Printer Friendly

The intensity and extensity of knowledge and the multinational corporation as a nearly recomposable system (NRS).

Knowledge and its management are increasingly seen as the stuff that economic activity and competition are made of.(1) The plea for such a focus is accompanied, although in my view not to an adequate extent, by efforts to analyze the properties and subcategories of knowledge and their consequences for firms and management.(2)

The aim of the present paper is to contribute to this discussion. Rather than reviewing previous conceptions and dimensionalizations of knowledge, I will suggest a simple categorization in two dimensions - knowledge intensity and knowledge extensity. I will argue that the modern multinational corporation (MNC) is working with knowledge that is both intensive and extensive, in senses to be explained below. This, in turn, requires the MNC to strive to become a nearly recomposable system (NRS), rather than a nearly decomposable one, as is the explicit or implicit principle guiding the design of most hierarchical systems.(3)

The main part of the paper is devoted to a discussion of the properties of nearly recomposable systems and the ability of the MNC and other socio-economic institutions in this regard.

The Intensity and Extensity of Knowledge

"Post-industrialism" is usually seen to involve not just the increased role of intangibles - such as services, information and knowledge - in relation to tangibles. It also suggests "higher" grades of intangibles, "intelligables" if one so wishes. The knowledge intensity of an activity or a firm is the degree to which it is dependent on an internal supply of advanced, complex and recent knowledge.4 A possible measure, all imperfections admitted, is the level of education of those involved in the firm or activity.

If this was all there was to it, the management challenge would be difficult but not that difficult. Firms have had to "manage intellectuals" and deal with complex knowledge for a long time, as have other institutions, notably the military. What is new is the increased dispersion of knowledge, its extensity. By knowledge extensity, I refer to the geographical, organizational (between individual firms and between their divisions, departments, subsidiaries, individuals, etc.), and "substantive" (over technical fields and types of knowledge) dispersion of knowledge relevant to the competitive distinctiveness of the firm.(5) Operational measurement is even more difficult than in the case of intensity. One index, for the geographical extensity, would be the spatial spread of units for research, development, top levels of management, and other strategic functions in the firm. Another would be a measure of the variance of knowledge intensity between geographically or organizationally separated units. High variance implies low knowledge extensity, as in the firm that undertakes all advanced functions in one place and disperses routine activities globally. Low variance may imply the dispersion of similar activities (as in an integrated system of globally dispersed R&D units for a product) or of quite different ones (as when design units in one country coordinate with component technology ones in another).

My hypothesis is that the world economy is moving toward higher degrees of both knowledge intensity (KI) and knowledge extensity (KE).(6) However, this broad trend contains much diversity, which can be discussed along the lines suggested by Figure 1.

I have chosen to concentrate on the geographical dimension concerning extension. The terms "creation" and "exploitation" should be taken in no theological and/or Marxist senses. Exploitation refers to the utilization of given resources and bases of competitiveness. Creation entails the search for and design of new sources of strength and ways to utilize them.

Taking the combination of extremes for the two dimensions (obviously, they are both continuous), we end up with four archetypes. I will not describe them closely here, but only note that three of the four quadrants harbour respectable MNCs. The Genius Firm may well be global, but its critical resources are concentrated, usually in the home country. This is a quite common model, particularly historically, and the one forming the basis for most theoretical models of the MNC.(7)

The Concept Globalizer could also be an MNC. Considering what allows international firms in the fast food, hotel, and retail business to hold their own against local competitors in basically low-skill industries and activities, two factors stand out. First, articulated and standardized packages of end product, production process and local management systems provide low costs and a reliable, easily communicable service, usually connected with a strong brand name (McDonalds, 7-11, Coca Cola). Second, and increasingly relatedly, modern information systems provide productivity gains and control mechanisms quite suitable to low-skill employees. For example, they don't have to be able to read through the introduction of iconic-buttoned cash registers.

Many mass-service industries not usually associated with the new proletariat behind fast-food and retail counters could be added to the list of potential Concept Globalizers. Banks and insurance are perhaps the most interesting, given their weight in national and global economies. Banks are full of people with relatively little formal education. It is "high-tech" in its use of advanced information systems. However, it is no more knowledge intensive in the human sense than in the case of a truck fleet company, where also the advanced knowledge is embodied in standardized machinery. The most explicitly global of the quadrants, however, is the north-east one, containing heterarchical, transnational firms, with globally dispersed strategic resources and an intention to use them in a coordinated fashion.(8) These last words are important. MNCs may have scattered resources, even all sharing a high knowledge intensity, with no intention or possibility of combining them, such as in the classical polycentric firm.(9)

Heterarchy/transnationalism implies dispersion coupled with integration. The basic idea in the polycentric firm, as well as in the conglomerate and multidivisional (M-form) firm, is that activities are separated so that interdependencies are minimized. This kind of firm cannot exploit the opportunities and live up to the demands imposed by the global, interactive use of simultaneous intensity and extensity of knowledge. I have suggested the N-form corporation as a concept covering firms configured in ways able to continuously combine and recombine knowledge.(10) The N-form corporation aims at near recomposability of its parts. The M-form and polycentrism relies on the alternative principle of near decomposability. The term "near decomposability" is, as far as I know, Simon's.(11) In an articulate analysis of the reasons for the prevalence of hierarchical organization - in the natural as well as the social world - the desirability of sealing off different areas of high interaction between elements of the system from each other is seen as a main argument for hierarchical order. Indeed, the two concepts become almost synonymous, or at least they imply each other.

A special case is the division of work. In Simon's famous clockmaker example, the wise manufacturer arranges his production so that long series of individual parts are produced in sequence - a method with purportedly hierarchical properties.(12) This works, I would argue, under conditions of stability. When things like technology or market demand change, the supposedly stupid clockmaker may well feel better. The assembly-line hero gets stuck with a million pendula when the Japanese send him a battery-powered clock for comparison, whereas the primitive assembler of whole clocks in totality is stuck with just one unsaleable museum piece.

When such changes, or opportunities perceived within a firm, are present, more often than not there is a need to rearrange knowledge parts (competencies, practical skills, theoretical knowledge, chunks - this is Simon's term for the use of expert knowledge, etc.) over geographically and organizational boundaries and technical fields. The problem or opportunity will mostly not be most fruitfully addressed by further decomposition. (To make just a part of the pendulum in ten million units.)

I suggest the term recomposition to indicate the rearrangement of knowledge elements. A nearly recomposable system (NRS) serves to create rather than exploit. It generates novelty rather than reproduces the old. It puts things together rather than takes them apart. It is more like multiplication than division. It implies temporariness, not eternity (the re-part of it).(13)

The choice of words deserves some comment. I use creation of knowledge here. In a piece almost 10 years ago, I and Dag Rolander posed exploitation against experimentation, inspired by Prigogine's analysis of ant colonies. Jim March has proposed exploitation vs exploration. Both experimentation and exploration are terms of search modality, whereas creation, like exploitation refers to the purpose of activity. Thus, symmetry is better upheld.(14)

Why then recomposable? There are three reasons to prefer it to concepts deriving from combination or integration. First, it rhymes well with decomposable, and it relates to and extends the logic of decomposition. Second, it implies the temporary and ever-changing nature of connecting parts of knowledge. Third, it is more active than words such as combining and integrating, and it implies a sort of activity rather than just hope for good consequences. Considering frameworks in terms of "dynamic capability" by Teece and colleagues, integrative capability by Henderson, or combinative capability by Kogut and Zander, one is struck by their emphasis on path dependency, stability and coherence of firms.(15) For example, Kogut and Zander, building on Nelson and Winter, emphasize inertial routines and the firm as a repository of knowledge. In spite of the call for combination, the focus is on explaining the historical path rather than starting conditions for getting out of the beaten track.(16) The notion of a repository of knowledge is, in my view, too static to capture the creative efforts better intimated by the term nearly recomposable system.

A musical analogy may be enlightening. Composition is something quite different from combination. Beethoven's ninth is not a combination of 7 and 8! There are computer programs producing Bach-like fugues, through combination of common figures and chord sequences. However, they all sound hollow. Good examples of recomposition are found in variations, such as Bach's Das Musikalische Opfer, Beethoven's Diabelli variations, and Schubert's charming Wiener Damen Landler. Also Jimmy Hendrix' rendition of the US national anthem and much jazz improvization is really recomposition, is this last case in a social context with coordination between individual members. An alternative to recomposition might have been reconstellation, "putting stars together in new ways", focusing on the quality of individual members and their inputs rather than the activity. Much creative artistic work builds on creative reconstellation, in this social sense.

What, then, are the likely properties of nearly recomposable systems (NRS)?

Requirements for Nearly Recomposable Systems (NRS)

I will try to delineate the properties of social NRS, i.e., systems containing human beings in interaction. Also the characteristics of the technology and machinery involved are of importance, but the focus is on the users and producers of knowledge. Occasionally, non-social systems will be mentioned for purposes of clarification and comparison.

Of the twelve desirable properties to be discussed, two - the first and the last - pertain to the structure of the entire knowledge base and the basic posture adopted in utilizing it, respectively. Four (No. 2-5) involve aspects of internal communication; four (No. 6-9) have to do with organizational intention and motivation and their consequences and roots in organizational systems; and, two (No. 10-11) relate primarily to the system's sensitivity to the structuring of knowledge chunks within as well as between organizations.(17)

Productive Diversity

In order to recompose, there has to be something to work with in the first place. The system has to contain enough diversity for search for (re)composition to be profitable. The global system of McDonald franchises is not a likely NRS because of the imposed standardization. The diversity is in terms of knowledge and packages of knowledge. Acquisition-driven globalization processes are likely to generate (sometimes more than) enough diversity, whereas organic growth may be too conservative in this sense. The principle of productive diversity implies systems strategies that extend the knowledge base beyond what is currently perceived as "rational".(18)

Common General Language

For knowledge to be exchanged and combined, there has to be a shared medium of communication. People have to be able to make sense to and of each other. 19 One aspect of this is a shared spoken general language, like Hindi, Bantu, or English. In history, strong international institutions have always fought to uphold a shared language. When the Reformation successfully challenged Catholicism and its institutions, one main point was to insist on the giving of sermons in the local language. This contributed to the decline of Latin as the accepted medium of religious as well as secular communication, and it paved the way for the ascendancy of the European style nation state as the dominant organizing principle of the modern world. Other examples abound, from the secret languages of gypsies and other travelling peoples to the dominance of French in diplomacy.

Common Domain - specific Language

The tower of Babel is not the only problem, however. For recomposability, there is also a need for a shared domain-specific language, with its own vocabulary, syntax and semantics. A very open and flexible NRS is the (or an) alphabet. However, not all letter combinations are allowed in a natural language, and still fewer in a specialized, professional field. You only have to consult even a popular journal on, for example, personal computing to appreciate the degree of specialization of vocabulary and semantics (and often very implicit and mysterious syntax).

