Management Learning http://mlq.sagepub.com/ Contrasting Perspectives on the Diffusion of Management Knowledge : Performance Management in a Norwegian Multinational Jon Erland Lervik and Randi Lunnan Management Learning 2004 35: 287 DOI: 10.1177/1350507604045607 The online version of this article can be found at: http://mlq.sagepub.com/content/35/3/287 Published by: http://www.sagepublications.com Additional services and information for Management Learning can be found at: Email Alerts: http://mlq.sagepub.com/cgi/alerts Subscriptions: http://mlq.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations: http://mlq.sagepub.com/content/35/3/287.refs.html Downloaded from mlq.sagepub.com at SAGE Publications on October 12, 2010 Management Learning Copyright © 2004 Sage Publications London, Thousand Oaks, CA and New Delhi www.sagepublications.com Vol. 35(3): 287–302 1350–5076 Jon Erland Lervik and Randi Lunnan BI Norwegian School of Management, Norway Contrasting Perspectives on the Diffusion of Management Knowledge Performance Management in a Norwegian Multinational Abstract This article presents an in-depth case study of the adoption patterns of performance management within a Norwegian multinational enterprise. The case allows the contrasting of a set of theoretical perspectives that provide different accounts of management knowledge diffusion. We review four perspectives named Conformity, Transfer, Translation and Local Modification, and find that each perspective can account for only parts of the observed adoption patterns in the multinational. The perspectives complement each other and together they provide more comprehensive ‘lenses’ for studying the diffusion of management knowledge. For future research, we suggest three strategies: to develop multi-perspectives typologies of diffusion outcomes, to develop multi-perspective process models, and to develop a meta-theory with propositions specifying conditions when different outcome types will be prevalent. Key Words: conformity; local modification; management ideas; multinational enterprises; performance management; translation; transfer Over the last few decades, media, consultants and business schools have provided organizations with an increasing number of new ideas and recipes for organizing (Sahlin-Andersson and Engwall, 2002). Often, planned organizational change efforts are led by the introduction of externally derived ‘best practices’ (Røvik, 1998). This has spurred our interest in how management knowledge that has been developed elsewhere may affect recipient organizations. Previous studies from the New Institutionalism perspective have taken more interest in carriers and sources of management knowledge than in its impact on individual organizations (Meyer, 1996; Sahlin-Andersson and Engwall, 2002). Business history approaches have primarily focused on the impact of management knowledge at regional, national or industry levels (e.g. Kipping and Bjarnar, 1998; Westney, 1987; Zeitlin and Herrigel, 2000), although a couple of case studies DOI: 10.1177/1350507604045607 Downloaded from mlq.sagepub.com at SAGE Publications on October 12, 2010 288 Management Learning 35(3) exploring organization level outcomes exist (Cailluet, 1998; Glimstedt, 1998). Other theoretical perspectives have examined knowledge transfer between subunits in multi-unit firms (Szulanski, 1996) and multinational enterprises (Kogut and Zander, 1993). However, different perspectives suggest a variety of ways in which ‘external’ management knowledge affects recipient organizations: as rationalized myths with little impact beyond formal structures and ‘talk’ (Brunsson, 1989; Meyer and Rowan, 1977/1991), or as rational tools to improve the effectiveness of organizations (Huselid, 1995; Szulanski, 1996). In addition, accounts vary regarding whether diffusion leads to homogenization of organizational structures and processes (Engwall, 1999; Meyer and Rowan, 1977/1991), or whether it leads to recombination and hybridization, where variety and innovation of organizational forms can be the result (Westney, 1987; Zeitlin and Herrigel, 2000). These contrasts represent only a short outline of the different theoretical traditions that approach the same phenomenon. Different theoretical perspectives give most attention to particular types of outcomes, because they start from different assumptions about the nature of management knowledge. We find perspectives that conceive management knowledge as a predefined, reified object adopted by organizations/subunits (Meyer and Rowan, 1977/1991; Szulanski, 1996) and other perspectives conceiving management knowledge as constructed via a process of diffusion (Cummings and Mohrman, 1987; Czarniawska and Sevón, 1996a). Perspectives also differ in whether they focus on technical or symbolic aspects of management knowledge (Alvarez, 1998). However, cross-fertilization and debate between these perspectives may be stifled as these traditions address different levels of analysis (organizational fields, organizations), or conceive their object of study in different ways (e.g. as transfer of ‘best practices’ or as diffusion of ‘management ideas’). We acknowledge these differences, but argue that the identified perspectives share a sufficiently common basis that enables them to be combined. This warrants closer inspection of the relative merits of each perspective in relation to our specific research question: What happens within an organization when it takes up management knowledge developed somewhere else? Our article draws on a rich, multifaceted case of the transfer of a management practice within a Norwegian multinational enterprise (MNE). Such a case study is relevant, as MNEs have been identified as central carriers in the international diffusion of management knowledge (Dunning, 1998; Sahlin-Andersson and Engwall, 2002). Also, MNEs constitute a suitable arena for studying the complexities of modern organizations, and studies of MNEs have contributed central insights to organization theory (Ghoshal and Westney, 1993). However, MNEs have received less attention from the new institutionalist perspective in sociology (Engwall, 1999). Studies within international business, evolutionary economics and the resource-based view have examined transfer of innovations and best practices within multinational and multiunit organizations (Ghoshal and Bartlett, 1988; Kogut and Zander, 1993; Szulanski, 1996). These perspectives differ from new institutionalist sociology, as they primarily conceive transfer as replication of technical/economic aspects of knowledge, with less focus on possible