowledge. As Cohen and Levinthal argue (1990: 131; italics added), ‘learning performance will be greatest when the object of learning is related to what is already known . . . [while] a diverse background . . . increases the prospect that incoming information will relate to what is already known.’ Thus, both knowledge richness and knowledge diversity interact to determine in which domains a firm or unit can potentially learn and how well it can do so. Indeed, studies have empirically shown that technological relatedness increases a firm’s ability to absorb new knowledge (e.g. Bierly et al., 2009; Mowery et al., 1996). Since efficiency and strategic considerations marshal firms and units to emphasize the most relevant related domains, relatedness deepens and enriches a firm’s knowledge base and reduces its diversity.

Since units are main entry points for knowledge, firms wishing to absorb knowledge and develop absorptive capacity require interfaces between units and external constituents as well as across units. An important side effect of the path dependent nature of absorptive capacity is that units may come to experience difficulty to share knowledge with peer units as they diverge in developing their own knowledge bases. Cohen and Levinthal (1990) make, therefore, a distinction between outward-looking and inward-looking absorptive capacity. Outward-looking absorptive capacity facilitates externally-driven research and development and the potential to absorb knowledge, whereas inward-looking absorptive capacity fosters internal research and development and the realization of absorptive capacity into exploited knowledge (cf. Nagarajan and Mitchell, 1998; Nicholls-Nixon and Woo, 2003; Rothaermel and Alexandre, 2009; Zahra and George, 2002). Firms need to maintain a delicate balance between both as ‘excessive dominance by one or the other will be dysfunctional’ (Cohen and Levinthal, 1990: 133).

Cohen and Levinthal (1990: 134) argue that the ‘ideal knowledge structure for an organizational subunit should reflect only partially overlapping knowledge complemented by nonoverlapping diverse knowledge.’ Knowledge may be differentiated into different types. Prior related knowledge is not only characterized by substantive technological and product knowledge, but also by basic skills, learning skills, problem solving methods, prior learning experiences, and a shared language. In that sense, that two units in a diversified firm pursuing two seemingly unrelated technologies may be able to acquire knowledge from each other if they have a shared language and operate using similar problem-solving methods. Likewise, units may have unrelated technological knowledge but may have similar market and managerial knowledge and capabilities that provide an anchor for learning new, non-overlapping knowledge (cf. Sammara and Biggiero, 2008). Henderson and Cockburn (1996) found a positive effect of the knowledge held by a firm in one research domain on the knowledge developed in technologically related domains as well as on the number of research domains pursued by firms.

Intensity of effort is required if a firm or unit seeks to acquire non-overlapping knowledge and to develop absorptive capacity beyond domains it is familiar with and in which it has prior knowledge (Cohen and Levinthal, 1990; Kim, 1998). When knowledge is complex and comprised of multiple interconnected components, firms and units will experience difficulty acquiring and using such knowledge (Levitt and March, 1988). However, if some of these knowledge components are shared or can be understood through analogical reasoning they may be able to gradually understand knowledge that is completely new to them (Gavetti et al., 2005). Learning beyond familiar domains with the aim to create new knowledge is most effective at the individual level. Matusik and Heeley (2005) found that individual absorptive capacities are more likely to lead to the creation of new knowledge, whereas collective knowledge is more prone to the extension of existing knowledge. Similarly, Argote and Ingram (2000) argue that transferring knowledge from one site to another will be more effective if accompanied by moving people. People are more capable of adapting knowledge to the new context as their knowledge of how to use the transferred knowledge is interconnected. Consequently, characteristics of the knowledge to be transferred illustrate that firm absorptive capacity is dependent on the interplay between individuals and units as well as the links between them (see also Figure 13.1).

Organizational characteristics

Since the distribution of knowledge cannot be untied from the organization, Cohen and Levinthal (1990) argue that absorptive capacity is dependent on a set of organizational characteristics. Van den Bosch et al. (1999) found that organizational form influences absorptive capacity in that organizations with mechanistic properties are more likely to be efficient in absorbing knowledge, while organic organizations enjoy benefits in the scope and flexibility of knowledge absorption. In mechanistic organizations, organizational processes are formalized, which allows firms to exploit knowledge and realize absorptive capacity (Jansen et al., 2005), and research and development is more likely to be centralized, which leads to a broadening of knowledge and capabilities (Argyres and Silverman, 2004) and under stable environmental conditions enables firms to build on existing knowledge efficiently (Cohen and Levinthal, 1990; Van den Bosch et al., 1999). In organic organizations, local managers are closely located to action and are more likely to participate in decision making, which allows them to adapt to new circumstances and develop absorptive capacity (Gavetti, 2005; Jansen et al., 2005). In other words, firms that seek to develop absorptive capacity both to build on existing knowledge and to create new knowledge require both mechanistic and organic characteristics.