The principle of shared domain-specific language has a very important consequence: An NRS must be relatively narrow. Espacing this limitation requires well-defined and widely accepted inter-domain interfaces. This is why the digitalization of most everything provides expansion room for the NRS. The merging of computer, telecommunications, and radio technology through common strategies of digitalization gives new opportunities for combinations not even contemplated only a few years ago. So far, pharmaceutical innovation has been largely a matter of "the molecular roulette", knowledge in one field or about one substance not giving much leeway in and about others. The development in genetical engineering ("re-combinant DNA"...) may move biologically based business closer to the situation in the IT industry.(20)

A Flexible Communications Infrastructure

A shared language does not guarantee communication. Particularly when new compositions are needed, there have to exist platforms where different parts of the knowledge base meet. Usually, a social system has designed its lines of communication according to the dictates of running day-to-day business. Encouraging the unexpected requires investments in "unnecessary" communications platforms and links, and great if not total openness, giving everybody access to the total knowledge of the organization.(21) The fact that much individual as well as organizational knowledge is tacit requires interaction to be intensive and extended in time for demerging, merging and remerging of knowledge elements to be possible.(22) Also more explicit knowledge may need to be "massaged" and adapted to fit together with other knowledge.

Modern information technology has done much, and will do more in the future, to extend and enrich possibilities for global (re)combination of knowledge. However, also the composition of corporate and subsidiary boards, travel schedules, international rotation, executive and technical training programs, and, mostly for intranational communication, physical architecture have important roles to play.(23) For international chance as well as regular encounters, perhaps MNCs should have their own offices at main airports. An interesting experiment is under way at Samsung, which is building a huge ocean-going ship, where much of their management training and development will take place.

Know-who Even if recomposition partly relies on random encounters, the main part will have to take place on the initiative at one node or other searching for complementary knowledge in the entire system. Particularly if the sought-after knowledge is likely to contain large doses of tacitness or complexity, the most efficient storage as well as retrieval mode is probably through single individuals. Therefore, an NRS library of information is perhaps primarily a library of people and their competencies.

More important in the long run is the richness of the personal networks of acquaintances and friends that are built up through constant efforts to utilize the potential of the NRS. The intimacy required for transfer and transformation of knowledge favours friendship rather than acquaintance as the desirable depth of relationship. In the bureaucracy, you are just colleagues, indeed, personal closeness has traditionally been seen as a weakness likely to corrupt the efficient machine. Know-who also implies a relatively constant total pool of people. Long tenure in organizations, at least for many core people and not only at the top, is desirable. The disadvantage is, obviously, the danger of insularity and group-think.(24) The NRS may compensate for this through very intensive and open exchange with the interorganizational domain.(25)

Strong Knowledge Vision

An NRS may turn into a hotchpotch of frantic and inefficient experimentation unless there are search rules guiding the efforts to recompose. In management theory as well as economics, there is a strong assumption, well corroborated in empirical research as well as practice, that organizational action is - on the one hand - rather chaotic, ambiguous and fractionated and - on the other - that it is subject to identifiable stable patterns, easy to observe but difficult to change. (Search for solutions in the neighbourhood of problems, sequential attention to goals, the tyranny by the committed, quasi-resolution of conflict, etc., to take examples from the most influential organization theory source(26).) The paradox for an NRS is to both encourage and channel the diversity and sometimes chaos.

This is pretty much accepted in popular as well as scholarly management texts today, although not always with the focus on knowledge and its use and creation.(27) Calls for shared visions, missions, goals, strategies, etc. are loudly proclaimed. A full discussion of what kind of such guiding concepts are appropriate cannot be undertaken here.(28)

Shared Identity

In order not to disintegrate in the commotion and perhaps chaos of seeking and managing new opportunities, an NRS needs to instil a strong feeling of identity with the organization and its members. This keeps together ("we're in the same boat") as well as limits search ("we don't work with just anybody"). It encourages people to stay even if and when they find potentially more lucrative possibilities outside, particularly together with internal colleagues and friends. It also gives psychological security through the common ground established for the ever changing recompositional efforts, often implicit and changing roles, organizational positions, geographical location, reporting patterns, etc.(29)

Rewards for New Compositions A potentially recomposable system may never recompose unless there are mechanisms countering the strong tendencies to continue in old ways. The difficulty of relying totally on selection by the market - because of long time lags, non-linear effects, the importance of active creation of markets, etc. - forces an NRS to reward also for internal completion of projects or other combinatorial efforts. Rewards may have to do with pecuniary gain, but often more importantly with status, prestige, access to resources, and good turnover between temporary tasks.(30) The usual dominance by the "line" organization must be tempered for composition to take place.

Cross-border Communication

This is implied in the above but deserves special mentioning. In the old type of M-form firm, as well as in the theories of it, the aim is to restrict communication horizontally (through divisionalization, or, in the U-form, functional organization) as well as vertically (through the top leadership taking a hands-off strategic role). An NRS cannot trust such borders to optimally circumscribe beneficial areas for communication. The principle should be open access in all directions. The image of an intelligent corporation should be "the firm as a brain" rather than "the brain of the firm", where thinking goes on only at the top. Even the mere notion of "levels" becomes problematic in an NRS. Action recomposes structures and roles incessantly, and freezing the system of such processes at an instant in time may not be possible, or desirable.(31)

Context Robustness

Efforts to combine competencies often fail because each part is too sensitively grounded in an original context to be transplanted and married into a new one. For example, when the model of venture capital companies to successful - at least in a total system sense, most of them may have gone bankrupt - in the USA was introduced in Europe, in most cases there was great disappointment. Apparently, the formula is strongly context-bound. Similarly, something as mundane as the habits of washing clothes in Europe and the USA provide different contexts that severely inhibit relatively simple - from the technical point of view - transfer of knowledge.(32)

Context rubustness is helped by well defined global standards and intra-company technical interfaces. Modularization as a solution to balancing simultaneous demands for standardization and adaptation is often called for.(33) This meta-style response only touches the surface of the problems, however. More serious is probably the context-boundedness of people. This, in turn, affects the ability to communicate knowledge and to use one's expertise in a new context. One obvious remedy is rotation schemes of various kinds: geographical, functional, between roles, etc.

Openness to the Outside

The concentration and limitation required by the need for a shared domain and language force the NRS to rely on external suppliers for missing pieces in a recomposition. This paradox between focus and external openness is probably one of the most difficult in a managerial sense. Particularly the demands for modification of old components often necessary in innovation force collaboration to an intimacy similar to that of successful intraorganizational relations. The border of an NRS is thus soft and malleable. Issues of secrecy, cheating, shirking, and property right protection appear, but also possibilities to extend the organizational family through "adoption", temporary job swap schemes, taking the supplier physically into one's premises, taking the customer into the supplier's centre, building shared incentive systems, instilling a project - rather than organizational - identity, etc.

Experimentation as Search Modality

A number of methods may be used to generate ideas for recomposition. Some of them have been extensively used in technological forecasting, like matrices of technologies, where each technology's possible synergy with each of the others is assessed. Ideas may also be derived from vision-inspired broad master plans.(34) Still others may come from individual hunches or chance meetings with colleagues, or drinking with friends. Organized creativity-enhancing processes also have roles to play.

However, the mere identification of such opportunities is easy in comparison with their implementation. Given the uncertainty and confusion surrounding all novelty, all NRS must be prepared for an experimental approach, trying many small things rather than betting on one big project. Experimentation suggests both trying the unlikely, evaluating its results, and drawing conclusions for a broader set of issues. Thus, an NRS cannot just let the creative forces loose and hope for the best, but must all the time diligently learn from success as well as failure.(35)

What Real Social Systems Look Like NRS?

In my examples, I have mostly alluded to corporations. However, the list is meant to be more general, suggesting the properties of any social NRS. What social systems might then fulfil the criteria? (I will return to the question of firms, and particularly MNCs, later.)

Specialized Academic Communities

A perhaps surprising candidate is the specialized global profession built on academic community. In, for example, a medical speciality such as brain surgery, you find a very articulated and shared language, with Latin terms often compensating for the lack of universality of English, the accepted medium of communication in the main journals, at conferences, etc. You also find shared visions (although there is much competition for the interpretations of manifest destiny and its related technologies!), and a very strong feeling of community. Rewards are severely biased in terms of innovation. There are dedicated mechanisms for information sharing: conferences, journals, limited sejours in other countries, international research cooperation, etc. The circle of top people know or know of each other globally.

Medical technology works on the human being, who provides a relatively constant context across locations. Experimentation is the accepted way of judging what works, and large risks are not shunned, since there are many opportunities where the absence of action leads to certain undesirable consequences.

On the weak side, the openness to the outside is limited in many academic disciplines. Even in medicine, where the suffering patient serves to focus everybody's attention, utilization of knowledge between specialists is often shallow. Partly, this is because communication codes are different and sheer factual knowledge inadequate. Sometimes, it has to do with the difficulty of associating with people of a different identity.

Purer academic communities, lacking the possibility of constant experimentation, for example in the humanities or social sciences, are usually relatively distant from a challenging external market, weeding out bad inventions and rewarding good ones. On balance, however, relatively specialized academic communities meet most of the criteria for an NRS. Whether a university does is another matter. A university should probably be seen as the relatively stable and hierarchical positional structure harbouring a plethora of action initiatives undertaken against the background of a global knowledge system. At any rate, academic life and its institutions belong to the most long-lived and successful institutions we know, their job has been the generation of new knowledge, and they have mostly been very international in orientation. It should be no surprise that they look like NRS. Of interest is also that many of the principles of heterarchical, N-form organization are exemplified in the management of academic communities: top leaders are usually appointed by faculty; deanship is seen as a burden that should be rotated; real work is organized in projects; positional ascendancy is more a reward than a license to tell others what to do; symbols, ceremonies, and stories play a large part in upholding the identity; sabbaticals serve to eradicate old knowledge and force new encounters, etc., etc.

Professional Firms

Consulting and law firms, and other associations of experts with specialized yet complementary knowledge evidence many of the same strong points (for NRS status) as academic communities. They may be, at least potentially, even stronger on common goals, rewards, communicative infrastructure, and, particularly, openness to an outside market. The main problems may lie in exploitation of present strengths driving out recomposition, and in the perhaps weak loyalty with the firm, rather than with the profession. (Note the turbulence of manpower turnover in management consulting, investment banks, etc.)