modification/adaptation and on symbolic aspects of management knowledge (Kostova and Roth, 2002). Downloaded from mlq.sagepub.com at SAGE Publications on October 12, 2010 Lervik & Lunnan: The Diffusion of Management Knowledge 289 Our study aims to contribute in the following ways: first, it will be a robust ‘test’ of variation in adoption outcomes, since the empirical setting of a single MNE case is expected to minimize variation. We examine a prescribed management practice transferred from corporate headquarters to subunits in a direct hierarchical relation to the sender. In this setting we would expect considerably less variation than in the case of an abstract management idea adopted by recipients in a nonhierarchical relationship vis-à-vis the sender (Oliver, 1991; Røvik, 1998). Second, our theory review and case point out uncharted aspects of management knowledge in the different existing theoretical perspectives, and suggest avenues for expanding these. Building on existing approaches, our study demonstrates that management knowledge is both technical and symbolic; in one context symbolic aspects can be most salient, in other contexts the technical aspects come to the forefront. Third—on a more general level—our article identifies theoretically important similarities and differences between existing perspectives, which might lead to more informed cross-fertilization and thus prepare the ground for more multi-perspective studies. In the next section we present the case, with four types of adoption patterns in the MNE’s subunits. Thereafter we review theoretical perspectives that can each account for different types of observed adoption patterns. In the last section we discuss the boundaries between perspectives, managerial implications, and suggest areas for further research. Performance Management in a Multinational: Four Adoption Patterns Multi is a Norwegian multinational conglomerate operating in energy-intensive industries, with 50,000 employees in 60 countries worldwide (2003). In 1998, the corporate management team decided to introduce a standardized performance management (PM) system throughout its divisions. We can identify multiple, partly overlapping, purposes for this decision: to develop a culture of increased accountability and performance-orientation, to achieve higher consistency and quality in the leadership development processes and, more indirectly, to strengthen the capacity of headquarters to monitor and control managerial resources in the subunits. Additionally, use of the PM process was intended to strengthen the internal standing and the strategic role of the corporate human resource department, and to strengthen the company reputation towards existing and potential employees. The research team has conducted ongoing research on the MNE since 1998. Here we draw on approximately 40 interviews conducted from March 1998 through December 2000, and documentation concerning the company’s HR strategy and practices, including the design process for the new PM system, covering the period 1995 to 2000. In addition, the authors have carried out extensive participant observation in corporate HR staff meetings, project groups/ task forces, and PM review meetings in all three business areas of Multi. Finally, we have conducted three company-wide surveys on PM implementation in 1999, 2000, and 2001. The case reported in this article was developed gradually from chronological accounts with data and low-level inferences of individual business units to a more theoretical/interpretative case account (Pettigrew, 1990). The final Downloaded from mlq.sagepub.com at SAGE Publications on October 12, 2010 290 Management Learning 35(3) case descriptions are the result of both authors’ data/observations and joint analysis and case writing. The PM process is in widespread use in Multi; it was formally adopted by 15 of the company’s 18 divisions. Here we present four adoption patterns in selected subunits. Intermediate forms could also be observed, but the reported outcomes represent ‘archetypes’ that can throw light on different aspects of this multifaceted phenomenon. Conformity Several units followed prescribed procedures exactly by the book, but only partially integrated the PM process with the existing management practices. Procedures were followed, leaders held appraisals with subordinates, jointly evaluating performance and leadership skills and agreeing on development plans. The units held meetings as prescribed, sent representatives to all the common corporate meetings, and provided the data (on participants, processes and outcomes) as requested by headquarters. These actions, however, had little impact on the ‘real’ activities and management practices within the units. They seemed to be more a ritual to satisfy corporate expectations for what constituted legitimate HR practices than meaningful activities to improve efficiency and learning. One unit went through a large restructuring; plants were closed and 20 percent of the employees laid off. This unit faced considerable challenges and important decisions concerning HR: to identify the right people to keep, to focus on core business activities, and to carry out the restructuring process fast, for example by putting together well performing teams. In this process, the new PM practice could have provided the basis for many tough personnel decisions, since a lot of information regarding individual strengths and weaknesses had already been collected. The PM practice was not used, however. Apparently, those in charge did not regard the information in PM as sufficiently tailored and detailed for the hiring and firing decisions. Thus the PM process was carried out, but it did not have an impact on the wider management processes in the unit. We label this pattern ‘Conformity’: the prescribed idea was enacted faithfully, but decoupled from other management practices and the ongoing operations of the business unit. Efficiency outcomes were not prevalent, but gaining legitimacy with HQ seemed to be an underlying motivation. As a large division within the strategic core of the company, this unit might have experienced a higher conformity pressure than, for example, smaller units as described in the case of ‘Modification’. Transfer Another diffusion pattern of PM was the complete implementation of the prescribed procedures and their integration with existing management practices. This occurred in a business unit that was central in the PM design and pilot phase in 1998. By 2000, the PM process was well established in its prescribed yearly cycle with appraisals, review and