OrganizationScience Vol. 22, No. 1, January���February 2011, pp. 81���98 issn 1047-7039 eissn 1526-5455 11 2201 0081 informs �� doi 10.1287/orsc.1100.0525 ��2011 INFORMS Microfoundations of Internal and External Absorptive Capacity Routines Arie Y. Lewin The Fuqua School of Business, Duke University, Durham, North Carolina 27708, ayl3@duke.edu Silvia Massini Manchester Institute of Innovation Research, Manchester Business School, University of Manchester, Manchester M15 6PB, United Kingdom, silvia.massini@mbs.ac.uk Carine Peeters European Center for Advanced Research in Economics and Statistics and Centre Emile Bernheim, Solvay Brussels School of Economics and Management, Universit�� libre de Bruxelles, B-1050 Brussels, Belgium, carine.peeters@ulb.ac.be T(Cohen, he 20 years following the introduction of the seminal construct of absorptive capacity (AC) by Cohen and Levinthal W. M., D. A. Levinthal. 1989. Innovation and learning: The two faces of R&D. Econom. J. 99(397) 569���596 Cohen, W. M., D. A. Levinthal. 1990. Absorptive capacity: A new perspective on learning and innovation. Admin. Sci. Quart. 35(1) 128���152) have seen the proliferation of a vast literature citing the AC construct in over 10,000 published papers, chapters, and books, and interpreting it or applying it in many areas of organization science research, including organization theory, strategic management, and economics. However, with very few exceptions, the specific organizational routines and processes that constitute AC capabilities remain a black box. In this paper, we propose a routine-based model of AC as a first step toward the operationalization of the AC construct. Our intent is to direct attention to the importance of balancing internal knowledge creating processes with the identification, acquisition, and assimilation of new knowledge originating in the external environment. We decompose the construct of AC into two components, internal and external AC capabilities, and identify the configuration of metaroutines underlying these two components. These higher-level routines are expressed within organizations by configurations of empirically observable practiced routines that are idiosyncratic and firm specific. Therefore, we conceptualize metaroutines as the foundations of practiced routines. The ability of organizations to discover and implement complementarities between AC routines may explain why some firms are successful early adopters and most firms are imitators. Success as an early adopter of a new management practice or an innovation is expected to depend on the extent to which an organization evolves, adapts, and implements the configuration of its internal and external absorptive capacity routines. Key words: absorptive capacity routines capabilities microfoundations complementarities innovators imitators History: Published online in Articles in Advance May 7, 2010. Introduction The 20 years following the introduction of the semi- nal construct of absorptive capacity (AC) by Cohen and Levinthal (1989, 1990) have seen the proliferation of a vast literature citing the AC construct in over 10,000 published papers, chapters, and books, and interpreting it or applying it in many areas of organization science research, including organization theory, strategic man- agement, and economics (see, for example, Mowery and Oxley 1995, Lane and Lubatkin 1998, Van den Bosch et al. 1999, Lane et al. 2006). AC is the ���ability of a firm to recognize the value of new, external infor- mation, assimilate it, and apply it to commercial ends��� (Cohen and Levinthal 1990, p. 128) and is attributed to moderating or mediating a range of phenomena relating to firm-level innovation, adaptation, and performance. Over time, AC has evolved from being a moderating factor of firms��� ability to develop and adopt technolog- ical innovation to being an enabler of firms��� adaptation and change in general. But, although a widely used con- struct, with very few exceptions (e.g., Szulanski 1996) the specific organizational routines and processes that constitute AC capabilities remain a black box (e.g., Lane et al. 2001, Zahra and George 2002, Lewin and Massini 2003, Todorova and Durisin 2007, see in particular Lane et al. 2006). This is perhaps because AC has acquired the charac- teristics of an umbrella concept (Hirsh and Levin 1999, Meyer 1991). Its development and evolution over time are consistent with the three-stage model described by Hirsh and Levin (1999). In the first stage, the ���emergent excitement��� follows the initial articulation by Cohen and Levinthal (1989, 1990), which attributed innovative per- formance to the firm���s absorptive capacity. The second 81

Lewin, Massini, and Peeters: Microfoundations of Internal and External Absorptive Capacity Routines 82 Organization Science 22(1), pp. 81���98, ��2011 INFORMS stage, the ���validity challenge,��� attracts research intended to operationalize or create direct measures of AC or in other ways empirically validate the construct. How- ever, over time the absence of direct measures leads to the third stage, ���tidying up with typologies,��� which is reflected in recent publications by Zhara and George (2002), Lane et al. (2006), and Torodova and Durisin (2007). In the absence of progress on operationalizing an umbrella concept, Hirsh and Levin (1999) expect three possible outcomes, ���override challenges��� (alterna- tive constructs emerge), ���permanent issue��� (unresolved problems), and ���construct collapse��� (usefulness of con- structs diminishes). The model