Specialized Geographical Regions

Somewhat puzzling, given that phenomena such as Silicon Valley firms were part of the high intensity/low extensity quadrant of Figure 1, such regions appear to meet most of the criteria for an NRS. The distinction is between one firm in the region (not NRS, at least not in the sense of utilizing knowledge sources globally) and the region itself (NRS). Research on Silicon Valley as well as less known and fashionable industrial clusters demonstrate the presence of the communications and structural attributes I have enumerated: diversity, shared language and knowledge domain, dense communications networks, a great deal of know-who, sharing across boundaries (sometimes involuntarily!).(36) The weak points have to do with the identity items: there is rarely a clear joint view as to where the system should head; feelings of belonging is more with the part (firm, organization) than with the totality; rewards are not in line with systems development. Also structurally, there are a few problems: context robustness is hampered by the diversity of knowledge bases and strategies; and, most importantly, the geographical openness to the outside may be low, also in terms of novel "ways of doing things". If the region can become a magnet, pulling in important peripheral elements, the problem is alleviated, but not solved. There will be a natural tendency for Silicon Valley to focus on US market needs rather than Japanese, and the attraction of partners, lead customers, etc., will be thus affected.

Nevertheless, the specialized economic region is an important class of institution, with a specific role in (re)combining knowledge, and with an ambience encouraging rapid experimentation. A problem is that such regions take very long to grow, most planned ones fail, and they are probably most useful for areas where the domain is rather narrowly defined in terms of science and technology, and where the globalization of competence and competition has not already fragmented opportunities.

The absence of a strong central actor is the hallmark of both academic communities and regions. The concoction becomes something like reproduction in a fertile ecology, with much intra-species activity producing better quality and quantity. A few "lucky monsters" will also come out of forays and mutations with, primarily, outsiders to the system. With a central actor, a more engineered approach is possible, and learning can be systematic at the level of the individual organization, rather than rather unsystematic and at the level of the total system.

Networks of Dedicated Individuals (NIDs)

The origin of many first-wave IT firms owes a lot to the almost incredible entrepreneurship, energy, and vision of single individuals. I suggest that we already see, and will continue to see, the formation - mostly unconsciously and informally - of loose networks of individuals who together will compose NRS. The first wave was in terms of "nerds", the second is in terms of "NIDs", networks of individuals dedicated to a common vision, and to each other. This is not only, or even primarily, a matter of CEOs and the already successful. The bright, aspiring - and perhaps bored - people at lower levels within organizations have the most to gain. Whether they break loose to form their own organization or not may not be that important. They still can produce much together, using the collective strengths of their "homes". The relation to the "mother organizations" may be parasitic or symbiotic.

This kind of global coterie scores high (for NRS status) on most criteria. The problem is, rather, the lack of administrative and coordinative functions. Where the step from idea to "product" is close, and does not involve much investment, it can work (a new consulting approach, a new journal, a trading system). When scale-up problems are large, a NID is not a good or likely vehicle.

A special kind of NID consists of people in the arts, culture, and media. It is obvious how, in the past as well as today, cultural pages in newspapers, the assignment of Nobel prizes in literature, opera and theatre production, and film making are dominated by clearly but also tacitly circumscribed circles of colleagues. Often, such circles are global, although they may have started in a 17th century salon manner. The apparent success in terms of creativity, as well as the equally apparent risk of crowding out talent and reinforcing established styles, hold lessons for NRS contenders. The main weakness of the global artistic circle as an NRS have to do with openness to the outside, rewards, and lack of common vision. To insist on the latter is mostly explicitly considered a simplistic insult to the members, anyway.

Global Network Firms

A number of conceptions of a new kind of firm, relying to a large extent on a network of contacts and relations rather than either arms-length contracting or internalization, have been proposed recently. Without discussing attribution and nuance, I will just mention terms such as: the virtual organization, the hollow corporation, spider-web structure, fishnet organization, shamrock organization. There is also a strong Scandinavian tradition of studying buyer-seller interaction and corporate networks as the appropriate unit of analysis.(37)

Firms with extended and intensive global networking capability may be flexible NRS. They avoid some of the problems that the imposition of a large bureaucracy burdens the traditional MNC with. Very small and yet globally connected firms - "midget MNCs" - start populating the economic landscape.(38) It may be a matter of less than a dozen core people, with almost all specific and relatively routine activities - such as production, marketing, accounting - subcontracted around the world. At the center, which may also be geographically dispersed, key knowledge components such as design, technology, and market image and imaginization are kept. Trading networks resemble global network firms, although the focus often has been on arbitrage rather than creation.

Global network firms score highly - for NRS status - on diversity, know-who, cross-border communication, openness, and, perhaps, experimentation. The problems lie in lack of domain specificity - given the often functional division of labour between actors - knowledge vision, rewards, and context robustness.

Diasporic Networks

Networks of individuals and organizations built on, usually, ethnic ties, exhibit many of the NRS-supporting properties: Common language, identity and communications items, are strong points. On the weak side, we find the lack of knowledge vision or of a natural focus on knowledge at all. Most such networks - Phoenician, Jewish, Armenian, Hansa, Overseas, Chinese, etc. - have focused on trade and arbitrage rather than on knowledge assimilation and transfer, and on innovation.

Technological Waves and Paradigms

Sofar, NRS have consisted of organizations and people. However, you may also define an NRS as a process. Two such dynamic unfoldments seem of interest. The first is the emergence of related technological fields, as when electrification changed most every industry and way of living, or when standards for mechanical industry and work developed, so that an entrepreneur could "simply" set up a mechanical shop and be hopeful that there would be many possibilities to combine his skills and equipment with others - customers, suppliers, collaborating firms, and individuals sharing a similar background. Today, digitalization may offer the same prospect. The frantic deal-making and cooperation agreements, at a global level surpassing the dynamics of economic regions, indicates that there is much recomposition going on in the IT field, broadly defined. However, also in narrower areas of technology or demand, correlated micro-processes serve to create global systems of opportunity, with identifiable existing actors and niches for new ones. Visual image processing, high fashion as well as sportswear, and adventure tourism constitute examples.(39)

A large body of literature in economic and technology history documents the focusing effect of technological waves at the grand scale.(40) Considering such processes, one is struck by their conformance to many of the NRS traits. They create a common domain-relevant language in their making. There are often visions of where the whole field is going, although also competition about such destiny. There is a strong social aspect, with increasingly global communication patterns. Initial diversity is almost guaranteed, and an experimental atmosphere purveys. Reliance on an openness to the outside is necessary. The weak points have to do with rewards for individual actors, genuine sharing because of trust-building problems, and context robustness, both human and technical. In history, the latter has often lead to competing standards, effectively sealing off groups of actors from each other, precluding any recombination of competence and burdening standard-spanning activities with costly duplication.

National Mobilizing Projects

The building of a strong national economic base is the second type of "process NRS". The history of countries as unlike as Bismarck Germany, Sweden and Japan illustrates how ambitious visions of industrial catch-up catalyze and almost organically coordinate efforts by many actors. Sometimes, this function is undertaken primarily by the state, sometimes mesolevel industrial groups or capitalist empires provide the impetus and ideas. The projects range from relatively technical, such as building infrastructure, to very political and cultural, such as war and the creation of ascendancy in terms of economy and global leadership.(41) It is remarkable how fragmented systems are able to come together and combine their competencies when the need is strongly felt. And, even if the immediate objectives are narrow, there are good chances of spin-offs that provide NRS capability for the long haul.(42)

Mobilizing projects often build much of their force on almost demagogic use of common language, identification of critical domains of excellence, and the establishment of novel platforms and modes of communication. The nazi era indicates the potential, and the thought of Goebbels with IT and Internet to complement UFA, radio, and Nurnberg makes one shudder. Of the communications items, know-who is necessarily limited in the grander projects, but at the elite level it may be very developed. The way the "best and brightest" in the USA in the 1950s and 1960s moved between sectors in society seems to be an example, as does MITI's history in Japan.(43) Also on motivational items, national mobilization is very strong, except perhaps for shorter term goals, where sacrifice is called for.

Economic history shows that learning from the outside and creative combinations with internal factors play a large role in development. Often, there have been ambitious programs of sending young people to the best seats of learning as well as to places where administrative innovation, or at least perceived excellence, is located.(44) Experimentation is also strongly in evidence.

The weaknesses of mobilizing projects as NRS have to do with structural characteristics. Context robustness is low, since almost by definition there are many unique components to consider. Different parts of the composition do not easily fit together. Intensive cross-border sharing of knowledge does not come as naturally as in the systems that have already built mechanisms for the purpose of communication, albeit of a kind initially lending themselves to the exploitation rather than the creation of knowledge. Of course, there is no necessity of projects being national. Local, regional, and global ones can also appear as NRS candidates. Indeed, the lack of global sharing and combination of knowledge bases is one of the more pressing problems in addressing questions of poverty environment, and medicine. The challenges here are beyond the scope of this paper. Probably, only real or imminent global disaster can provoke the "moral equivalent of war", shared interpretation of issues, and structures of reward necessary for real utilization of the world's knowledge.

The introduction of processes as systems, with part-processes as well as structured forms as elements, may seem puzzling. An alternative conceptual strategy may be to see processes as enabling/helping conditions at, mostly, the macro level, and reserve NRS status or the more traditional social systems discussed earlier. The combination of the two would be the most effective for recomposition. Perhaps one reason for the long era of domination for the idea of decomposability has been the lack of inspiring megavisions. The digital future and cyberspace is perhaps a candidate, as might be the formation of regional economic blocs, such as the EU and ASEAN. The latter category resembles traditional instincts to address issues at a higher and higher level, coordinating and standardizing rather than rearranging, however.

The MNC as an NRS

The initial discussion has already prefigured some reflections on the MNC, and I have made reference to the kind of new entity baptized heterarchical, transnational, horizontal, multifocus, differentiated network, etc.(45) In the ensuing, this is what I mean by an "MNC". The analysis is furthermore biased to firms from the West (Europe, North America), with long and extensive international experience, that have grown at least partly by acquisitions, that involve much foreign direct investment rather than only exports, that have knowledge bases spread globally, and that do not base their claim to fame primarily on conglomeration or buying and selling assets.

This does not mean that it would be uninteresting to consider non-Western, small, young, etc. organizations. On the contrary, organizational innovation is as or more likely to come from such quarters, particularly as many of the weak points (for NRS status) of the MNC have to do with historical inertia and embeddedness, and an orientation that deemphasizes novelty and the necessity of experimentation, often for good reasons. I will return to this issue later in the paper.

A second limitation is that multinationality may be less of a factor than the nature of the activities the MNC is or gets involved in. It is clear that recomposition is more feasible in the IT industry than in, say, cement production. It is mostly more appropriate in highly knowledge intensive than in more mundane businesses. The need for rapid reconstellations may be greater or younger, rapidly developing arenas. Some areas of need and demand are more global than others, etc. One can identify at least four areas and types of MNCs that seem rich in recomposition opportunities:

1. The whole IT/electronics/digitalization/communications gamut. Firms such as Philips, NTT, NEC, Motorola, Ericsson, Nokia, Nintendo, CNN, Intel belong here. Possibly, and in the future probably, the pharmaceutical and life-science based fields will show the same picture. 2. More traditional systemic infrastructure providers in electrical engineering, building and construction, and mass transporation. Here, we find ABB, General Electric, Hitachi, Bechtel.