follow-up meetings. The process involved people on all levels, down to the foreman in some plants. The division manager participated actively in reviewing and giving feedback to people in all parts in the organization. Downloaded from mlq.sagepub.com at SAGE Publications on October 12, 2010 Lervik & Lunnan: The Diffusion of Management Knowledge 291 Several managers who participated actively in appraisals and the review meetings commented on their importance for improved cross-department cooperation and knowledge sharing. The meetings were seen as very useful in verifying assessments of employees through input from internal ‘customers’ who had first-hand experience with the candidates discussed, and in gaining more knowledge about people and competences in other parts of the organization. In this case, the division went beyond meticulously following prescribed procedures in the PM process and achieved a considerable degree of integration with the corporate business planning process. The division and business unit objectives taken from the PM plan formed the basis for setting performance objectives for managers and professionals in their individual appraisals. In addition, information collected through appraisals and review meetings was used actively by HR managers, such as when selecting candidates for a position or for high-profile management development programmes. We label this pattern ‘Transfer’: the prescribed practice was fully implemented and integrated with existing management practices. The division’s participation in the design and pilot phase may be an important reason for this faithful enactment; it ensured high congruence between the division’s own problem formulation and the PM as the corporate ‘solution’. Translation In two business units, we could see a different reaction to the symbolic and the technical elements of the PM process. Procedures, meeting requirements and forms were adopted without reluctance and rapidly replaced existing appraisal systems. However, we found small but significant modifications to the symbolic aspects of the PM, namely the notion of ‘high potentials’—or promising young employees. The two units rejected the label ‘high potentials’. One vice president said: We will not use the term ‘high potentials’ since we have no capacity for programs for them. We had burned ourselves here. We have used the word previously; people got expectations, became disappointed and left the company. This symbolic aspect of PM was downplayed to maintain consistency between official talk and the actual systems and processes the business unit had at its disposal. Using the label ‘high potentials’ was seen as having possible disruptive effects on the unit’s professionals and managers. In another unit, the rationale for rejecting the label was primarily an issue of identity. Many employees talked about a ‘Norwegian corporate culture’, and selecting out ‘high potentials’ was seen to go against the established norms and values of egalitarianism of the organization. According to the division HR executive: ‘It could be very negative and discouraging for those who were not singled out as high potential; does this mean that they do not have a future in the organization?’ The problem was not the selection in itself, but the idea of openly differentiating between two groups of employees. In this unit, certain professionals were actually singled out and discussed in management team review meetings, but without the candidates knowing. The candidates in question used the same ‘employee’ appraisal form as others instead of the ‘leadership’ appraisal forms that Downloaded from mlq.sagepub.com at SAGE Publications on October 12, 2010 292 Management Learning 35(3) were part of the PM process. Thus the official image of equal treatment was upheld. In these two units we found significant symbolic modifications to the PM process, while the technical, procedural aspects were taken up and carried out as intended by headquarters. This symbolic modification therefore did not constitute a threat concerning legitimacy with top management. We label this pattern ‘Translation’. Local Modification We also observed cases of innovation and hybridization in the implementation of the PM process. One business unit went much further in utilizing the tool for competence development purposes than corporate prescriptions suggested. This unit had its headquarters in Norway but most of its production and sales activities abroad. They seized PM as an opportunity to get a complete overview of managerial resources, and conducted a large two-day meeting, where all senior managers were present. In this meeting, all middle managers and especially the high potentials were reviewed—their competence and performance were discussed as well as development opportunities, rotations, training, and career interests. These activities covered much more territory than the corporate PM policy suggested. The foreign business unit of another division used the PM tool to support management team development. The new appraisal forms and leadership criteria caught the interest of the subunit manager, who saw how these could help him not only in discussing individual performance, skills and careers but also with team processes. While the PM had been designed as an individual level tool, it was here taken up for other purposes. These cases show how the standardized PM process was extended and applied in a different way from that originally intended by corporate headquarters. We found significant modifications to the technical elements of the PM, while at the same time the symbolic aspects were played down. We label this pattern ‘Local Modification’. These units displaying a higher degree of local initiative were relatively smaller and in lower echelons of the hierarchy, they may therefore have experienced less pressure for conformity and were left more to their own device. Discussion: Contrasting Perspectives on Diffusion Our case illustrates four adoption patterns that were subunits’ reactions to externally derived management knowledge. Conformity is characterized by complete imitation, but with a high degree of decoupling; transfer consists of complete imitation plus a thorough integration with other management practices; translation is characterized by significant symbolic modifications; and finally, local modification reflects considerable modification to technical aspects of the introduced management knowledge. These four patterns may not constitute the entire domain of possible outcomes, but they represent theoretically important variations in how organizations utilize