of AC proposed by Zahra and George (2002) has proved an important contribution in describ- ing AC as bundles of routines and capabilities. It focuses on adaptation of companies to new knowledge originat- ing in the external environment. However, it is silent on the processes and capabilities underlying internal knowledge combination, recombination, transformation, exploitation, and assimilation. In this paper, we advance a more proactive view of AC as a bundle of routines that encompass the ability of companies to initiate change from within as well as identify and assimilate ideas from the external environment. In their original con- tribution, Cohen and Levinthal (1989) already empha- sized the twofold purpose of AC (in their model research and development (R&D) investments) of generating new knowledge internally and enabling absorption of exter- nally generated knowledge. The present paper returns to the original dual conceptualization of AC and advances a model of AC incorporating internal and external metaroutines that make up an organization AC capabili- ties (i.e., their microfoundations), as well as the expres- sion of these internal and external AC metaroutines in the form of practiced routines. In doing so, we are also addressing the validity challenge of AC becoming an umbrella concept by proposing a routine-based model of AC that in our view will advance future empirical research on operationalizing the AC construct. The concept of routines has been applied in a wide range of settings, and it has a central place in A Behav- ioral Theory of the Firm (Cyert and March 1963, Argote and Greve 2007) and in evolutionary economics theory (Nelson and Winter 1982). In this literature stream, rou- tines are seen as the building blocks of organizational capabilities (Dosi et al. 2000, Winter 2003), and their systematic generation and modification in response to past experience and environmental changes is at the core of firm dynamic capabilities (Teece et al. 1997, Zollo and Winter 2002). Moreover, if such capabilities are firm specific, idiosyncratic, and/or complex and unobserv- able, they are the source of competitive advantage and must be difficult to imitate (Nelson and Winter 1982 Barney 1991 Rifkin 2000, 2001). Prime and Butler (2001) argue that it is difficult if not impossible to test a theory that accounts for heterogeneous performance out- comes on the basis of unobserved capabilities. Barney (2001) counters this criticism by suggesting that direct measurement of capabilities is not necessary if the the- ory can specify the origins and consequences of capabili- ties. In this paper, we propose to overcome this challenge by advancing a model of metaroutines (Cyert and March 1963, Nelson and Winter 1982, Feldman and Pentlands 2003) underlying AC capabilities, i.e., higher-order rou- tines that define the purpose of AC capabilities, and dis- cussing their actual expression in the form of observable and measurable practiced routines. In the sections that follow, we briefly review the literature on AC and the lit- erature on organizational routines. Then, we distinguish between internal and external AC routines, propose a taxonomy of AC metaroutines, and illustrate the opera- tionalization of the metaroutines with examples of prac- ticed AC routines. This is followed by a more general discussion of the role of AC routines and capabilities for mediating innovation performance, and a summary and conclusions. Revisiting Absorptive Capacity and OrganizationalRoutines Absorptive Capacity The concept of AC was first advanced by Cohen and Levinthal (1989, 1990) as the ���ability of a firm to recog- nize the value of new, external information, assimilate it, and apply it to commercial ends��� (1990, p. 128). They further argued that AC mediates the speed, frequency, and magnitude of innovation, and that the evolution of firms��� AC capabilities has a strong path dependency on the prior R&D investment and knowledge base of a firm. AC and learning are often described as coevolv- ing and mutually reinforcing (Barkema and Vermeulen 1998, Simonin 1999, Autio et al. 2000). AC enables firms to innovate and learn, and the new knowledge adds to the existing AC (Kim and Kogut 1996, Helfat 1997, Van den Bosch et al. 1999). Van den Bosch et al. (1999) further argue that the AC���learning���new AC feedback loop suggested by Cohen and Levinthal (1990) is medi- ated by the environment in which the firm operates and how it copes with it. Since its introduction by Cohen and Levinthal (1989, 1990), the concept of AC has been further elaborated. Zahra and George (2002) refine the concept of AC as a set of capabilities that underlie the processes identi- fied by Cohen and Levinthal (1990), from the acquisition and assimilation to the transformation and exploitation of external knowledge. They propose that the first two dimensions make up an organization potential AC, which is intended to capture the uncertainty associated with how well the firm will be able to exploit the knowl- edge. The last two dimensions make up an organiza- tion realized AC. Todorova and Durisin (2007) suggest