3. Complex assembly, with larger and stronger doses of intangibility and intelligent components. The automobile industry is a prime example, but also aerospace and much defence industry. Boeing, GM, Rockwell, Daimler Benz, Komatsu are representative, although some of the industries of such firms have historically not been very globally conceived, configured, coordinated or oriented.

4. Services with a high information/knowledge content, or with IT as an important factor of production. Here, we may deviate from the logic of discussing the high KI/high KE quadrant in Figure 1. However, firms such as Anderson Consultants and McKinsey, Lehman Brothers, Federal Express, Benetton and Progressive Insurance are potential NRS.

I will, in discussing the NRS status of MNCs below, concentrate on categories 2 and 3, because they are more problematic and therefore provide a harder "test" of the ideas. Furthermore, there is more research on their international activities and organizational experiments in general. Finally, I know them better empirically myself.

My own assessment, which will have to - at this stage - go without references to research on even explicit reasoning, is summarized in Table 1. Everything should be seen as an illustration of a research program and hypotheses to be discussed and possibly refuted. I have selected to focus on four main questions, covered in the four columns of the table, respectively.

1. What are the MNCs main strengths in the particular trait hypothesized to contribute to recompositional ability?

2. What are the corresponding weaknesses?

3. What is the overall assessment of the balance between strengths and weaknesses?

4. What are the possibilities for action that can be undertaken quickly and that is likely to produce improvement rapidly?

The hypotheses in Table 1 give rise to four more general conclusions. First, the MNC as an NRS has "hard" advantage and "soft" disadvantages. The former have to do with the almost definitional diversity at its disposal, its targeted competence base and associated domain-specific language, and its communications infrastructure, particularly the "wired", hard parts. The drawbacks - at least empirically, to judge from the research there is - pertain mostly to questions of identity and attitude. Few MNCs can boast a coherent and shared identity and sense of belonging among the employees globally, or a forceful vision for knowledge and its development. Openness to the outside as well as an experimental attitude are hampered by the historical weight of successfully exploiting old advantages.(46) However, there are also "harder" obstacles, generated primarily by structural rigidities at the firm as well as the industry or larger levels. To the former belong the often pathological reward systems - if you want international recomposition - and lack of sharing of knowledge across organizational boundaries. At the meso and macro levels, the fact of thousands of languages being spoken in the world and of differing standards prohibiting shifts between contexts are formidable barriers.

The distinction hard/soft is not cast in stone, or so tightly woven into the fabric of managerial reality, however. Many of the "soft" weaknesses originate in hard-wired organizational systems and routines that over time build culture-like edifices and are seen in forms such as weak identity, bad attitude, and institutionalized [TABULAR DATA FOR TABLE 1 OMITTED] refusal to cooperate. The various NRS components constitute an interactive system, rather than just a list of nice things to have. In extreme cases, the lack of one of the attributes may block any reconstellation, and whether it is the hard that causes the soft to flounder or vice versa is not the essential point.

Second, the overall picture looks bleak, but there is a large potential for quick improvements. In eight of the twelve dimensions, the MNC scores low, to judge from empirical research and, admittedly, much conjecture and averaging over a very multi-faceted corporate landscape. Concerning potential for rapid - in the sense of initiation as well as expectation of results - action to upgrade NRS capability, matters look more promising. Seven of the twelve aspects seem amendable to, if not quick fixes, at least short term and substantial amelioration.(47) It is mostly a matter of leadership. Unsurprisingly, it is the harder items that are most influenceable. Reward systems can, technically, be changed over night. Interaction across organizational boundaries takes longer, but utilization of the strengths in communication and know-who would help, as would a general demobilization of the arms at the disposal of the positional organizational hierarchy. The really tough anticompositional nuisances to address and reform correspond to those that are serious obstacles in the first place: the spoken, natural language; the lack of sense of belonging; closure to the outside world; a proclivity for stalling experimentation; and, the limitations posed by barriers between contexts. For the latter, the structural aspects - standards, user habits, and the irritating inertia of atoms compared to bits - are less influenceable than the human ones, for an MNC in isolation. Overall, prospects for improving (re)composing competence seem positive, but attitudinal problems remain, and the situation varies much between areas. ABB may not be as bound by context disparity as Bechtel.

Third, there seems to be a genuine choice of role for the MNC. One alternative is to close most gates of synergistic search and concentrate on the efficient utilization of given, traditional strengths. This could be complemented by determined programs to identify and spread best practice around the units and individuals in the global system. There is a good case to be made for the MNC as a "hygiene factor" in the global economy, shaping up standards wherever it descends, in the process perhaps also forcing local competitors to do the same. In an interesting international research project on the Power Transformer business area in ABB, a strong impression is that this kind of efficiency increasing, standardizing, modernizing influence from the traditional cores of the firm, supported by determined program management by business area leaders, produces great effects in rather short time.(48) Efforts to integrate rather than standardize, increase diversity rather than stamp it out, and innovate are less in evidence, both in terms of intention and consequence in this particular case.

Thus, the MNC, or parts of it, can select to become an efficient "global machine", leaving drastic renovation to other actors. In diversified firms, some areas will probably not be ripe, or too ripe, for global recomposition, whereas others will not. The trend to decompose large firms into semi-independent parts helps in this respect. The recent split-up of ATT has many motives. One consequence is that the global dynamics of intra- as well as interfirm collaboration for knowledge development can proceed with the parts' differing requirements and potentials more clearly in focus. For example, "strategic alliances" at the grand, diversified scale suffer from being too weak for genuine knowledge transfer and synthesis in some areas, and uselessly strong or even detrimental in others.(49)

The alternative to the vision of the MNC as a Global Machine Corporation (50) is to see the MNC's specific role, today and even more in the future, as exactly the institution able to add value by integrating globally dispersed and combinable pools of knowledge, becoming a sort of Recomposing Agent Firm.(51) Creation involves a spectrum of activities from "simple" transfer of competence and knowledge, over their standardization, adaptation, improvement, expansion, synthesis, novel creation, and dissemination.(52) It can take place through systematic programs or more experimentally, probably in conjunction. Novelty in the socio-economic world can be initiated and diffused in creationist fashion as well as through blind selection, artificial breeding and adaptation.(53) Intelligent utilization of ali four modes seems quite possible in an MNC context.

Fourth, the analysis in Table 1, if it is correct and relevant, suggests a strategy of change for improving the MNC's total and global utilization of its knowledge. In the most influential recent work on MNCs, there is a distinction between the anatomy (for example, organization structure), physiology (for example, flows of financial data), and psychology (for example, culture of cooperation) of the corporation.(54) To move towards the transnational ideal - for many firms, it depends on industry and a host of other factors - it is recommended that one does not start, as many real firms do, with anatomy, for example by fiddling with the formal structure and command chains.(55) Instead, one should start with psychological (affect culture) and physiological (build interdependencies through joint tasks) matters.

Whatever the merit of the claims for transnational change strategy, the one most appropriate for NRS capability building seems related, but a bit different. The following should be seen as a systemic hypothesis of such a strategy.

1. Start initiatives on all dimensions simultaneously. The systemic whole may be corrupted by just one disjointed element. If the problems is long-term, so much greater the reason to start early. This also shows conviction and dispels doubts. There have been too many management fashions floating around not to have inspired justified cynicism by those subjected to them.(56)

2. Expend most resources - financial, management time, human energy, etc. - on and make clear that the initial focus is on three aspects:

* To realign incentives and motivation structures for individuals and organizational units. If you are paid for global knowledge transfer, you may attempt it, and such schemes can be rapidly put in place, given will from top management.

* To invest aggressively in communications infrastructure. Intra- and internets are parts of the picture, and particularly training in using such opportunities.(57) Also executive and other education should become much more global, strengthening know-who and providing conditions for joint creation rather than only transfer, and for the sharing of tacit or idiosyncratically structured knowledge and skills rather than just those that go easily down optical fibres.

* Urgently set about the formulation of a knowledge vision for the firm. This channels the power of incentives and communications platforms in a coherent direction. Catalogizing, structuring, retrieving and storing - as well as stating ambitions for assimilation, transfer, transformation, creation and destruction of knowledge and its modes of dissemination to the market- should be part of the vision. Its development has to be thoroughly participative and continuous, and generate also operational, short-term goals.

Considering the three items - incentives, communication and providing direction - one is immediately reminded of Barnard's(58) three executive functions. The difference lies in the focus on knowledge and its (re)composition, and on generating novelty rather than administering the given.

3. Do not reorganize in the short term, unless it is absolutely necessary. The exception to this rule is when there are clearly distinct knowledge management requirement in different parts of the MNC. Large-scale reorganization of the formal reporting structure creates havoc for particularly the work of ironing out specific and targeted knowledge development vision, strategies, and goals.

4. Deal with the attitudinal problems by exploiting pockets of enthusiasm and initiative.(59) For example, highlight cases where two subsidiaries manage to produce something new, by top management support and recognition, and tangible and visible rewards. Support includes money. There are infrastructural ("intra-firm external economies", to bastardize economics) aspects that justify some subsidy. Mostly, however, collaborative efforts go un- or underfunded anyway, because of the power of the positional, line organization.

5. To increase context robustness, initiate and engage in field-wide ("industry", technical field, user domain, etc.) efforts to establish standards and open interfaces, including shared vocabulary. This amounts to intellectual leadership, and initiators have more to gain than laggards.(60)

In Bartlett's and Ghoshal's terms, the change strategy amounts to leaving the psychology to rather indirect and long-term action, and start with a fourth system, which we may call the intentional one.(61) It includes general direction as well as short-term goals and individual, pecuniary rewards. Simultaneously, a particular part of the physiology - the central nervous system of the firm - is influenced, at the technical as well as human level. We should not be surprised, I believe, that metaphors in terms of nerves and brains are appropriate for a knowledge processing system.


Three-Dimensional N-form Heterarchy. an Arithmetic Interpretation of Chandler, Recomposability as a Systemic Property. and the Institutional Status of NRS

The analysis leads to four more general conclusions, or rather beginnings of discussions. They pertain to, in turn: the understanding of organization structure, and the biases various conceptions give rise to; the trends in corporate development and associated management models over the last century or so; the necessity to consider recompositional ability at a systems level, in order to understand firms' location choices as well as national competitiveness and public policy; and, a comparison between N-, M and U-forms and markets in terms of de- and recomposability. Necessarily, the exposition is sketchy, and the ambition is to suggest fruitful approaches rather than apply them in detail. The emphasis is on the first question, since this relates most naturally to the discussion of the MNC.