Consistent with Van den Bosch et al. (1999), Yayavaram and Ahuja (2008) found that a nearly decomposable knowledge base, which is characterized by clusters of knowledge which are connected and integrated by ties, is most advantageous when it comes to useful inventions, and they discuss how such knowledge bases can be created. Since knowledge bases should be viewed as networks of knowledge elements in which the ties are as important as the elements themselves, they argue that organization structures should permit differentiation across clusters and include integration mechanisms, such as personnel transfers across units, gatekeepers, and rewards for cross-unit innovations. Since ‘firms are constrained to local search due to cognitive limitations and the lack of absorptive capacity required for long “jumps”, . . . integration between clusters ensures that local moves at the level of a cluster can lead to adaptive walks that span cluster boundaries’ (2008: 357).

Indeed, the evidence that mechanisms facilitating integration between individuals and units contribute to absorptive capacity is strongly growing. Creating a mosaic of links between individuals and units enables them (1) to realize that knowledge they assimilated is transformed and exploited (Rothaermel and Alexandre, 2009; Zahra and George, 2002), (2) to enlist others to absorb knowledge in case the knowledge is unrelated to their existing knowledge base, and (3) to relieve others from evaluating and assimilating knowledge (Cohen and Levinthal, 1990). An important way in which firms achieve integration is through their combinative capabilities (Kogut and Zander, 1992). To understand the impact of combinative capabilities on absorptive capacity, Jansen et al. (2005) and Van den Bosch et al. (1999) differentiate between system, coordination, and socialization capabilities. Based on two case studies, Van den Bosch et al. (1999) found that systems capabilities drive the efficiency at which firms absorb knowledge, whereas coordination and socialization capabilities foster the scope and flexibility in which firms absorb knowledge. Jansen et al. (2005) found that different types of combinative capabilities influence the different capabilities and dimensions constituting absorptive capacity. Specifically, they found that coordination capabilities, as manifested by cross-functional interfaces, participation in decision making, and job rotation, mostly influence the acquisition and assimilation of knowledge and a small portion of transformation of knowledge. Socialization capabilities, such as connectedness and socialization mechanisms, were found to be better predictors of the transformation and exploitation of knowledge. The effect of systems capabilities was more complex in that formalization contributed strongly to the exploitation of knowledge, but routinization had a negative impact on acquisition, assimilation, and transformation.

The study of Jansen et al. (2005) is one of the few exceptions having sought to understand the antecedents of the different capabilities or dimensions that make up absorptive capacity. Most studies empirically differentiating the capabilities constituting absorptive capacity have made a distinction between the transfer of knowledge from an outside source to the firm and the application of that knowledge, which are broadly understood as potential and realized absorptive capacity respectively (Bierly et al., 2009; Minbaeva et al., 2003; Mowery et al., 1996). Others have made a distinction between the three capabilities constituting absorptive capacity initially forwarded by Cohen and Levinthal (1990), and underscored that each capability requires discrete organizational processes and that using broad dimensions may be suboptimal (Lane et al., 2001). An important insight gained from these studies is that training fosters the ability to assimilate and acquire knowledge (Lyles and Salk, 1996; Minbaeva et al., 2003), and that competence in training facilitated the commercial application of that knowledge (Lane et al., 2001). Training extends and alters organizational members’ knowledge bases, which enables them to assimilate knowledge they were previously not familiar with. Competence in training increases the efficacy in which firms and units are able to have organizational members understand, retain, and apply knowledge that is complex and causally ambiguous. The ability to acquire and assimilate knowledge is further enhanced by the degree of flexibility in adapting to changing circumstances and contexts (Lane et al., 2001). A firm’s ability to transform and apply knowledge is dependent on internal communication between units and individuals (Cohen and Levinthal, 1990; Minbaeva et al., 2003), as well as on the strategy pursued by a firm (Lane et al., 2001) and the presence of performance-based compensation schemes and incentives (Minbaeva et al., 2003). Bierly et al. (2009) focused particularly on the antecedents of applying knowledge and made a distinction between exploitation and exploration. They found that financial leverage and the presence of technological capabilities expedites the application of knowledge to exploitative innovations, whereas entrepreneurial orientation fosters the application of knowledge to exploration.

Network characteristics

Lane and Lubatkin (1998) argue that absorptive capacity is not absolute but essentially relative, and should be considered at the dyad level since the capacity for learning is dependent on characteristics of the interacting parties involved. They found that absolute absorptive capacity as measured by research and development intensity has limited explanatory power as a predictor of knowledge absorption. Correspondingly, Dyer and Singh (1998) argued that absorptive capacity is inherently partner-specific. A firm or unit may have a substantial knowledge base, but its ability to learn from its partner depends on whether part of that knowledge base overlaps with the knowledge base of its partner (Lane et al., 2001; Mowery et al., 1996). In their study of research and development alliances between pharmaceutical and biotechnology firms, Lane and Lubatkin (1998) found that knowledge overlap determines a firm’s ability to learn but that the effect of knowledge overlap is limited to overlap in basic knowledge of biochemistry since overlap in specialized knowledge had no effect. Overlap in basic knowledge facilitates the valuation of external knowledge extending or related to biochemistry. A specialized knowledge base is more likely to be rich and deep and overlap limits the opportunity set of firms to learn new knowledge. Corroborating this notion, Kotabe et al. (2007) found that the relative quality of knowledge has a positive impact on knowledge transfer, whereas a firm’s absolute quality of knowledge diminishes knowledge transfer. Instead, the presence of a qualitatively extensive knowledge base leads to more innovation. Moreover, their evidence shows that prior experience with knowledge transfer contributes strongly to both knowledge transfer and innovation. Prior experience with knowledge transfer may serve as a template that shapes firms’ ability to obtain and apply knowledge to innovation, even when such knowledge is unrelated to existing knowledge.