management knowledge. Downloaded from mlq.sagepub.com at SAGE Publications on October 12, 2010 Lervik & Lunnan: The Diffusion of Management Knowledge 293 Figure 1 Four types of perspectives on diffusion Knowledge as reified object Ontological assumptions about management knowledge Knowledge constructed Symbolic aspects Conformity Translation Transfer Local Modification Highlighted aspect of management knowledge Technical aspects In the existing literature, we find theoretical perspectives that correspond to our four patterns. These perspectives vary according to (1) their emphasis on symbolic or technical aspects, and (2) whether they see management knowledge as reified or constructed. These two dimensions form a theoretical space illustrated in Figure 1. They are continua rather than dichotomies. The four quadrants in Figure 1 constitute ‘ideal-types’ characterized by different assumptions about management knowledge. Conformity: Management Knowledge as Reified Symbols New Institutionalism in organizational analysis has dealt extensively with the diffusion of management ideas. Meyer and Rowan’s (1977/1991) initial formulations were a reaction to rational, efficiency accounts of why firms adopted new management practices. They argued that ‘institutionalized products, services, techniques, policies, and programs function as powerful myths, and many organizations adopt them ceremonially’ (p. 41). As new rules arise in the institutional environment, organizations incorporate them as structural elements in their formal organization. With competing technical-economic forces, loose coupling is conceived as the prevalent pattern, where talk and action are separate domains, and management knowledge affects the formal structure of organizations more than their operational aspects (Brunsson, 1989). DiMaggio and Powell (1983/1991: 64) asked ‘why there is such startling homogeneity of organizational forms and practices’. Similarity, not variation, was of interest in this research, and diffusion of management knowledge—techniques, policies and programmes—was a central area for empirical study (see e.g. Downloaded from mlq.sagepub.com at SAGE Publications on October 12, 2010 294 Management Learning 35(3) Fligstein, 1985; Orrú et al., 1991; Strang, 1987). Researchers went on to demonstrate ‘the very existence of these processes of diffusion by measuring the resulting formal similarity of both influencing and influenced, diffuser and receiver organizations’ (Alvarez, 1998: 23), and empirical studies captured this formal similarity. The implicit view is that organizational fields define and shape management knowledge, with little scope for negotiation or modification by individual firms. Actors are conceptualized as passively adapting to external, structural influences, with a ‘theory of individual action, which stresses the unreflective, routine, taken-for-granted nature of most human behavior and views interests and actors as themselves constituted by institutions’ (DiMaggio and Powell, 1991: 14). Some recent contributions have emphasized the discretion of firms in adoption (Oliver, 1991); however, the original argument from New Institutionalism represents management knowledge as a predefined, reified object, where a firm’s options are to adopt or not to adopt. Furthermore, this perspective highlights symbolic rather than technical aspects of management knowledge, by emphasizing ‘myth and ceremony’ and loose coupling of formal structure and operations. It corresponds to our pattern of ‘conformity’, where the subunit closely imitated directives from headquarters. The comprehensive enactment of appraisals, review meetings with information gathering and analysis did not have a wider impact on other management practices and decisions, though; it was decoupled. This suggests that legitimacy towards headquarters was the main motivation for and most prevalent outcome of this diffusion pattern. Transfer: Management Knowledge as Reified Techniques Management knowledge can be seen as a potential source of sustainable competitive advantage, a scarce resource residing in one part of an organization, which needs to be replicated and transferred to other parts to achieve synergies and scale economies. We find the main proponents of this perspective within the resourcebased view of strategy (Hansen, 1999; Szulanski, 1996, 2000) and international management (Ghoshal and Bartlett, 1988; Kogut and Zander, 1993). A core contribution is Szulanski’s work on the transfer of best practices, which examines both technical and organizational innovations. This perspective views management knowledge not as abstract ideas or discourses, but mainly as constituted in idiographic organizational practices, where practice ‘refers to the organization’s routine use of knowledge and often has a tacit component, embedded partly in individual skills, partly in collaborative social arrangements’ (Szulanski, 1996: 28). Synergies are achieved through complete and perfect imitation of a complex bundle of organizational routines and practices between multiple locations in the firm. For Szulanski, deviations from plans and modifications to ‘best practices’ are seen as resistance and result in a ‘less-thansuccessful’ transfer. The underlying rationale is obtaining economic gains through effective replication. This perspective does not pay attention to adaptation, reworking and local innovation. Instead, it emphasizes structural explanations for transfer success. Several characteristics of knowledge influence the transfer, such as its codifiability, teachability, complexity, provenness, and causal ambiguity (Hansen, 1999; Szulanski, 1996; Zander and Kogut, 1995). Characteristics of senders, recipients and the relationship have also been examined (Ghoshal and Downloaded from mlq.sagepub.com at SAGE Publications on October 12, 2010 Lervik & Lunnan: The Diffusion of Management Knowledge 295 Bartlett, 1988; Gupta and Govindarajan, 2000). Szulanski identifies causal ambiguity of a practice, strained relationships between sender and recipient, and recipients’ lack of absorptive capacity as the most significant barriers to knowledge transfer. This view corresponds to our pattern of ‘transfer’, where prescribed practices were not only imitated closely, but also integrated with existing management practices. Central actors in the unit had deep knowledge of the PM process and saw it as efficient, emphasizing the technical aspects of PM. The unit spent time integrating PM within its organization, thus fulfilling both the detailed prescriptions and the more general technical rationale behind the idea. Management knowledge was replicated and integrated with operations for efficiency purposes. Translation: Management Knowledge as Constructed Symbols Two edited books convey different flavours of this constructed view of management knowledge. Czarniawska and Sevón (1996b) were inspired by Latour’s translation model (1987). Their volume includes several case studies showing how management ideas and recipient organizations mutually influenced and modified each other. Sahlin-Andersson and Engwall (2002) focus on carriers and sources of management knowledge, with contributions at national and industry level that emphasize the importance of actors in diffusion. Management knowledge is mainly seen as circulating ideas, discourses and texts: ‘[M]anagement knowledge only exists in and through the processes of circulation’ (Sahlin-Andersson and Engwall, 2002: 6); what is circulating are idealized, de-conceptualized accounts of other organizations’ management practices (Sahlin-Andersson, 1996). Management ideas can be transformed and become institutionalized in recipient firms; but this perspective emphasizes the institutionalization of new language, concepts and categories rather than actions. The notion of loose coupling is upheld to a large degree (Meyer, 1996). Translation is a central concept used in two ways; as modification of texts (Røvik, 1998), and as acts of political persuasion to enroll support for an idea (Latour, 1987). Management knowledge can be seen as the local, temporary result of ongoing translation processes, where other texts and discourses are enrolled to strengthen the persuasiveness of a management idea (Czarniawska and Joerges, 1996; Czarniawska and Sevón, 1996a). Travelling management ideas are strengthened by legitimizing narratives (Czarniawska and Joerges, 1996), and by quasiresearch (Røvik, 1998). The identification of recipients with senders is an important precondition for diffusion (Strang and Meyer, 1994); but in this perspective this is not considered only an exogenous contextual characteristic. Sahlin-Andersson (1996) shows how a sense of identity results from the work of actors in editing texts, promoting similarities and downplaying differences between sender and potential recipients. Thus texts, symbols and discourses are what characterize management knowledge and, at the same time, are the materials and resources drawn on for strengthening the persuasiveness of circulating management knowledge. These studies have emphasized the idealist, symbolic aspects of how ideas influence organizations; how the adoption of ideas can (re)shape firm identity (Sevón, 1996). Czarniawska and Joerges (1996) emphasize institutionalization of Downloaded from mlq.sagepub.com at SAGE Publications on October 12, 2010 296 Management Learning 35(3) new concepts as a central organizational outcome. Other studies emphasize carriers and sources, and pay less attention to the impact on recipients (SahlinAndersson and Engwall, 2002). To sum up, Scandinavian institutionalism focuses on management knowledge as constructed, and giving emphasis to the symbolic aspects of diffusion. This perspective points to the salient aspects of the ‘translation’ pattern we have presented: specific symbolic aspects—publicly declaring certain employees to be ‘high potentials’—were modified to fit with the recipient culture or with current management practices. This minor modification in objective terms was significant in altering the meaning of the PM process; it was democratic, and it did not promise special treatment to a chosen elite. Scandinavian institutionalism provides a well-suited framework for capturing and accounting for this type of diffusion outcome. Local Modification: Management Knowledge as Constructed Techniques Management ideas are seen to have important effects on organizations, though not always the intended effects (Beer et al., 1990; Hennestad, 1998). Organizations introduce management ideas with the aim of efficiency gains. Scholars have examined the impact of received management knowledge on, for example, productivity and product quality (Beyer et al., 1997), safety (Marcus, 1988), task specialization, deskilling/reskilling (Zuboff, 1988), and autonomy/work democracy (Levin, 1997). In these studies, management knowledge is not seen as predefined, but rather as locally constructed from abstract principles. Organizations ‘need to learn how to translate the general practices into organizationallyrelevant behaviors, structures and processes’ (Cummings and Mohrman, 1987: 276). Cummings and Mohrman (1987: 283) recommend a ‘self-design strategy’, emphasizing that received ideas are relatively underdeveloped, that implementation entails unanticipated consequences, and that this demands considerable organizational learning. Autonomy, participation and local experimentation are important to make ideas and innovations work in the local context (Levin, 1997). This constructed view encompasses elements of both concretizing and modification. Abstract ideas must be transformed into concrete recipes for action, but also the content of ideas is likely—or supposed—to be modified. Surprises and deviations from the original idea are sources of learning and reflection, rather than errors to be corrected. Beyer et al. (1997) show how two similar firms in similar institutional and competitive environments end up with very different results when implementing total quality management (TQM). In one firm, management encouraged local experimentation and partial adoption suited to the needs of different departments. The other firm had a tightly controlled implementation process, where deviations were considered errors. The first organization experienced more long-lasting effects on the quality of work processes and products from adopting TQM. Marcus (1988) shows a similar pattern when new standardized safety rules were issued for all nuclear reactors after the Three Mile Island accident. The sites with an autonomous approach to adoption improved their safety statistics compared with sites that adopted the new policies in a rulebound manner. This perspective puts the locally constructed, ‘workable’ processes and activities at the centre of attention. Management ideas are general prescrip- Downloaded from mlq.sagepub.com at SAGE Publications on October 12, 2010 Lervik & Lunnan: The Diffusion of Management Knowledge 297 tions and building blocks that have to be transformed and constructed to fit the local setting. Our pattern of ‘local modification’, using PM for team development or in-depth personnel reviews, exemplifies important aspects in this implementation perspective: creative local solutions are developed, where local needs and external management knowledge qua ideas meet; received ideas can be a source of inspiration, not only for direct imitation. Management knowledge is used to deal with ‘real’ local problems, and symbolic aspects receive less attention. Transition Zones between Perspectives The theoretical perspectives differ on two critical dimensions, conceiving management knowledge as reified/constructed and symbolic/technical. These dimensions are not new or revolutionary; rather they are well established and recognized, to some extent defining boundaries that researchers in the respective areas often do not cross. We hope that our study can yield new insights into how different perspectives can complement and build on each other. The two dimensions form two different types of boundaries. Horizontally, ‘constructed’ perspectives position themselves in relation to ‘reified’ perspectives— criticizing stringent, unrealistic assumptions of the latter and emphasizing additional, complementary insights of the former. The ‘conformity’ perspective is criticized by ‘translation’ perspectives (Clark and Geppert, 2002; Sahlin-Andersson and Engwall, 2002), and the ‘transfer’ perspective is criticized by ‘local modification’ perspectives (Cummings and Mohrman, 1987). Along the symbolic–technical continuum, the picture is more complex. We find intermediate positions on the reified side and perspectives viewing knowledge as constructed, but otherwise little cross-referencing between perspectives. Kostova studied the transfer of TQM into MNE subsidiaries, and introduced the concept ‘strategic organizational practice’ to capture practices that were people rather than technology oriented. In the same way that organizations can be ‘infused with value’ (Selznick, 1957), such practices can gain a status beyond their assumed technical efficiency. Successful transfer ‘is determined by the transferability of meaning and value, in addition to the transferability of knowledge’ (Kostova, 1999: 311). Value-impregnated practices are subject to institutional duality in MNE subsidiaries, and Kostova and Roth (2002) found that coercive, normative and mimetic factors of subsidiaries’ national environments had a strong impact on transfer success. Competing ‘motivations’ for adoption were recognized, and Kostova and Roth distinguished between ‘active adoption’ and ‘ceremonial adoption’, where active adoption referred to units where subsidiary members were committed to and believed in the practice. These studies thus integrate aspects of the ‘transfer’ and ‘local modification’ patterns; however, these studies still see TQM as a reified object, and replication is the desired outcome. Another transition zone can be identified in business history approaches. Westney’s (1987) account of the industrial and social transformation of Meiji Japan demonstrated that emulating others entailed both imitation and innovation; that is, seeing management knowledge as constructed rather than as reified. The social transformation described by Westney encompassed changing traditional values and norms, as well as industrializing and improving military capacity in Japan, so both Downloaded from mlq.sagepub.com at SAGE Publications on October 12, 2010 298 Management Learning 35(3) the symbolic and the material impact of the emulation of European organizational models are studied. Recent research on the Marshall Plan and the rebuilding of Europe after World War II also takes a broad perspective encompassing both symbolic and technical effects of diffusion of management knowledge (Djelic, 1998; Kipping and Bjarnar, 1998; Zeitlin and Herrigel, 2000). Other studies explore outcomes in individual firms. Cailluet’s account of scientific management in the aluminium producer Pechiney is a story of both efficiency gains and political-symbolic aspects, such as how leaders associated with new management knowledge increased their influence in the firm, and how ‘management considered it very important to adopt a modern and ‘“US-like” organization’ (Cailluet, 1998: 204). Similarly, Glimstedt’s (1998) account of Volvo introducing a US concept for remuneration—methods–time–measurement or MTM—shows how the new idea provided both a technical and cultural challenge (it was seen to go against established worker norms of appropriateness) and a practical challenge by integrating MTM with established administrative routines for pay. These accounts illustrate how management ideas at the same time have both important symbolic and technical aspects. Conclusion Earlier studies of diffusion have primarily tended to start from one theoretical perspective and examine specific phenomena in different empirical domains. Our contribution is to draw on multiple perspectives in examining one case, in exploring the multifaceted nature of diffusion within a multinational enterprise. The MNE case also highlights the pluralistic nature of complex organizations, and demonstrates the importance of multiple perspectives for understanding diffusion in organizations. Our case study demonstrates considerable variation in the ways in which one PM practice is taken up, even by subunits of the same MNE. Furthermore, different perspectives can each account for only parts of the observed variation. We believe that our approach can yield new insights into how different perspectives can complement and build on each other. By drawing on multiple perspectives, we see how management knowledge is both symbolic and technical, and in some instances can be described as reified, in others as constructed. Much of the literature tends to emphasize particular aspects based on the specific research issues examined. However, efforts at combining these views may yield additional insights. Adler and Borys (1993) point out that integration of political-symbolic approaches and technical-economic approaches is especially important in explaining the behaviour of organizations: ‘[A]t the societal level such forces are to some extent institutionalized in distinct, relatively autonomous institutional systems such as “the economy” and “the State”’ (p. 661, citing Polanyi, 1944), while these forces meet and interrelate at the organization level. We therefore argue that it is even more important to aim for the integration of perspectives when we address diffusion at organization level. Our review has identified specific points of intersection between perspectives that can stimulate further research. Our suggested typology for diffusion outcomes within organizations is slightly different from others, such as the one proposed by Kostova and Roth (2002). Classifications from different perspectives can be refined Downloaded from mlq.sagepub.com at SAGE Publications on October 12, 2010 Lervik & Lunnan: The Diffusion of Management Knowledge 299 and integrated to develop a richer vocabulary. We used a cross-sectional approach, looking at differences in patterns at one particular time. Another way to increase the understanding of diffusion is to do longitudinal studies of diffusion processes into and within individual organizations. These studies could also draw on stage models of process taken from various perspectives (e.g. Czarniawska and Joerges, 1996; Szulanski, 1996, 2000). The main aim of this study has been to demonstrate variation rather than to try to explain it. When, and under what conditions, certain outcomes are most likely remain important questions that should be addressed in further research. Our proposed typology can be a promising start for initiatives towards a meta-theory of diffusion patterns in MNEs. Current theories can give some indications as to when we can expect symbolic rather than technical aspects to dominate; for instance late adopters in an organizational field (Fligstein, 1985; Rogers, 1995), firms with less clear instrumental goals (Brunsson, 1989), and recipients in a resourcedependency situation vis-à-vis senders (Kostova and Roth, 2002). Similarly, our case suggests that not only the objective dependency but also the visibility and size of recipients may influence their perceived need to cater to external legitimacy concerns. On the second dimension, we can expect higher degrees of modification when cultural and institutional differences are large (Kedia and Bhagat, 1988), and with recipients experiencing conflicting institutional pressures (Oliver, 1991). These are only some indications, and efforts to develop a consistent theoretical framework could constitute a way forward in integrating various theoretical traditions. It is beyond the scope of this article to point to managerial prescriptions, but we can draw attention to how these different perspectives on management knowledge convey very different messages to managers. Is adoption of management knowledge about ‘myth and ceremony’ or about rational efficiency seeking? Is diffusion a predictable process of replicating predefined knowledge, implying a high degree of standardization, predictability and control, or is diffusion inherently innovative, where local learning and adaptation are critical? Our research supports previous studies in that the transfer of management knowledge may fulfil many roles, even at the same time: gaining external legitimacy, shaping internal identity, and achieving synergies through replication. Alternatively, it may change an organization’s work processes in less predictable ways, when received knowledge is used more as a source of inspiration than as a concrete template for imitation. In this article, we have focused on how different theories conceive management knowledge as reified or constructed, as technical or symbolic. These categories can be powerful concepts for managers, consultants or other change agents when doing planned change. Transfer efforts guided by unreflective or simplified notions of, for example, the replicability of management knowledge may yield disappointing results. Managers perceiving only technical knowledge and efficiency outcomes may overlook other important effects of diffusion on firms’ identity (Sevón, 1996) or shared language (Czarniawska and Joerges, 1996). Similarly, managers thinking only in terms of replicating may stifle important processes of local learning and experimentation in their quest for ‘best practice’. Our hope is that the article increases awareness of the many faces of management knowledge, and thus facilitates reflective, thoughtful diffusion and appropriation of management knowledge developed elsewhere. Downloaded from mlq.sagepub.com at SAGE Publications on October 12, 2010 300 Management Learning 35(3) References Adler, P. S. and Borys, B. (1993) ‘Materialism and Idealism in Organizational Research’, Organization Studies 14(5): 657–79. Alvarez, J. L. (1998) ‘The Sociological Tradition and the Spread and Institutionalization Knowledge for Action’, in J. L. Alvarez (ed.) The Diffusion and Consumption of Business Knowledge, pp. 13–57. Basingstoke: Macmillan. Beer, M., Spector, B. and Eisenstat, R. A. (1990) The Critical Path to Corporate Renewal. Boston, MA: Harvard Business School Press. Beyer, J. M., Ashmos, D. P. and Osborn, R. N. (1997) ‘Contrasts in Enacting TQM: Mechanistic vs. Organic Ideology and Implementation’, Journal of Quality Management 2(1): 3–39. Brunsson, N. (1989) The Organization of Hypocrisy: Talk, Decisions and Actions in Organizations. Chichester: Wiley. Cailluet, L. (1998) ‘Selective Adaptation of American Management Models: the Long-term Relationship of Pechiney with the United States’, in M. Kipping and O. Bjarnar (eds) The Americanization of European Business, pp. 190–207. London: Routledge. Clark, E. and Geppert, M. (2002) ‘Management Learning and Knowledge Transfer in Transforming Societies: Approaches, Issues and Future Directions’, Human Resource Development International 5(3): 263–77. Cummings, T. G. and Mohrman, S. A. (1987) ‘Self-designing Organizations: Towards Implementing Quality-of-Work-Life Innovations’, Research in Organizational Change and Development, vol. 1, pp. 275–310. Greenwich, CT: JAI Press. Czarniawska, B. and Joerges, B. (1996) ‘Travel of Ideas’, in B. Czarniawska and G. Sevón (eds) Translating Organizational Change, pp. 13–48. Berlin: de Gruyter. Czarniawska, B. and Sevón, G. (1996a) ‘Introduction’, in B. Czarniawska and G. Sevón (eds) Translating Organizational Change, pp. 1–12. Berlin: de Gruyter. Czarniawska, B. and Sevón, G. (eds) (1996b) Translating Organizational Change. Berlin: de Gruyter. DiMaggio, P. J. and Powell, W. W. (1983/1991) ‘The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields’, in W. W. Powell and P. J. DiMaggio (eds) The New Institutionalism in Organizational Analysis, pp. 63–82. Chicago: University of Chicago Press. DiMaggio, P. J. and Powell, W. W. (1991) ‘Introduction’, in W. W. Powell and P. J. DiMaggio (eds) The New Institutionalism in Organizational Analysis, pp. 1–38. Chicago: University of Chicago Press. Djelic, M.