The Hierarchical Malaise and the N-form Heterarchical Alternative

Most of the drawbacks of the MNC, or any large corporation, as an NRS originate in its fundamentally hierarchical design, all useful experimentation with "boundarilessness", "virtuality", "hypertexteralism", "networking", "lowerarchy", or "heterarchy" notwithstanding.(62) Empirical corporate reality is mostly solidly bound to notions of hierarchy, many of which unconsciously. A useful starting point is with the accepted originator of the concept, Dionysius the (pseudo) Areopagite.(63) Among a dozen or so telling and, for organizations, still largely relevant properties of the structure in heaven, there is one remarkable insight: "The divine hierarchy is an order not only of beings but also of knowledge and action." The doctrines of divine omniscence and omnipotence are directly implicated here. The analogy - or, I would claim, adoption in milder form - in corporations is a conjunction of three systems or structures: a Positional one of deep (in heaven, there are nine levels) super- and subordination; an Action one structured in terms of overarching strategy, deduced tactics, and further deduced implementation and discrete acts; and, a Knowledge one, where synthetic and encompassing notions and language as well as competencies rule over increasingly narrow, specialized terms, and where people lower down are not privy to scripture through secrecy and "need-to-know" policies. Admittedly, this is somewhat of a caricature of real firms, but for those not at the top of them a quite recognizable reality. At the top: highly paid general managers with strategies, encompassing knowledge and open access to all files. In the middle: moderately paid people with technical and tactical roles and specialized, restricted knowledge. At the bottom: lousily paid people with tasks of implementation and sometimes almost no need to know much. The point is the symmetry and overlap of those systems. They are all shaped and thought of as pyramids, and the strength of the structure is the clarity that goes with it. The overlap is illustrated in the way of talking about competence, a central notion in the Weberian bureaucracy, which is a modern restatement of many of Dionysius' principles. At least in Germanic languages, when somebody in a government office says that something "is not within my area of competence", it means "I don't know" as well as "I am not in a position to comment" and "I cannot do anything about it".

Hierarchy in this strong sense is beautifully adapted to a world where decomposibility is the adaptive strategy.(64) However, it is antithetical to recomposability. An NRS needs to be flat to speed up cross-border communication - decomposition leads to increasing depth almost by definition. An NRS needs horizontality - hierarchy almost defines verticality. An NRS works with temporary projects and constellations - hierarchies are supposed to provide stability. An NRS needs circularity and recursiveness - hierarchy is highly non-recursive.(65) You don't talk back to God.

The main point, and implicit in the above, is that recomposition requires the disjunction of the three structures. Knowledge chunks catalogized in a certain way in the positional structure and involved in certain action packages need to be reallocated, relocated, and realigned with others, and the three dimensions cannot be expected to coincide.(66) This is not the only reason for the disjunction of organizational dimensions, but its importance for building NRS capability is paramount.

Considering the MNC's weaknesses as an NRS, it is striking how many of them have to do with the basic, hierarchical structure, in the strong sense here defined. The dominance of the exploitation-oriented positional line organization over those with special competence and over experimental action projects is one example. The scarcity of strong statements about knowledge visions and goals in another. General business goals are considered enough, and "vision" is a good thing to have, but it is expressed in terms that do not necessarily concern knowledge management at all. The still primarily vertical use of communications systems and reporting systems testify to the tendency for the potential for new models - in this case in favour of horizontality and recursiveness - to be channeled into the old channels and ditches.(67)

Thus, the disjunction of dimensions is the basic problem for the MNC as an NRS. A full exposition of how this happens, how it can be avoided, and a further specification of the three-dimensional model is beyond the scope of this paper.(68,69) The important point is that NRS process requires a new understanding of organizational structure. It is at least three-dimensional, and the interaction between the dimensions in a curse, primarily through the dominance of the Positional system, based primarily on historically successful exploitation strategies. It is also a blessing, through the dynamism of utilizing the dimensions' different and potentially complementary and symbiotic properties. For the latter to happen, hierarchy needs to give way to heterarchy as an organizing principle. Heterarchy implies, apart from the general admonition to subscribe to the three-dimensional model: many - and different - centeredness; for the MNC, geographical dispersion of critical resources (increases variety); an experimental rather than exploitative attitude; utilization of the broad scanning ability (increases openness to the outside); structuring rewards, and even employee ownership of the corporation, for global ambitions; focusing on learning from specialized, dispersed nodes; horizontal communication; normative integration through corporate culture; a brain-like view of the firm; and, an emphasis on communications infrastructure and human resource management policies. It has many of the desirable traits of an NRS. I have suggested the term N-form corporation (70) as a short-hand for a firm build on heterarchy. The historical M-form is a short-hand term for a firm built on the principle of dividing complexity into quasi-independent parts - the principles of hierarchy and nearly decomposable systems. Thus, there is a symmetry between the terms M-form/hierarchy/decomposability and N-form/heterarchy/recomposability. Whether one needs all three in each case is a matter of conceptual parsimony versus trying to cover too much with one concept. M/N form refers to empirically observable actors, hierarchy/heterarchy to ideal-type structuring principles, and decomposability/recomposability to one main aspect, namely, the posture towards complexity.

The two distinguishing characteristics of the N-form corporation are its focus on knowledge management and the strategy of dealing with complexity as an asset and an arena of choice, rather than something to be avoided and/or simplified. The latter point deserves a speculative, historical excursion.

Chandler Read as Arithmetics

It is curious to consider how well business history can be interpreted in terms of a very simple arithmetic analogy.(71) The emergence of the functionally organized, vertically integrated corporation can be seen as adding resources together that were earlier separated. The importance of the U-form increased with volume growth - again addition. Diversification increased complexity and diversity and, with great pains and time-lags, lead to the splitting up of corporations into semiindependent parts - division. The parts are often even called divisions. Unhampered combination of diversification and growth for many firms lead to disinvestment, "downsizing" and refocusing - subtraction. My argument is that it is now time for an era of multiplication, where parts are put together rather than primarily separated, grown, or shrunk. This is what "N-form" is meant to convey. Historically, the sequence F (separate functions), U (unified functions), M (multidivisional), D (downsized and refocused), and N (multiplicative) can be dated to before the mid-1800s (F), from this until the 1920s (U), from then until the recent past (M), with a great overlap with (D) in later years, and (N) still at an avant-garde stage.

The various conceptions and dominant strategies have historically left traces in the language of business. A function is something you do, implying the concrete work going on in production or sales. M-forms always are characterized by words of partition: divisions, segments, sectors, areas, niches. The whole is characterized as the union, rather than the intersection: conglomerate, business group, holding compancy, concern, sphere. The neologism of the D-form uses words pointing downwards, negations, and concepts implying going back to the old: downsizing, disinvestment, demerging, deconglomerating, refocusing. Now, for the N-form, we hear some words close to the sense of multiplication, primarily combining and integrating. I suggest that there will be much linguistic and organizationally useful innovation in the near future. Mathematics may not be enough. Knowledge management has to do with leverage, synthesis, catalysis, symbiosis, parasitism, selection, mutation, migration, osmosis, metamorphosis, etc., etc.

For an NRS, the connotations of a term such as "global product division" implying chopping up wood rather than putting together knowledge - or even "matrix" (an economical but static exposition of discrete, passive elements) - are inappropriate. Most practicing managers certainly do not care about etymology, so the problem is not overwhelming. However, why not establish an organization built not on Global Product Divisions but on Global Competence Factors? Factors multiply, and they imply that something specific is done.

Global Knowledge Factors would be more consistent with the encompassing use of "Knowledge" in this essay, but here "Competence" sounds better. Also, for most people, the practical, skill-related, tacit parts of knowledge are not easily communicated by "knowledge". In Germanic and Latin languages, there is commonly a pair of words to cover the two parts: wissen/konnen (German), savoir/pouvoir (French). There is also a third sense, implying recognition and familiarity: kennen (German), connaitre (French). In English, there is know/can/recognize, but not with exactly the same meaning and syntax. Recomposability as a Systemic Property

The enumeration of possible candidates, apart from the MNC, for NRS prowess on pp. 11-17 indicates some of their relative strengths and weaknesses on the 12 dimensions. For example, inter-specialist sharing of knowledge is weak in academic fields and strong between networks of dedicated individuals. Openness to the outside is a curse of some specialized economic regions, like rustbelts and Soviet-style combinats, whereas it is usually strong in diasporic networks. The basic balance between exploitation and creation it tilted in the former direction for professional firms - although there are spurts of creativity when the old 2x2 matrix becomes unpopular - and in the latter for academic communities - although the production of 10n papers on the basis of an idea of magnitude n/2 is not uncommon. In addition, there is great variation within the classes.

Therefore, the optimal - both to encourage recomposition of knowledge and to assure both exploitation and creation - is a system of institutions.

At a national level, in order to increase NRS capability, utilizing both national and global knowledge inputs, something like the following would seem to be desirable:

1. A healthy dose of Global Machine Corporations (GMCs), ensuring efficient exploition of past strengths, and assimilation of best practices and dissemination of the fruits of (re)composition through straight-forward transfer of experiences - the "hygiene factor".

2. An avant-garde of Recomposing Agent Firms (RAFs) with a bias for knowledge creation, particularly the more ambitious forms of knowledge transformation through recomposition and synthesis. RAFs may be small, but there should be many of them, because failure rates are likely to be high.

3. Magnet Firms, pulling in competence and human talent from a global resource pool. The focus is on creation, with an emphasis on assimilation. Perhaps IBM was such a firm a few decades ago, and Microsoft and the best investment banks may be ones today.

4. Some Magnet, specialized economic Regions, attracting firms and individuals from the whole world within a certain, rather specific knowledge domain. Both exploitation and creation are involved, as in Hollywood or the Stuttgart-centered automobile/component complex. The risk is, obviously, traps of competence and path dependencies leading down narrowing and finally nonviable roads.(72)

5. National firms, MNCs or not, with access to and membership in global networks - with competitors, suppliers, customers, and related fields of science. The aim is for both exploitation and creation, and useful diversity is strengthened.

6. Strong and specialized academic systems and membership in leading global academic communities. Here, the definite focus is on creation of high order, with recomposition of resources as a main strategy. The Swedish medical technology firm Elekta is a good example of the fruits of such investments. It started from the invention of a system for intensive, targeted gamma-radiation by one of the many geniuses fostered at Karolinska Sjukhuset in Stockholm. It now invests in technologies of image processing and precision robotics, all over the world, and continuously supports medical research. The original impetus, however, was a determined government policy of supporting education and research in general, and medicine in particular.

7. A home for a diaspora of closely, usually ethnically knit individuals and organizations. This provides for increased possibilities of primarily exploitation of existing advantages, but also for the dissemination and assimilation of extra-diasporic generated knowledge.