Along with prior knowledge, firms and units need in place similar knowledge processing systems to facilitate knowledge absorption, especially when such knowledge is new and dissimilar to existing knowledge. In keeping with Cohen and Levinthal’s (1990) original argument that absorptive capacity has elements that are typically organizational, Lane et al. (2001) and Lane and Lubatkin (1998) also studied organizational characteristics at the dyad level. Lane et al. (2001) found a positive effect of trust and cultural compatibility, in addition to related knowledge bases, on the ability to understand knowledge. However, these effects disappeared when other antecedents were included in the model. Lane and Lubatkin (1998) found that similarities in lower management formalization, in centralization of research and development, and in compensation practices facilitate the assimilation of knowledge. The degree of upper management formalization and management centralization were, on the other hand, found to have a negative effect. These findings are in line with other studies that found that since local managers are more closely located to the action they should be able to participate in decision-making processes so that they can monitor the environment and broaden a firm’s knowledge base (Gavetti, 2005; Jansen et al., 2005), while centralizing research and development enriches and deepens knowledge (Argyres and Silverman, 2004; Cohen and Levinthal, 1990). When firms or units are similar with regard to these organizational aspects, communication and understanding between them is facilitated. Altogether, current insights obtained from dyad level studies seem to substantiate that similarities in knowledge mainly influence the ability to evaluate knowledge, whereas organizational characteristics facilitate the assimilation and commercialization of knowledge.

In addition to examining the similarities and differences between pairs of firms, studies have assessed the effect of network-level characteristics (see Chapter 22 by Van Wijk et al. for an overview). A common approach in such studies is to distinguish between the structural, relational, and cognitive elements of the networks in which firms and units operate and from which they derive social capital (Adler and Kwon, 2002; Nahapiet and Ghoshal, 1998). Firms and units with a structurally central position in the network possess information benefits in that they have ties and access to redundant and non-redundant knowledge and are able to bridge these ties. For example, Cockburn and Henderson (1998) found that absorptive capacity is driven not only by internal basic research but also by the connectedness of researchers to a wide external community of other researchers. While redundant knowledge overlaps with existing knowledge, a non-redundant tie increases the range, novelty, and diversity of knowledge to which a firm or unit has access. By balancing redundant and non-redundant knowledge, firms and units receive valuable knowledge and know how to apply such knowledge to rewarding opportunities (McEvily and Zaheer, 1999).

While the ties reaped by a strong structural position in the network facilitate the transfer of non-complex, explicit knowledge, they are too weak for the transfer of complex, tacit knowledge (Hansen, 1999; Li et al., 2010). Firms and units also need strong relational and bonding ties characterized by trust. Trust facilitates knowledge absorption because it shapes the confidence of the recipient that the knowledge of the source is reliable and valuable, and increases its willingness to extend its efforts in absorbing the knowledge. Consequently, strong relational ties enable joint problem solving and the gradual learning of new and often tacitly held knowledge (McEvily and Marcus, 2005).

The ease with which new, unrelated knowledge can be understood, acquired, and applied is ameliorated by cognitive social capital, which manifests itself in the presence of a shared language, shared goals, and shared systems (Nahapiet and Ghoshal, 1998). While shared systems and a shared language facilitate understanding and acquiring new knowledge (cf. Cohen and Levinthal, 1990), shared goals allow firms and units to understand how knowledge consisting of multiple interconnected components can be applied. Marking the presence of shared values, cultural compatibility also has a positive effect on firms’ abilities to absorb knowledge, although Lane et al. (2001) found that its effect is smaller than the effect of related knowledge bases. In the same vein, geographic proximity has been found to positively relate to the capacity to absorb knowledge (Frost, 2001; Gupta and Govindarajan, 2000). Geographic proximity enhances the strength of ties as opportunities to interact with partners are multifold. In that sense, Feinberg and Gupta (2004) found that the decision of MNCs to locate research and development in local subsidiaries is driven by potential knowledge spillovers in local markets.

Finally, prior research has examined the role of the type and governance structure of the collaboration. Li et al. (2010) found that formal contracts not only assist firms in the acquisition of explicit knowledge, but also govern the effect of shared goals and trust on knowledge acquisition. Arora and Gambardella (1990) suggest that firms seeking to access basic knowledge and to keep track of technological developments in domains in which they lack internal knowledge enjoy efficiency benefits if they enter into non-equity alliances with universities and minority equity investments in other firms. Non-equity agreements with other firms are effectively used to access products that are ready for commercialization.