Lewin, Massini, and Peeters: Microfoundations of Internal and External Absorptive Capacity Routines Organization Science 22(1), pp. 81���98, ��2011 INFORMS 83 an elaboration of the model proposed by Zahra and George (2002) by reinstating the role of recognizing the value of external information, transformative pro- cesses, and regimes of appropriability. In addition, they reposition the role of social integration mechanisms and power relationships, and suggest the need for incorpo- rating feedback loops for a dynamic representation of absorptive capacity. Lane et al. (2006) propose a refinement to the original Cohen and Levinthal model of AC by intro- ducing a sequential process: recognizing and under- standing potentially valuable new external knowledge through exploratory learning, assimilation of new knowl- edge through transformative learning, and using the assimilated knowledge to create new knowledge through exploitative learning. These three competing elabora- tions of the AC construct clearly correspond to the ���tidy- ing up with typologies��� stage discussed by Hirsh and Levin (1999), and consistently emphasize the dynamic nature of AC. Interestingly, Lane et al. (2006) also rec- ognize that AC has gone through a process of reification in the extant literature. Lane et al. (2006) identified 14 academic journals that published five or more papers citing the Cohen and Levinthal (1990) paper between July 1991 and June 2002. Of the 289 papers, surprisingly, they identified only 6 papers that they considered to be most central in the absorptive capacity literature. Of the six papers, four (Mowery et al. 1996, Dyer and Singh 1998, Koza and Lewin 1998, Lane and Lubatkin 1998) address knowl- edge sharing and assimilation in interfirm relationships. One paper (Szulanski 1996) focuses on the impediments to the transfer of best practices within the firm. The sixth paper (Zahra and George 2002) provides the basis for the process model of AC proposed by Lane et al. (2006). Lane et al. (2006) highlight three major shortcomings of the existing literature on AC. First, few researchers have attempted to revise the definition of AC. Second, little attention has been given to the processes underlying AC. And third, it has almost exclusively been measured in the context of R&D. Empirical studies make attributions to the role of AC in innovation, firm adaptation, successful strategic alliances, and knowledge creation (e.g., Mowery et al. 1996, Helfat 1997, Simonin 1999, Autio et al. 2000, Steenma and Corley 2000, Ahuja and Katila 2001, Hill and Rothaermel 2003). Lewin et al. (2008) provide an overview of research that addresses the construct of AC conceptually and empirically. Following Cohen and Levinthal (1989), most empirical studies that make attributions to the AC concept rely on an R&D indi- cator (e.g., Veugelers 1997, Rocha 1999, Stock et al. 2001, Wenpin 2001), patents (cross-citations of patents in alliances Mowery et al. 1996), or coauthored papers as a mediating factor for connectedness between actors (Cockburn and Henderson 1998). Although these vari- ous proxy measures of AC are only an indirect mea- sure of AC, they have been used as both dependent and explanatory variables in empirical studies of high-tech firms and industries. The organization and processes of R&D are very likely only one component of internal AC capabilities for developing and adopting technological innovations. In a broader sense, innovation encompasses product, process, organizational, and market innovations (Schumpeter 1942). If these are the object of the anal- ysis, R&D, patents, and citations only indirectly repre- sent and capture partial aspects of capabilities related to valuing new, external information, its assimilation, and its application to commercial ends. Another stream of empirical studies has used case studies or survey instruments, normally using self-reports to make attribu- tions about AC (e.g., Szulanski 1996, Lane and Lubatkin 1998, Lane et al. 2001). But, similar to measures of R&D or patents, these are not informative of organi- zational structures or specific routines or processes that constitute AC and distinguish between AC capabilities of different organizations. The empirical studies using the AC construct summarized by Lewin et al. (2008) treat AC as a black box or as an exogenous variable (see also Foss and Pedersen 2002). In other words, organi- zational practices are used as outcomes or as indirect measures for making attributions about the AC of orga- nizations and its mediating role for innovation, change, resilience, flexibility, etc. There is a lack of direct obser- vation or measurement of the routines that make up AC, and a need for better understanding and operationaliza- tion of the AC concept (Joglekar et al. 1997, Matusik and Heeley 2001, Zahra and George 2002). Lewin and Massini (2003) represent a specific attempt at identifying the constitutive elements of AC. They decompose the concept of AC into two elements: inter- nal and external AC capabilities. They note that the prior literature focused on the exploration of knowledge in the external environment and its assimilation while neglecting the role of internal exploration of knowledge (new knowledge creation) and its assimilation. In their conceptualization, internal AC capabilities refer to man- aging the processes of internal variation, selection, and replication (VSR) described in evolutionary economics (Nelson and Winter 1982). External AC subsumes the management of exploration for new knowledge in the external environment and its assimilation. In this paper, we build on and extend the Lewin and Massini (2003) framework to identify the microfoundations of absorp- tive capacity in the form of a metaroutine taxonomy underlying absorptive capacity and their expression in organizations in the form of practiced routines. These are discussed in the section following the brief review of research on organizational routines below.