8. Inwards FDI, providing a base for MNCs from other countries with interrelationships with internal firms and individuals, contributing primarily to domestic exploitation ability, but also to a wider panorama of dissemination possibilities and some assimilation of best practices.(73)

9. A social system with "global" individuals who have been able to make contacts and form intimate bonds with people internally and, particularly, abroad. This implies investments in education, at all levels, in primarily science fields and language, youth exchange programs, and a mentality of supporting "nerds" as well as other idiosyncracies of personality. The creation potential inherent in possible recompositions is great, but also assimilation and dissemination of knowledge is boosted.

10. The definition of socially mobilizing, catalyzing projects of national (or regional, or global) development. They should at this level be formulated in terms of output (needs and demands) rather than input (resources and knowledge categories).(74) The universal call for competence development in the Western world as an answer to unemployment evades the question of competence for what. The main reason is, I believe, the lack of responsibility and vision among the elites in the political as well as economic and cultural realms. Bell's disjunction of realms(75) is one root cause.

The issue of whether and how to initiate and participate in technological paradigm shifts is more difficult. Experience shows that national level action is full of failure. At any rate, bets on the industry of the future should be low and diversified.

Neither are professional firms such as in consulting, law, or investment banking particularly important. They constitute potentially good NRS themselves, but their role and influence on the environment is less, and to the extent there is one - and a beneficial one at that - it does not require domestic origin or abode. What did all the chartered accountants do for Great Britain? The investment bankers and lawyers for the USA? Japan's, Germany's, and Sweden's industrialization proceeded well without much help from internal "professionals". To the extent they were needed, they could be enlisted from abroad and used and constructively misunderstood in the local environment.


An interesting analysis would be what countries, or parts or groups of them, can muster the whole arsenal of NRS mechanisms. Research on national innovation systems dwells on such questions, but with other conceptual lenses than the knowledge management/NRS one.(76) To indicate the type of discussion that might be useful, some rough guesses about 16 countries or their subregions are offered in Table 2. The scores are meant as examples and build on my superficial knowledge of most of the areas, just to show how a serious research program on NRS capability might be designed. For whatever it is worth, the ranking list, using a simple five-degree scale (-2 to +2) with equal weight for all items, is reported in Table 3.

Again, the "analysis" is only meant as an illustration. Also, a genuinely systemic argument cannot just add scores, but has to consider priorities, interactions, and Gestalt properties. Such considerations must hamper any quick conclusions about public policy. Only empirical research, where also the "dependent variable", NRS capability, is operationalized and measured can throw light on such matters. A less problematic hypothesis is that the view of the MNC as an NRS should be complemented with an analysis of the MNC in an NRS. Questions of the assimilative capacity of an MNC arise.(77) Also, there are issues of the distribution of gains from effectively utilizing a global knowledge pool. It may be that an effective economic world order would mean that individuals, academia, small firms and certain regions specialized on knowledge creation, and large multinational firms on exploitation, with large doses of assimilating and adapting innovations from the former group. However, the benefits from such a possible optimum would be affected by power relationship and appropriability regimes.(78) Moreover, one can of course challenge the notion of specialization of roles along the creation/exploitation axis. In order to create something new, you may have to be involved in manufacturing and selling it.(79)

The Recomposing N-form as an Alternative to Market, U-form, and M-forms

One can combine the ideas of de- and recomposability, giving rise to another speculation on a macro scale. According to business historians, and in the logic of internalization theory and transaction cost analysis, in the beginning there were markets.(80) Or at least there were factories selling their wares to wholesalers and further to retailers on the market. Next came the functionally integrated firm - the U-form, to use Chandler's vocabulary.(81) If one compares a market to the U-form, the former is often considered as atomistic and can, even etymologically, not decompose further. There are single buyers and sellers, and supply comes in discrete units and is not bunched together in packages, and the same applies to demand. A market does not decompose further. And, it is not arranged in a hierarchy of either demand or supply. Thus, it is neither an NDS nor an NRS in its ambitious form. However, it has properties of recomposability, but of a rather humble kind. It is here, I believe, that the early Schumpeter's entrepreneur has a role, in combining (the word Schumpeter used) opportunities and resources available for the alert eye. It is not a matter of an organization putting very related, domain-specific things together, a process I have termed recomposition. The archetypical entrepreneurial act is arbitrage, benefitting from the geographical separation of sources of supply and demand. Also the exploitation of an innovation, the future of which needs investment and human energy to materialize, is an example of the Schumpeterian entrepreneur's role. Those are functions with some "multiplication", but not as targeted, ambitious, continuous, and for the long haul as in the case of, say, a computer firm's continuous forays into communications and radio technology and entertainment production. With this qualification, one can still classify the market as a system that does not decompose further, but that is able to at least combine. Recomposition requires decomposition as a first step. The market does not do this, but works with the atomistic raw material. The U-form does neither. It may grow, or shrink, but the main architecture is set. Even if a department like R&D divides into more specialized parts as it grows, the ambition is to integrate the substance of work that goes on. When the internal diversity becomes too large to handle, it is - Chandler has convincingly shown time for the next step, the M-form. The U-form, at least as it developed historically, was not strong on recomposition. Essentially, it was an organization for vertical integration and growth. One-time large steps were taken by, for example, buying suppliers, but after that the idea was primarily to exploit efficiencies, not to generate novelty. With some exaggeration of the lack of dynamism of the large functionally integrated firm, we can regard the U-form as scoring low on both de- and recomposability.

The M-form explicitly does not rely on its divisions interaction or, much less, rearranging relationships. It does, however, entail the idea of further decomposition. A kind of fractal logic of mirroring the large scale pattern at a micro level is visible in the continuous division into ever sharper niches, business areas etc. in most firms. Thus, the M-form excels in division and shunts recomposition.

I have argued for a type of firm, the N-form, that sees recomposition of knowledge, utilizing internal as well as external sources, as one of its main claims to fame: Decomposing is necessary but not sufficient for recomposing, and the N-form firm is an organizational totality that, in contrast to an atomistic market, can be decomposed in many fashions.

The four institutions can be portrayed in a simple 2x2 matrix. Empirical research in business history suggests the sequence described by the arrows. The lacking relationship that catches the eye is the one between market and N-form. Indeed, there is a possibility that an organization set only on recomposition will degenerate to a set of elements without cohesion, i.e., a market. A downward arrow from N-form to market is a prospect and a sort of peril of experimentation. The antidote is to control the recompositional efforts by conscious design along the lines suggested in this paper.

There is also a possibility of the market "coagulating" around complementary resources to form an NRS with capability of continuous reconstellation. Perhaps the NIDs (Networks of Dedicated Individuals) is a case in point, starting from an assemblage of heterogeneous, "atomized" individuals (the expression is really a tautology: individual = not divisible!) but possibly developing to a coherent group, with continuity and strategy beyond those of any one member. Thus, much dynamism seems likely between the N-form and market poles. However, most Schumpeterian type entrepreneurs and innovators will select, or be encouraged by established institutional rules and patterns, to form a company, initially usually in U-form, around an idea or opportunity, rather than bet on successive and continuous recomposing of knowledge.

A Final Reflection

Whether one approaches the problem of organizational knowledge management and creation through a critique of hierarchy, or analysis of the consequences of simultaneous intensity and extensity of knowledge, or consideration of the properties of nearly recomposable systems, one tends to draw similar conclusions. The balance between exploitation and creation is shifted in the direction of the latter. An alternative archetypical organizational principle (heterarchy) is suggested. A specific critique, in terms of requirements for knowledge management, of empirically common forms of corporations is intimated. And, outlines of an alternative, operationally meaningful organizational model are drawn. The MNC, by virtue of its ambition to both intensify and extend its knowledge, and given the strains it experiences in and through its traditional structures and systems, is one main laboratory for such social innovation. The present paper is an effort to contribute to research dialogue and practical experimentation in this direction.

Table 3. Illustrative Ranking List of Societal Global NRS Capability

1. North California 2. South-East France 3. North Italy 4. Stuttgart area 5. Singapore 6. UK 7. Japan 8. Sweden 9. India 10. South China 11. South Africa 12. Wallonia 13. Korea 14. Russia 15. Oklahoma 16. Mexico


1 For one recent example among many, see Drucker 1991.

2 For a relatively early attempt that has won many followers, see Winter 1987.

3 This argument, with reference to sources, will be developed later in the paper.

4 For an enlightening discussion of knowledge intensity, see Starbuck 1992.

5 A fuller treatment should distinguish between the consequences of at least these three types of extension. Jonas Ridderstrale has suggested the terms dispersion, diversity, and durability to cover geography, technical field and change rate of technology, to be added to depth, covering intensity as dimensions of a new kind of contingency framework.

6 See Hedlund 1996, for arguments and a more detailed discussion of the KI/KE framework.

7 In different ways, the "eclectic model", Dunning 1979; the "product life cycle" one, Vernon 1966; the "internationalization process" view, Johanson/Vahlne 1977; the "monopolistic advantage" model, Hymer 1976/1960; and the "internalization" approaches, Buckley/Casson 1976, Rugman 1981, and Hennart 1982, all share the preoccupation with this type of firm and firm evolution.

8 For "heterarchy", see Hedlund 1986, and for transnational, see Bartlett/Ghoshal 1989. An earlier statement of the principle is in Bartlett 1986. Doz 1986, and White/Poynter 1990, suggest similar conceptions.

9 See Perlmutter 1965, whose ethno/poly/geocentrism notions prefigure much of what has been written later.

10 See Ridderstrale/Hedlund 1992/1997, and Hedlund 1994.

11 See Simon 1962.

12 I have discussed the example at length in Hedlund 1993, ibid. I refrain from repetition here.

13 For the analogy with arithmetics, see Hedlund 1994, ibid, and below, pp. 31-32.

14 See Hedlund/Rolander 1990, for experimentation, Ridderstrale/Hedlund 1992/1997, ibid, for creation and March 1991, for exploration.

15 See Kogut/Zander 1995, Teece/Pisano 1994, Henderson 1994.

16 One reason is, I believe, the inspiration from Schumpeter. His early work emphasizes the individual entrepreneur seeing one opportunity/combination and then exploiting it. The idea that large firms would be able to continuously renew themselves was alien to Schumpeter, at least in later days.

17 The three categories closely match Barnard's three main "executive functions", see Barnard 1938.

18 I have to use "productive diversity" rather than "requisite variety", since Ashby's concept is too closely tied to a mechanistic cybernetic frame of reference. It furthermore refers to a systems capacity to react to given environmental changes, rather than to create something new. The basic idea in Ashby (1956) is that a system must contain at least as many dimensions of reaction as the dimensions of environmental change. Basically, the discussion is in terms of control systems such as thermostats. Admittedly, the introjection of "productive", as "requisite" in the Ashby formulation, is a beauty spot, since it entails circularity and tautology. Empirically, the question is, of course, how much and what kind of diversity is productive.