-L. (1998) Exporting the American Model. Oxford: Oxford University Press. Dunning, J. (1998) ‘US-owned Manufacturing Affiliates and the Transfer of Managerial Techniques: The British Case’, in M. Kipping and O. Bjarnar (eds) The Americanisation of European Business, pp. 43–72. London: Routledge. Engwall, L. (1999) The Carriers of European Management Ideas. Uppsala: Företagsekonomiska Institutionen, Uppsala Universitet. Fligstein, N. (1985) ‘The Spread of the Multidivisional Form among Large Firms’, 1919–1979’, American Sociological Review 50(3): 377–91. Ghoshal, S. and Bartlett, C. A. (1988) ‘Creation, Adoption, and Diffusion of Innovations by Subsidiaries of Multinational Corporations’, Journal of International Business Studies 19(Fall): 365–88. Ghoshal, S. and Westney, D. E. (eds) (1993) Organization Theory and the Multinational Corporation. New York: St Martin’s Press. Glimstedt, H. (1998) ‘Americanization and the “Swedish Model” of Industrial Relations: The Introduction of the MTM System at Volvo in the Postwar Period’, in M. Kipping and O. Bjarnar (eds) The Americanisation of European Business, pp. 133–48. London: Routledge. Downloaded from mlq.sagepub.com at SAGE Publications on October 12, 2010 Lervik & Lunnan: The Diffusion of Management Knowledge 301 Gupta, A. K. and Govindarajan, V. (2000) ‘Knowledge Flows within Multinational Corporations’, Strategic Management Journal 21(4): 473–96. Hansen, M. T. (1999) ‘The Search-transfer Problem: The Role of Weak Ties in Sharing Knowledge across Organization Subunits’, Administrative Science Quarterly 44(1): 82–111. Hennestad, B. W. (1998) ‘Empowering by De-depowering: Towards an HR Strategy for Realizing the Power of Empowerment’, International Journal of Human Resource Management 9(5): 934–53. Huselid, M. (1995) ‘The Impact of Human Resource Management Practices on Turnover, Productivity and Corporate Financial Performance’, Academy of Management Journal 38(3): 635–72. Kedia, B. L. and Bhagat, R. S. (1988) ‘Cultural Constraints on Transfer of Technology across Nations: Implications for Research in International and Comparative Management’, Academy of Management Review 13(4): 559–71. Kipping, M. and Bjarnar, O. (eds) (1998) The Americanisation of European Business. London: Routledge. Kogut, B. and Zander, U. (1993) ‘Knowledge of the Firms and the Evolutionary Theory of the Multinational Corporation’, Journal of International Business Studies 24(4): 625–46. Kostova, T. (1999) ‘Transnational Transfer of Strategic Organizational Practices: A Contextual Perspective’, Academy of Management Review 24(2): 308–24. Kostova, T. and Roth, K. (2002) ‘Adoption of an Organizational Practice by Subsidiaries of Multinational Corporations: Institutional and Relational Effects’, Academy of Management Journal 45(1): 215–33. Latour, B. (1987) Science in Action: How to Follow Scientists and Engineers through Society. Cambridge, MA: Harvard University Press. Levin, M. (1997) ‘Technology Transfer is Organization Development: An Investigation into the Relationship between Technology Transfer and Organizational Change’, International Journal of Technology Management 14(2/3/4): 297–308. Marcus, A. A. (1988) ‘Implementing Externally Induced Innovations: A Comparison of Rule-bound and Autonomous Approaches’, Academy of Management Journal 31(2): 235–56. Meyer, J. W. (1996) ‘Otherhood: The Promulgation and Transmission of Ideas in the Modern Organizational Environment’, in B. Czarniawska and G. Sevón (eds) Translating Organizational Change, pp. 241–52. Berlin: de Gruyter. Meyer, J. W. and Rowan, B. (1977/1991) ‘Institutionalized Organizations: Formal Structure as Myth and Ceremony’, in W. W. Powell and P. J. DiMaggio (eds) The New Institutionalism in Organizational Analysis, pp. 41–62. Chicago: University of Chicago Press. Oliver, C. (1991) ‘Strategic Responses to Institutional Processes’, Academy of Management Review 16(1): 145–79. Orrú, M., Biggart, N. W. and Hamilton, G. G. (1991) ‘Organizational Isomorphism in East Asia’, in W. W. Powell and P. J. DiMaggio (eds) The New Institutionalism in Organizational Analysis, pp. 361–89. Chicago: University of Chicago Press. Pettigrew, A. M. (1990) ‘Longitudinal Field Research on Change; Theory and Practice’, Organization Science 1(3): 267–92. Polanyi, K. (1944) The Great Transformation. Boston: Beacon Press. Rogers, E. M. (1995) Diffusion of Innovations, 4th edn. New York: Free Press. Røvik, K. A. (1998) Moderne organisasjoner: Trender i organisasjonstenkningen ved tusenårsskiftet. Bergen: Fagbokforlaget. Sahlin-Andersson, K. (1996) ‘Imitating by Editing Success: The Construction of Organizational Fields’, in B. Czarniawska and G. Sevón (eds) Translating Organizational Change, pp. 69–91. Berlin: de Gruyter. Sahlin-Andersson, K. and Engwall, L. (2002) The Expansion of Management Knowledge: Carriers, Flows, and Sources. Stanford, CA: Stanford Business Books. Downloaded from mlq.sagepub.com at SAGE Publications on October 12, 2010 302 Management Learning 35(3) Selznick, P. (1957) Leadership in Administration. A Sociological Interpretation. New York: Harper & Row. Sevón, G. (1996) ‘Imitation and Organizational Identity’, in B. Czarniawska and G. Sevón (eds) Translating Organizational Change, pp. 49–68. Berlin: de Gruyter. Strang, D. (1987) ‘The Administrative Transformation of American Education: School District Consolidation, 1938–1980’, Administrative Science Quarterly 32(3): 352–66. Strang, D. and Meyer, J. W. (1994) ‘Institutional Conditions for Diffusion’, in W. R. Scott and J. W. Meyer (eds) Institutional Environments and Organizations, pp. 100–12. Thousand Oaks, CA: Sage. Szulanski, G. (1996) ‘Exploring Internal Stickiness: Impediments to the Transfer of Best Practice within the Firm’, Strategic Management Journal 17(Winter): 27–43. Szulanski, G. (2000) ‘The Process of Knowledge Transfer: A Diachronic Analysis of Stickiness’, Organizational Behavior and Human Decision Processes 82(1): 9–27. Westney, E. (1987) Imitation and Innovation—The Transfer of Western Organizational Patterns to Meiji Japan. Cambridge, MA: Harvard University Press. Zander, U. and Kogut, B. (1995) ‘Knowledge and the Speed of Transfer and Imitation of Organizational Capabilities: An Empirical Test’, Organization Science 6(1): 76–92. Zeitlin, J. and Herrigel, G. (eds) (2000) Americanization and its Limits—Reworking US Technology and Management in Post-war Europe and Japan. Oxford: Oxford University Press. Zuboff, S. (1988) In the Age of the Smart Machine: The Future of Work and Power. New York: Basic Books. Contact Addresses Jon Erland Lervik is at BI Norwegian School of Management, PO Box 580, N-1302 Sandvika, Norway. [email: jon.e.lervik@bi.no] Randi Lunnan is at BI Norwegian School of Management, PO Box 580, N-1302 Sandvika, Norway. [email: randi.lunnan@bi.no] Downloaded from mlq.sagepub.com at SAGE Publications on October 12, 2010