19 This may seem obvious. Still, there is much interest in organization as primarily providing codes of communication among economists and business scholars. Arrow 1974, makes this one of his main points. Later a whole school of thought analyzes management in terms of conversation. See, for example, Brunsson 1989.

20 Firms have always appreciated, and certainly often overestimated, the strength inherent in a corporate-wide grounding in some scientific or technological field. Company names bear witness: General Electric (electricity from light bulbs to nuclear reactors), ASEA-Allmanna (Common) Svenska Elektriska Aktiebolaget, AEG (Allgemeine Elektrische Gesellschaft), General Motors, etc.

21 The ideas of holographic organization (Morgan 1986, El Sawy 1985) and the firm as a brain metaphor rather than the brain of the firm one (Hedlund 1986) will not be pursued here, but it is intimately related to many of the arguments.

22 For tacitness, see Polanyi 1962.

23 For a modern example in the spirit of this paper, see Hagstrom/Hedlund 1994/1998. Contributions to Gagliardi 1990, provide more detailed consideration, albeit from a specialized, symbolic theoretical perspective.

24 See March 1991, for an analysis of effects of tenure. Janis 1982, is the classical reference for group-think.

25 See below and Hedlund/Nonaka 1993, ibid. The argument in the latter is that Japanese firms do exactly this.

26 See Cyert/March 1963.

27 See, for example, Nonaka/Takeuchi 1995, and Peters 1987.

28 For one view, see Collins/Porras 1996.

29 The notion that organizations dealing with drastic change need a disproportionate dose of normative, rather than just calculative and coercive, coordination is a standard one. For a classical statement based on Weber's still more classical one, see Etzioni 1961.

30 For empirical illustrations in the MNC context, see Hedlund/Ridderstrale 1995, and Ridderstrale 1997.

31 This leads to a fundamental reconsideration of notions of organizational structure. I will return to this issue on pp. 24-26.

32 Ridderstale 1997, ibid.

33 See, for example, Hout/Porter/Rudden 1982.

34 For the latter, see the examples in Prahalad/Hamel 1990, and Nonaka/Takeuchi 1995, ibid. Particularly the NEC C&C story is illustrative, and used by both sources.

35 This seems to be one of the most difficult aspects of knowledge management. See Peters 1994, and, for the MNC product development context Hedlund/Ridderstrale 1995, ibid, and Ridderstrale 1997, ibid.

36 The importance of industrial agglomerations is well established in economic geography. A recent synthesis is Porter 1990. For Silicon Valley and Route 128, see Saxenian 1994.

37 See, for example and for the MNC context, Johanson/Mattsson 1988.

38 Such firms were the subject of the analysis in Lindqvist 1991.

39 For visual image processing, see Lundgren 1991.

40 For a selection of readings and summaries of the debate, see Kondratieff 1984, Mensch 1979, Freeman 1984.

41 For the former, Dahmen's "development blocks" is a category in-between technological waves and mobilizing projects. See, Dahmen 1988. For war, see the fascinating analysis in McNeill 1983. For Japan's modernization, see Westney 1987.

42 The civilian fruits of US military spending is a case in point, although probably more for history than for the future. The entertainment industry utilizing advanced consumer electronics is now spearheading technical requirements and combining technologies fervently.

43 For MITI, see Johnson 1982.

44 For example, Hitotsubashi University outside Tokyo still has a professor of "German management" as a vestige of the effort to assimilate the then supposedly most effective methods in the 1930s.

45 See footnote 11. See also Nohria/Ghoshal 1997.

46 The dominant theoretical ideas about the MNC emphasize this aspect. See, in particular Dunning 1979, ibid, and works mentioned in footnote 10.

47 Empirical research provides support. See Ridderstrale/Hedlund 1995, and Ridderstrale 1997, ibid.

48 Belander et al.

49 The alliance and attempted merger between Volvo and Renault is an example of such a mismatch, in my view.

50 No pun intended, should an acronym be attempted...

51 Again, no pun intended, although the acronym would carry some of the meaning!

52 Cf the fuller discussion in Hedlund/Nonaka 1993.

53 Perhaps we could talk about Creationist, Darwinian, Veterinary, and Lamarckian (re)composition strategies.

54 See Bartlett/Ghoshal 1989, ibid.

55 This was a main point already in Bartlett 1982.

56 Of course, a campaign for NRS capability should not be called this. The project is to utilize the total global knowledge and competence resources of the corporation in a forceful manner, and appropriate banners and slogans can be invented to express the intention.

57 Our own empirical work suggests that those firms that do initiate such programs for global communication in product development find them very useful. See Hedlund/Ridderstrale 1995.

58 See Barnard 1938, ibid.

59 Cf Birkinshaw 1997.

60 See Hamel/Prahalad 1994.

61 The same emphasis is applied by Nonaka/Takeuchi 1995, ibid.

62 I will refrain from references or comparisons.

63 For a full discussion and history of evolution of the model suggested below, the following may serve as a guide: Hedlund 1988, poses the three-dimensional structure on the basis of Dionysius' original insight about hierarchy, and suggests that today and for the future, a three-dimensional heterarchy is appropriate. The main reason is the exploitation-biased nature of hierarchy and the experimentation-biased potential in heterarchy discussed in Hedlund/Rolander 1990. Hedlund 1990/1993, provides a detailed critique of hierarchy, where the disjunction of the three dimensions and their new roles are main points. Ridderstrale/Hedlund 1992/1997, extends the discussion of exploitation and creation strategies and their differing requirements. Hagstrom/Hedlund 1994/1998, is a detailed specification and analysis of the three-dimensional model with an empirical example of the Danish hearing-aid manufacturer Oticon. Hedlund 1994 gives the name "N-form corporation" to creation-prone, combinatorial forms and reaffirms their three-dimensional character. Hedlund/Ridderstrale 1995, is an empirical analysis of product development projects in ABB and Electrolux, arrive at conclusions consistent with the idea: stable positional structure, an action structure focused around temporary projects, and a knowledge one which is fluid, partly extra- and inter-organizational, with catalytic and infrastructural roles, but also requiring internal structuring.

64 The original and strong statement about the relations between the two notions is Simon 1962, ibid. My own view about its applicability to the modern firm is expressed in Hedlund 1990/1993, ibid.

65 This was the essence of McCulloch's original idea of heterarchy, where neurons in the brain were hypothesized to be organized in flat and recursive structures, rather than deeply and non-recursively, as computer programs at that time. See McCulloch 1965.

66 "Chunk" is Simon's term, although I have not been able to find the original written source.

67 On the vertical bias, see Galbraith 1973.

68 The logic of the disjunction of dimensions resembles Be!l's analysis of the perils of moving towards a post-industrial society. He speaks about the "disjunction of realms". There are three realms: a techno-economic one, where the "axial principle" is functional rationality; a political one, the axial principle of which is legitimacy; and, culture, dominated by the principle of the expression and remaking of the "self" for realization and fulfillment (Bell 1973, pp. 11-12). While the dimensions do not match those of organization structure, the analysis of how they were once in conjunction, and now out of joint, does. During the rise of industrialization, self-expression and political legitimacy were necessary to free social and individual energy in a technically rational direction. Today, the focus on self-expression threatens the latter. You are supposed to be a disciplined and diligent worker between 9 and 5, and a lazy, reckless and easily manipulated consumer on free-time.

69 The three-dimensional model resembles the "Hypertext organization" in Nonaka/Takeuchi 1995. A comparison with the works mentioned in footnote 63 will establish similarities, differences, and priorities, but this would require a lengthy discussion. I am encouraged by their adoption of a three-dimensional framework: "business system", roughly my Positional system; "project teams", roughly Action system; and, "knowledge base", roughly Knowledge system. However, there are also important differences.

70 See Hedlund 1994.

71 I rely on my interpretation of, primarily, Chandler 1962, 1977. See also Hedlund 1994, ibid.

72 For competence traps, see the long line of work by March. For a discussion in the context of a review of a larger literature, see Levitt/March 1988.

73 However, large FDI in no way guarantees local assimilation, as a large body of research shows. See Kokko 1992, for a review and an empirical study among many. It is also obvious that inwards FDI is not a necessary condition for assimilation of foreign knowledge, as the histories of countries as different as Japan and Sweden show.

74 Because the latter easily leads to favouring either incumbent and firms likely not to contain the seeds of new things, or fashionable "industries of the future" spotted by everybody else. Also, the risks of forms of corruption are not negligable.

75 Cf the argument by Lasch 1996.

76 See contributions in Freeman/Lundvall 1988.

77 For a critical view on assimilative capability, see Solvell/Zander 1994/1997.

78 For a recent treatment of appropriability issues, see Teece/Pisano 1994.

79 Cf the debate on relations between science and technology and their roles for economic progress in Rosenberg/Birdzell 1986. See also Ridderstrale/Hedlund 1992/1997.

80 I do not subscribe to this dictum myself, but the development of the functionally integrated firms in the 1800s roughly fits the idea. Even cursory study of older economic history shows the perils of markets, and the necessary "hierarchical" support necessary to sustain them. See, for example, Braudel 1982.

81 The sequence described, up to the M-form, is of course based on Chandler's pioneering work. See Chandler 1962, 1977.

References Ashby, W. R., An Introduction to Cybernetics, London: Chapman & Hall 1956.

Arrow, K. J., The Limits of Organization, New York: Norton 1974.

Barnard, C. I., The Function of the Executive, Cambridge, MA: Harvard University 1938.

Bartlett, C. A., How Multinational Organizations Evolve, Journal of Business Strategy, 3, 2, 1982, pp. 20-32.

Bartlett, C. A., Building and Managing the Transnational: The New Organizational Challenge, in Porter, M. E. (ed.), Competition in Global Industries, Boston, MA: Harvard Business School Press 1986, pp. 367-401.

Bartlett, C. A./Ghoshal, S., Managing Across Borders. The Transnational Solution, Cambridge, MA: Harvard Business School Press 1989.

Belander, J./Berggren, C./Bjorkman, T./Kohler, C. (eds.), Producing Beyond Frontiers - ABB and the Meaning of Being Local World-wide, Ithaca, NY: Cornell University Press (forthcoming).

Bell, D., The Coming of Post-Industrial Society, New York: Basic Books 1973.

Birkinshaw, J., Entrepreneurship in Multinational Corporations: The Characteristics of Subsidiary Initiatives, Strategic Management Journal, 18, 3, March 1997, pp. 207-229.

Braudel, F., The Wheels of Commerce, Collins 1982.

Brunsson, N., The Organization of Hypocrisy, Chichester: John Wiley & Sons Ltd. 1989.

Buckley, P. J./Casson, M., The Future of the Multinational Enterprise, London: MacMillan 1976.

Business Week, October 7, 1996.

Collins, J. C./Porras, J. I., Building Your Company's Vision, Harvard Business Review, 74, September-October 1996, pp. 65-77.

Chandler, A. D., Strategy and Structure, Cambridge, MA: MIT Press 1962.

Chandler, A. D., The Visible Hand, Cambridge, MA: Harvard University Press 1977.

Cyert, R. M./March, J. G., A Behavioral Theory of the Firm, Englewood Cliffs, NJ: Prentice-Hall 1963.

Dahmen, E., 'Development Blocks' in Industrial Economics, Scandinavian Economic History Review, 36, 1, 1988, pp. 3-14.

Doz, Y. L., Strategic Management in Multinational Companies, Oxford: Pergamon Press 1986.

Drucker, P. F., The New Productivity Challenge, Harvard Business Review, 69, 6, November-December 1991, pp. 69-79.

Dunning, J., Explaining Changing Patterns of International Production: In Defense of the Eclectic Theory, Oxford Bulletin of Economics and Statistics, 4, November 1979, pp. 269-295.

El Sawy, O. A., From Separation to Holographic Enfolding, Paper presented to TIMS Meeting, Boston May 1985.

Etzioni, A. A., A Comparative Analysis of Complex Organization, New York, NY: The Free Press 1961.

Freeman, C., Prometheus Unbound, Futures, February 1984.

Freeman, C./Lundvall, B-A., (eds.), Small Countries Facing Technological Revolution, London: Pinter Publishers Ltd. 1988.

Gagliardi, P. (ed.), Small and Artifacts: Views of the Corporate Landscape, Berlin: Walter de Gruyter & Co. 1990.

Galbraith, J. R., Designing Complex Organization, Reading, MA: Addison-Wesley 1973.

Hagstrom, P./Hedlund, G., The Dynamic Firm: A Three-Dimensional Model of Internal Structure, Paper presented at the Prince Bertil Symposium: The Dynamic Firm, Stockholm May 1994, in Chandler, A. D., Jr./Hagstrom, P./Solvell, O. (eds.), The Dynamic Firm, Oxford University Press 1998.

Hamel, G./Prahalad, C. K., Competing for the Future, Cambridge, MA: Harvard Business School Press 1994.

Hedlund, G., The Hypermodern MNC -A Heterarchy?, Human Resource Management, 25, 1, 1986, pp. 9-35.

Hedlund, G., The First Theory of Hierarchy Contemplation of Its Pervasiveness in Modern Business Life, Research Paper 88/4, Institute of International Business, Stockholm School of Economics 1988.

Hedlund, G., Assumptions of Hierarchy and Heterarchy: An Application to the Multinational Corporation, Research Paper 90/4, Institute of International Business, in Ghoshal, S./Westney, D. E. (eds.), Organization Theory and the Multinational Corporation, London: MacMillan Press Ltd., and New York: St. Martins Press 1993, pp. 211-236.

Hedlund, G., A Model of Knowledge Management and the N-form Corporation, Strategic Management Journal, 15, Special Issue, Summer 1994, pp. 73-90.

Hedlund, G., The Intensity and Extensity of Knowledge: Implications for Possible Futures of the Global Firm, CEMS Business Review, 1, 1996, pp. 111-126.

Hedlund, G./Nonaka, I., Models of Knowledge Management in West and Japan, in Lorange, P./Chakravarthy, B. G./Roos, J./Van de Ven, H. (eds.), Implementing Strategic Processes, Change, Learning, and Cooperation, London: Basil Blackwell 1993, pp. 117-144. Hedlund, G./Ridderstrale, J., International Development Projects - Key to Competitiveness, Impossible or Mismanaged?, International Studies of Management and Organization, 25, 1-2, Spring/Summer 1995, pp. 158-184.

Hedlund, G./Rolander, D., Action in Heterarchies: New Approaches to Managing the MNC, in Bartlett, C. A./Doz, Y. L./Hedlund, G. (eds.), Managing the Global Firm, London: Routledge 1990, pp. 15-46.

Henderson, R., The Evolution of Integrative Capability. Evidence from Product Development in Automobiles and Mainframe Computers, Industrial and Corporate Change, 3, 3, 1994, pp. 607-630.

Hennart, J. F., A Theory of Multinational Enterprise, Ann Arbor, MI: University of Michigan Press 1982.

Hour, T./Porter, M. E./Rudden, E., How Global Companies Win Out, Harvard Business Review, 60, 5, 1982, pp. 98-108.

Hymer, S. H., The International Operations of National Firms: A Study of Direct Investment, Cambridge, MA: The MIT Press 1976 (previously unpublished doctoral dissertation, MIT, 1960).

Janis, I. L., Group Think, 2nd ed. Boston, MA: Houghton Mifflin 1982.

Johanson, J./Mattsson, L.-G., Internationalisation in Industrial Systems - A Network Approach, in Hood, N./Vahlne, J.-E. (eds.), Strategies in Global Competition, London: Croom Helm 1988, pp. 287-314.

Johanson, J./Vahlne, J.-E., The Internationalization Process of the Firm - A Model of Knowledge Development and Increasing Market Commitments, Journal of International Business Studies, 8, 2, 1977, pp. 23-32.

Johnson, C., MITI and the Japanese Miracle: The Growth of Industrial Policy, 1925-1975, Stanford, CA: Stanford University Press 1982.

Kogut, B./Zander, U., What Firms Do? Coordination, Identity, and Learning, Organization Science, December 1995.

Kokko, A., Foreign Direct Investment Host Country Characteristics and Spillovers (Published doctoral dissertation), Stockholm EFI 1992.

Kondratieff, N., The Long Wave Cycle, New York: Richardson & Snyder 1984.

Lasch, C., The Revolt of the Elites: And the Betrayal of Democracy, New York, London: W. W. Norton & Company 1996.

Levitt, B ./March, J., Organizational Learning, Annual Review of Sociology, 14, 1988, pp. 319-340.

Lindqvist, M., Infant Multinationals - The Internationalization of Young, Technology-based Swedish Firms (Published doctoral dissertation), Stockholm: IIB 1991.

Lundgren, A., Technological Innovation and Industrial Evolution - The emergence of Industrial networks (Published doctoral dissertation), Stockholm: EFI 1991.

March, J. G., Exploitation and Exploration in Organizational Learning, Organization Science, 2, 1, 1991, pp. 71-87.

McCulloch, W., Embodiments of Mind, Cambridge, MA: Harvard University Press 1965.

McNeill, W. H., The Pursuit of Power, Oxford: Basil Blackwell 1983.

Mensch, G., Statemate in Technology - Innovations Overcome the Depression, Cambridge, MA: Ballinger 1979.

Morgan, O., Images of Organization, London: Sage Publications Inc. 1986.

Nohria, N./Ghoshal, S., The Differentiated Network: Organizing Multinational Corporations for Value Creation, San Francisco, CA: Jossey-Bass Inc. 1997.

Nonaka, I./Takeuchi, H., The Knowledge-Creating Company, Oxford University Press, Inc. 1995.

Perlmutter, H. V., L'entreprise internationale: trois conceptions, Revue Economique et Sociale, May 1965.

Peters, T., Thriving on Chaos, New York: Harper & Row 1987.

Peters, T., The Pursuit of WOW!, New York: Vintage Books 1994.

Polanyi, M., Personal Knowledge: Towards a Post-Critical Philosophy, New York: Haper Torchbooks 1962.

Porter, M. E., The Competitive Advantage of Nations, New York: The Free Press 1990.

Prahalad, C. K./Hamel, G., The Core Competence of the Corporation, Harvard Business Review, 68, May-June 1990, pp. 79-91.

Ridderstrale, J., Global Innovation - Managing International Projects at ABB and Electrolux, Stockholm: IIB 1997.

Ridderstrale, J./Hedlund, G., Toward the N-form Corporation. Exploitation and Creation in the MNC, Paper presented at a conference, University of South Carolina, May 1992, in Toyne, B./Nigh, D. (eds.), Perspectives on International Business, Theory, Research and Institutional Arrangements, Columbia, SC: University of South Carolina 1997.

Ridderstrale, J./Hedlund, G., Managing International Innovation Projects, Paper presented at the Academy of International Business annual meeting in Seoul, Korea, November 1995.

Rosenberg, N./Birdzell, Jr., L. E., How the West Grew Rich, New York: Basic 1986.

Rugman, A. M., Insides the Multinationals, London: Croom Helm 1981. Saxenian, A., Regional Advantage: Culture and competition in Silicon Valley and Route 128, Cambridge, MA: Harvard University Press 1994.

Simon, H., The Architecture of Complexity, Proceedings of the American Philosophical Society, 106, 1962, pp. 467-482.

Solvell, O. Zander, I., International Diffusion of Knowledge - Isolating Mechanisms and the Role of the MNE, Paper presented at the Prince Bertil Symposium: The Dynamic Firm, Stockholm May 1994, in Chandler, A. D., Jr./Hagstrom, P./Sovell, O. (eds.), The Dynamic Firm, Oxford University Press 1998.

Starbuck, W. H., Learning by Knowledge-intensive Firms, Journal of Management Studies, 29, 6, November 1992, pp. 713-740.

Teece, D./Pisano, G., The Dynamic Capabilities of Firms: An Introduction, Industrial and Corporate Change, 3, 3, 1994, pp. 537-556.

Vernon, R., International Investment and International Trade in the Product Cycle, Quarterly Journal of Economics, 80, 2, 1966, pp. 190-207.

Westney, D. E., Imitation and Innovation: The Transfer of Western Organizational Patterns to Meiji Japan, Cambridge, MA: Harvard University Press 1987.

White, R. E./Poynter, T. A., Organizing for a World-wide Advantage, in Bartlett, C. A./Doz, Y. L./Hedlund, G. (eds.), Managing the Global Firm, London: Routledge 1990, pp. 95-116.

Winter, S. G., Knowledge and Competence as Strategic Assets, in Teece, D. (ed.), The Competitive Challenge - Strategies for Industrial Innovation and Renewal, Cambridge, MA: Ballinger 1987, pp. 159-184.
COPYRIGHT 1999 Gabler Verlag
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1999 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:The Organisational Evolution of the Multinational Enterprise - A Tribute to Gunnar Hedlund
Author:Hedlund, Gunnar
Publication:Management International Review
Date:Apr 15, 1999
Previous Article:National influences on multinational corporation control system selection.
Next Article:Alternatives to decline, threat or scarcity: exit, voice, loyalty and institutional response.

Related Articles
Organisational evolution of the multinational enterprise: an organisational sociological perspective.
From Stopford and Wells's model to Bartlett and Ghoshal's typology: new empirical evidence. (1).

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |