Prospering in Dynamically-competitiveEnvironments: Organizational Capability

as Knowledge Integration

Robert M. GrantSchool of Business, Georgetown University, Washington, DC 20057

AbstractUnstable market conditions caused by innovation and in-creasing intensity and diversity of competition have resultedin organizational capabilities rather than served markets be-coming the primary basis upon which firms establish theirlong-term strategies. If the strategically most important re-source of the firm is knowledge, and if knowledge resides inspecialized form among individual organizational members,then the essence of organizational capability is the integra-tion of individuals' specialized knowledge.

This paper develops a knowledge-based theory of organi-sational capability, and draws upon research into competitivedynamics, the resource-based view of Ihe firm, organizationalcapabilities, and organizational learning. Central to the the-ory is analysis of the mechanisms through which knowledge isintegrated within firms in order to create capability. Thetheory is used to explore firms" potential for establishingcompetitive advantage in dynamic market settings, includingthe role of firm networks under conditions of unstable link-ages between knowledge inputs and product outputs. Theanalysis points to the difficulties in creating the "dynamic"and "flexible-response capabilities" which have been deemedcritical to success in hypercompetitive markets.{Knowledge; Organizational Capability; Competitive Ad-vantage)

IntroductionThe growing intensity and dynamism of competitionacross product markets has had profound implicationsfor the evolution of strategic management thoughtduring the 1980s and 1990s. Increasing turbulence ofthe external business environment has focused atten-tion upon resources and organizational capabilities asthe principal source of sustainable competitive advan-tage and the foundation for strategy formulation. Asthe markets for resources have become subject to thesame dynamically-competitive conditions that have af-flicted product markets, so knowledge has emerged asthe most strategically-significant resource of the firm.This paper seeks to extend our understanding of the

1047-7039/96/0704/0375/S01.25C\ipyright sf W d . Insliiutc for Operations Researchand Ihe Manngcmcnt Sciences

determinants of competitive advantage in dynamically-competitive market environments by analyzing the roleof knowledge in organizational capability. Buildingupon four major theoretical streams: competition as adynamic process., the resource-based view of the firm,organizational capabilities and competences, and orga-nizational knowledge and learning, this paper estab-lishes the rudiments of a knowledge-based theory ofthe firm. At the heart of this theory is the idea that theprimary role of the firm, and the essence of organiza-tional capability, is the integration of knowledge. Thepaper explores how knowledge is integrated to formorganizational capability, and goes on to identiiy char-acteristics of capabilities which are associated withcreating and sustaining competitive advantage in dy-namically-competitive markets, including the achieve-ment of flexible integration across multiple knowledgebases. Finally, I consider the relative merits of internalversus external knowledge integration and the benefitsof firm networks in coping with hypercompetitive mar-ket conditions.

BackgroundThe displacement of static theories of competitionassociated with neoclassical microeconomics and the"structure-conduct-performance" school of industrialeconomics by the more dynamic approaches associatedwith the Austrian school of economics, especially withSchumpeter's concept of competition as a process of"creative destruction" (Schumpeter 1934), has had pro-found implications for strategic management thought{Jacobsen 1992). During the early part of the 1980s,strategy analysis was focused upon the quest formonopoly rent through industry and segment selectionand the manipulation of market structure to createmarket power (Porter 1980). However, if market struc-ture is in a state of flux, and if monopoly rents quicklysuccumb to new sources of competition, approaches to

ORGANIZATION SCIENCE/VOL 7, No. 4, July-August 1996 375

ROBERT M. GRANT Dynamicalfy-competitive Environments

Strategy based upon choices of product markets andpositioning within them are unlikely to yield profitadvantages that are more than temporary. The impactof the resource-based view of the firm on strategicmanagement thinking can be attributed to two factors.First, given the lack of evidence that monopoly poweris an important source of profit (Rumelt 1991),Ricardian rents (returns to resources over and abovetheir opportunity costs) appear to be the primary sourceof interfirm profitability differences. Second, if exter-nal markets are in a state of flux, then the internalresources and capabilities of a firm would appear to bea more stable basis for strategy formulation than theexternal customer focus that has traditionally associ-ated with the marketing-orientation to strategy (Levitt1960).

This emphasis on the "supply-side" rather than the"demand side" of strategy has been closely associatedwith recent work on organizational capabilities.Prahalad and Hamel (1990) argue that sustainablecompetitive advantage is dependent upon building andexploiting "core competences"—those capabilitieswhich are fundamental to a firm's competitive advan-tage and which can be deployed across multiple prod-uct markets. Porter's recent work emphasizes the needfor firms and countries to broaden and upgrade theirinternal advantages in order to sustain and extendcompetitive advantages (Porter 1991, 1992).

While extreme forms of dynamic competition (termed"hypercompetition"by D'Aveni 1994) are characteristicof product markets, dynamically competitive conditionsalso are present in the markets for resources. Indeed,competitive conditions in product markets are driven,in part, by the conditions of competition in the marketsfor resources (Barney 1986). Thus, the speed withwhich positions of competitive advantage in productmarkets are undermined depends upon the ability ofchallengers to acquire the resources needed to initiatea competitive offensive. Sustainability of competitiveadvantage therefore requires resources which are id-iosyncratic (and therefore scarce), and not easily trans-ferable or replicable (Grant 1991). These criteria pointto knowledge (tacit knowledge in particular) as themost strategically-important resource which firms pos-sess (Quinn 1992). Thus, this paper's focus upon knowl-edge and its integration is justified by two assumptionsabout the success in dynamically-competitive marketenvironments:

First, under dynamic competition, superior profitabilityis likely to be associated with resource and capability-basedadvantages than with positioning advantages resulting from

market and segment selection and competitive positionsbased upon some form of "generic strategy";

Second, such resource and capability-based advantagesare likely to derive from superior access to and integrationof specialized knowledge.

The literature on organizational knowledge andlearning has explored the role of organizations in theacquisition, processing, storage, and application ofknowledge (Argyris and Schon 1978, Levitt and March1988, Organization Science 1991, Starbuck 1992). Theprimary emphasis of this literature is on the acquisitionof information by organizations. Nonaka (1994) pro-poses a theory of knowledge creation built arounddynamic interaction between two dimensions of knowl-edge transfer: transformations from tacit to explicitknowledge and vice-versa; and transfers between Indi-vidual, group, organizational, and interorganizationallevels. However, as Spender (1992) recognizes, firmsare engaged not only in knowledge creation but also inknowledge application. The distinction between thesetwo processes is crystallized in Demsetz's (1991) obser-vation that efficiency in the acquisition of knowledgerequires that individuals specialize in specific areas ofknowledge, while the application of knowledge to pro-duce goods and services requires the bringing togetherof many areas of specialized knowledge.

Much of the research into the management issuesconcerning the integration of different types of special-ized knowledge has been within the context of newproduct development (Nonaka 1990, Clark andFujimoto 1991, Wheelwright and Clark 1992). Whilesome innovations are the result of the application ofnew knowledge, others result from reconfiguring exist-ing knowledge to create "architectural innovations"(Henderson and Clark 1990, Henderson and Cockburn1995). This ability ofthe firm to "generate new combi-nations of existing knowledge" and "to exploit Itsknowledge of the unexplored potential of the technol-ogy" is what Kogut and Zander (1992, p. 391) describeas "combinative capabilities".

The theory of organizational capability which followsrepresents an extension and synthesis of these contri-butions, based upon the idea that the essence of orga-nizational capability is the integration of individuals'specialized knowledge.

The Model: Organizational Capabilityas Knowledge IntegrationMy model of organizational capability rests upon basicassumptions regarding the characteristics of knowledge

376 ORGANIZATION SCIENCE/VOI. 7, No. 4, July-August 1996

ROBERT M. GRANT Dynamicalfy-competitiveEnuironments

and its deployment. From these assumptions I developpropositions concerning the nature of organizationalcapability, the linkage of capability to organizationalstructure, and the determinants of competitive advan-tage.

AssumptionsThe focus of this paper is upon a single resource:knowledge. This emphasis is justified by the assump-tions that, first, knowledge accounts for the greaterpart of value added,' second, barriers to the transferand replication of knowledge endow it with strategicimportance. I define knowledge broadly to includeboth "explicit" knowledge which can be written down,and "tacit" knowledge which cannot. The emphasis ison tacit knowledge since, in the form of "know-how",skills, and "practical knowledge" of organizationalmembers, tacit knowledge is closely associated withproduction tasks, and raises the more interesting andcomplex issues regarding its transfer both within andbetween organizations.

The key managerial issues arising from the charac-teristics of knowledge stem from the observation thattacit knowledge is acquired by and stored within indi-viduals. Due to the cognitive limits of the human brain,knowledge is acquired in a highly specialized form: anincrease in depth of knowledge implies reduction inbreadth. Advances in knowledge tend to be associatedwith increased specialization. However, production —the creation of value through transforming input intooutput—requires a wide array of knowledge, usuallythrough combining the specialized knowledge of anumber of individuals.

Knowledge to FormOrganizational CapabilityThese assumptions provide the basis for a knowledge-based view of the firm. If knowledge is a critical inputinto all production processes, if efficiency requires thatit is created and stored by individuals in specializedform, and if production requires the application ofmany types of specialized knowledge, then the primaryrole of the firm is the integration of knowledge. Butwhy are institutions called "firms"needed for the inte-gration of knowledge? It is because the alternatives aretoo inefficient. An individual's ability to integrateknowledge is constrained by cognitive limits: it is notfeasible for each individual to try to learn the knowl-edge possessed by other specialists. Integration acrossmarkets is difficult: in the case of explicit knowledge itis difficult to appropriate the value of the knowledgethrough market contracts; in the case of tacit knowl-

edge, transfer is both difficult and necessitates transac-tion-specific investment. This view of the firm as aninstitution for knowledge integration establishes a viewof the firm based upon close integration between orga-nizational members implying stability, propinquity andsocial relationships, but it does not readily yield preci-sion definition of the firm and its boundaries. For thisreason, Demsetz (1991) refers to "firm-like organiza-tions".

Integration of specialist knowledge to perform adiscrete productive task is the essence of organizationalcapability, defined as a firm's ability to perform repeat-edly a productive task which relates either directly orindirectly to a firm's capacity for creating value througheffecting the transformation of inputs into outputs.Most organizational capabilities require integrating thespecialist knowledge bases of a number of individuals.A hospital's capability in cardiovascular surgery is de-pendent upon integrating the specialist knowledge ofsurgeons, anaesthetist, radiologist, operating-roomnurses, and several types of technicians. L.L. Bean'sorder processing capability, Rubbermaid's new productdevelopment capability and McDonald's Restaurants"capability in preparing and serving hamburgers are allexamples of organizational capabilities requiring theintegration of specialized knowledge across quite largenumbers of employees.

The Architecture of CapabilitiesThe integration of knowledge into organizational capa-bilities may be viewed as a hierarchy. This hierarchy isnot one of authority and control, as in the traditionalconcept of an administrative hierarchy, but is a hierar-chy of integration. At the base of the hierarchy is thespecialized knowledge held by individual organiza-tional members. At the first level of integration arecapabilities which deal with specialized tasks. Movingup the hierarchy of capabilities, the span of specializedknowledge being integrated broadens: task-specific ca-pabilities are integrated Into broader functional capa-bilities ̂ —marketing, manufacturing, R&D. and finan-cial. At higher levels of integration are capabilitieswhich require wide-ranging cross-functional integration—new product development involves especially wide-ranging integration (Clark and Fujimoto 1991). Figure1 illustrates this concept of hierarchy of capabilities byproviding a vertical segment of the hierarchically-arranged organizational capabilities of a manufacturerof private-branch telephone exchanges (PBXs).

The wider the span of knowledge being integrated,the more complex are the problems of creating and

ORGANIZATION SCIENCE/VOI. 7, No. 4, July-August 1996 377

ROBERT M. GRANT Dynamically-competitiveEnuironments

Figure 1 Organizational Capabiiities of a PBX Producer: A Partial Vertical Segment

CHOSS-FUNCTIONALCAPABILtTIES

BROADFUNCTIONALCAPABILITIES

ACTIVITY-RELATEDCAPABILITIES(Operations-related only)

SPECiALIZEDCAPABiLITIES(Manufacturingrelated oniy)

SINGLE-TASKCAPABILITIES(Only thoserelated toPCB assembly)

New productdevelopmentcapability

Customersupportcapability

Qualitymanagementcapability

Operationscapability

R & D anddesigncapability

MIScapability

MarKetingand salescapability

Humanresource mgt.capability

Manufacturingcapabiiity

TMaterialsmanagementcapabiiity

Processengineeringcapabiiily

I TProductengineeringcapability

Testengineeringcapabiiily

Printedcircuit-boardassembly

Automatedthrough-iiolecomponentinsertion

Manualinsertion olcomponents

Surfacemounting otcomponents

Wavesoiderlng

ii N D i V i D U A L S ' S P E C i A L i Z E D K N O W L E D G E

managing organizational capability. The "quick re-sponse capability" which Richardson (1996) identifiesamong apparel suppliers Benetton, The Gap, andGiordano is an important competitive advantage pri-marily because it is difficult to achieve—it involvesintegrating across multiple vertical stages. The difficul-ties experienced by the Bell operating companies intransferring the new capabilities developed in theiroverseas businesses back to their domestic operationscan be attributed to the fact that many of these newcapabilities (e.g. wireless communication, fiber-optics,marketing within competitive markets, and managingjoint ventures) require integration across broad-spansof knowledge and expertise (Smith 1996).

Although higher-level capabilities involve the inte-gration of lower-level capabilities, such integration canonly be achieved through integrating individual knowl-edge. This is precisely why higher level capabilities areso difficult to perform. New product development re-quires the integration of an extremely broad basis ofknowledge, but communication constraints imply that

the number of individuals who can be directly involvedin the process is small.̂ Cross-functional product de-velopment teams are not so difficult to set up, thechallenge (as confirmed by Imai et al. 1985 and Clarkand Fujimoto 1991) is for the team to access thebreadth and depth of functional knowledge pertinentto the product, and integrate that knowledge.

In most companies, hierarchies of capabilities do notcorrespond closely with their authority-based hierar-chies as depicted by organization charts. In particular,some top management capabilities such as capital bud-geting, strategic planning, and government lobbyingmay involve a limited scope of knowledge integration,and hence are closer to the base than to the apex ofthe capability structure. At the same time, if knowledgeis to be integrated effectively by the firm, the architec-ture of capabilities must have some correspondencewith the firm's structure of authority, communication,and decision making, whether formal or informal. Forexample, Clark and Fujimoto find that, within automo-biles, superior capabilities in new product development

378 ORGANIZATION SCIENCE/VO1. 7, No. 4, July-August 1996

ROBERT M. GRANT Dynamically-competitive Environments

require product managers with substantial influenceand decision making authority—what they term"heavyweight product managers".The need tor organi-zational capabiiities to be supported by firm structureposes difficulties for the creation of new capabilities.In the case of the Bell telephone companies, newcapabilities were created outside the formal structurethrough "garbage can" processes (Smith 1996).

Mechanisms for Integrating KnowledgeHow is knowledge integrated by firms to create organi-zational capability? Explicit knowledge involves fewproblems of integration because of its inherent com-municability. Advances in information technology havegreatly facilitated the integration of explicit knowledgethrough increasing the ease with which explicit knowl-edge can be codified, communicated, assimilated,stored, and retrieved (Rockart and Short 1989). How-ever, the most interesting and complex issues concernthe integration of tacit knowledge. The literature pointsto two primary integration mechanisms:

(/) Direction. Demsetz (1991, p. 172) identifies di-rection as the principal means by which knowledge canbe communicated at low cost between "specialists andthe large number of other persons who either arenonspecialists or who are specialists in other fields."To optimize the operation of a McDonald's restaurant,it is more efficient for McDonald's to create an operat-ing manual which covers almost every aspect of therestaurant's management than to educate everyMcDonald's manager in cooking, nutrition, hygiene,engineering, marketing, production management,human resource management, psychology, accountingand finance, and the other specialist areas of knowl-edge embodied in standard operating rules.

The more complex an activity, the greater the num-ber of locations in which that activity must be repli-cated, and the more stringent the performance specifi-cations for the outcome of that activity, the greater isthe reliance on knowledge integration through direc-tion. British Airways operates aircraft maintenance fa-cilities in 67 locations distributed across the globe.Service and repair at these facilities is guided by a hostof highly formalized procedures and directives basedupon the standards established by the major regulatoryauthorities (the Federal Aviation Authority, the BritishCivil Aeronautics Board, and others), guidance andtechnical information provided by aircraft manufactur-ers, and the company's own policies and procedures.These directives, policies, and procedures embody thetechnical knowledge of a large number of specialists.

ORGANIZATION SCIENCE/VOI. 7, No. 4, July-August 1996

(/7) Organizational Routines. Direction involves cod-ifying tacit knowledge into explicit rules and instruc-tions. But since a characteristic of tacit knowledge isthat "we can know more than we can tell" (Polanyi1966), converting tacit knowledge into explicit knowl-edge in the form of rules, directives, formulae, expertsystems, and the like inevitably involves substantialknowledge loss.

An organizational routine provides a mechanism forcoordination which is not dependent upon the need forcommunication of knowledge in explicit form. Marchand Simon (1958, p. 142) "regard a set of activities asroutinized to the extent that choice has been simplifiedby the development of a fixed response to a definedstimuli." Such patterns of stimulus and response maylead to highly complex and variable patterns of seem-ingly-automatic behavior. Within our knowledge-basedview, the essence of an organizational routine is thatindividuals develop sequential patterns of interactionwhich permit the integration of their specializedknowledge without the need for communicating thatknowledge.

Observation of any work team, whether it is a surgi-cal team in a hospital operating room or a team ofmechanics at a grand prix motor race, reveals closely-coordinated working arrangements where each teammember applies his or her specialist knowledge, butwhere the patterns of interaction appear automatic.This coordination relies heavily upon informal proce-dures in the form of commonly-understood roles andinteractions established through training and constantrepetition, supported by a series of explicit and implicitsignals (see Pentland and Rueter 1994, for a carefulanalysis). The advantage of routine over direction is ineconomizing on communication and a greater capacityto vary responses to a broad range of circumstances.

Competitive Advantage inDynamically-competitiveEnvironmentsCreating and Sustaining AdvantageUnder conditions of dynamic competition, the poten-tial of organizational capabilities to earn rents for thefirm through establishing sustainable competitive ad-vantage depends upon their capacity for both creatingand sustaining advantage. Competitive advantage isdetermined by a combination of suppiy-side and de-mand-side factors. On the demand side, a firm's pro-ductive activities must correspond to a market need.On the supply side, the firm must have the capabilities

379

ROBERT M. GRANT Dynamically-competitive Envimnments

not only to serve that market need, but to serve it moreeffectively or efficiently than other firms. For simplic-ity's sake, let us abstract from demand-side considera-tions and focus exclusively upon the supply side: theability to create unique advantages and to protect theseadvantages against imitation.

The first observation is that the critical source ofcompetitive advantage is knowledge integration ratherthan knowledge itself. Specialized knowledge cannot,on its own, provide a basis for sustainable advantage,first, because specialized knowledge resides in individ-uals, and individuals are transferable between firms;second, because the rents generated by specializedknowledge are more likely to be appropriated by indi-viduals than by the firm. Of course, some knowledge(patents, copyrights, trade secrets) is proprietary to thefirm, and is appropriable. However, empirical evidencesuggests that the value of proprietary knowledge de-preciates quickly through obsolescence and imitation(Levin et al. 1987). Hence, even in technology-intensiveindustries, the key to sustainable advantage is notproprietary knowledge itself, but the technological ca-pabilities which permit the generation of new knowl-edge.

If knowledge integration is the basis for competitiveadvantage under dynamic market conditions, what arethe characteristics of knowledge integration associatedwith the creation and sustenance of such an advan-tage? I identify three characteristics of knowledge inte-gration pertinent to the competitive advantage and therents associated with such advantage:

(i) The efficiency of integration—the extent to whichthe capability accesses and utilizes the specialist knowl-edge held by individual organizational members;

(ii) by the scope of integration—the breadth of spe-cialized knowledge the organizational capability drawsupon;

(iii) the flexibility of integration—the extent to whicha capability can access additional knowledge and re-configure existing knowledge.

My goal is to explore the performance requirementsof systems of knowledge integration conducive to at-taining competitive advantage. Given the uniqueness ofeach firm's stock of specialized knowledge and theidiosyncracy of each firm's institutional heritage, it isimpossible to specify the organizational arrangementsconducive to the formation of organizational capabilitythrough knowledge integration. Critical to the analysisof this paper is an equifinality view (Van de Ven andDrazin 1985) that, recognizing uniqueness of knowl-edge bases and institutional conditions, firms canachieve equally effective, yet highly differentiated ap-

proaches to knowledge integration. The key contribu-tion of our analysis is in recognizing the commonrequirements of these different approaches.

The Efficiency of IntegrationCompetitive advantage depends upon how productivefirms are in utilizing the knowledge stored within indi-vidual organizational members, which is dependentupon the ability of the firm to access and harness thespecialized knowledge of its members. Three factorsare important in determining the efficiency with whicha firm integrates the specialized knowledge availablewithin it:

(a) The Level of Common Knowledge. Both directionand routine require communication between individu-als. Demsetz (1991) identifies the prerequisite forcommunication between different specialists as thepresence of common knowledge between them. If spe-cialized knowledge must be reduced to common knowl-edge in order to communicate it, there is inevitablysubstantial information loss. The size of this loss de-pends upon the level and sophistication of commonknowledge. A basic prerequisite is a common language.Direction is almost entirely ability upon detailed articu-lation of instructions. Routines typically rely upon amuch more limited set of cues and responses whichserve not so much as to communicate knowledge, butto permit a sequencing of individual's application ofknowledge inputs. For both integration mechanisms,efficiency of communication depends upon commonal-ity of vocabulary, conceptual knowledge, and experi-ence between individual specialists. Shared behavioralnorms form a central part of the common knowledgewhich facilitates communication and understanding(Garfinkei 1967, Zucker 1987). Generally speaking, thewider the scope of knowledge being integrated (and,hence, the greater the diversity of the individuals in-volved), the lower is the level of common knowledge,and the more inefficient the communication and inte-gration of knowledge. Thus, the effectiveness of socialnetworks among biotechnologists as mechanisms forcommunicating and integrating knowledge reflectedtheir high level of common knowledge arising fromtheir comparatively narrow spread of knowledge andcommonality of behavioral norms (Liebeskind et al.1996). Organizational culture may be regarded as aform of common knowledge, one of the functions ofwhich is to facilitate knowledge integration within thecompany.

(b) Frequency and Variability of Task Performance.The efficiency with which organizational routines

380 ORGANIZATION SCIENCE/VOL 7, No. 4, July-August iy96

ROBERT M. GRANT Dynamicalfy-competitiueEnuironmenL's

integrate the specialized knowledge of team membersdepends upon the sophistication of the system of sig-nalling and responsiveness which develops betweenteam members as a result of repetition and improve-ment. The efficiency of an organizational routine de-rives from the fact that:

•- • I -

While each organization member must know his job, there is

no need for anyone to know anyone else's job. Neither is there

a need for anyone to be able to articulate or conceptualize the

procedures employed by the organization as a whole (Nelson

and Winter 1982. p. 105).

The critical requirement is the "ability to receive andinterpret a stream of incoming messages from othermembers and from the environment" (ibid, p. 100).Integrative efficiency depends upon the effectivenessof this communication in eliciting appropriate re-sponses from each organization member. This is afunction of the frequency with which the particularpattern of coordinated activity is performed. Thegreater the variation in the routine which is required inresponse to variation in environmental circumstances,the lower is integrating efficiency likely to be. Theineffectiveness of the response by the National Guardto the Los Angeles riots of 1992 and the Russian Armyto the Chechnya rebellion of 1994/95 reflects, in part,the infrequence with which these organizations wererequired to suppress insurrection.

(c) Structure. Efficiency of knowledge integration re-quires economizing upon the amount of communica-tion needed to effect integration. Organization struc-tures need to be designed with a view to organizingactivities such as to reduce the extent and intensity ofcommunication needed to achieve knowledge integra-tion. Bureaucracy is a structure which (under certaincircumstances) maximizes the efficiency of knowledgeintegration in an organization where direction is thepredominant integrating mechanism. A key feature oforganiziitional innovatitms such as Henry Ford's mov-ing assembly line, the kanban system for just-in-timescheduling, multidivisional structure (or "M-form") istheir promotion of efficiency through achieving higherlevels of coordination with lower levels of communica-tion.

The principle of modularity is fundamental to thestructuring of organizations to achieve communicationefficiencies. Simon's observation that:

. . . division of labor means factoring the total system of deci-

sions ihat need lo be made into relatively independent subsys-

tems, each one of which can be designed with only minimal

eoneern for its interactions with the others (Simon 1973,

p. 270),

and Williamson's "principle of hierarchical decomposi-tion" (Williamson 1981, p. 1550), may be viewed asorganizational conditions for optimizing the efficiencyof knowledge integration.

Modularity is especially important in organizinghighly complex capabilities which involve broad-scopeknowledge integration. Clark and Fujimoto (1991) showhow the hugely complex task of developing a newmodel of automobile is facilitated by means of organiz-ing the task:

• into sequential phases (concept development, vehicledesign and layout, component design, prototype build-ing, process engineering);• by function (marketing, product engineering, test en-gineering, process engineering);• by product segment (body, chassis, engine, transmis-sion, electrics and electronics).

The problem of many conventional approaches tomodularity is that they rest heavily upon time-sequenc-ing. Under conditions of hypercompetition such se-quencing is simply too time consuming. The organiza-tional challenge is creating modularity which permitseither overlapping phases or full simultaneity.

The Scope of IntegrationIncreases in the span of knowledge which are inte-grated within an organizational capability increases thepotential for both establishing and sustaining competi-tive advantages through two sources:

(i) Different types of specialized knowledge arecomplements rather than substitutes in production. Upto the point of diminishing relevance, the marginalrevenue product of a unit of specialist knowledge in-creases with the addition of different types of knowl-edge.

(ii) The greater the scope of knowledge being inte-grated within a capability, the greater the difficultyfaced by competitors in replicating that capability dueto increases in "causalambiguity"(Lippman and Rumelt1982) and time-based diseconomies of replication(Dierickx and Cool 1989). The complexities associatedwith broad-scope integration are further increasedwhen different types of knowledge require differentpatterns of integration. Toyota's lean production sys-tem combines cost efficiency, quality, flexibility, andinnovation. These different performance dimensionsinvolve different types of integration. While cost effi-

ORGANIZATION SciENCE/Voi. 7, No. 4, July-August 1996 381

ROBERT M. GRANT Dynamicalfy-competitiue Environments

ciency may be best served through organization around"sequential interdependence", flexibility is likely to re-quire more complex patterns of "reciprocal interde-pendence" (Thompson 1967, p. 40). Similar complexi-ties of integration are likely among suppliers of fashionapparel which combine low costs with fashion-baseddifferentiation and quick response capability(Richardson 1996).

The Flexibility of IntegrationWhile integration across a wide scope of specialistknowledge is important in sustaining competitive ad-vantage, hypercompetitive conditions ultimately resultin all positions of competitive advantage being erodedby imitative or innovative competition. Hence, main-taining superior performance ultimately requires thecontinual renewal of competitive advantages throughinnovation and the development of new capabilities.Within the context of our model, there are two dimen-sions to such renewal: extending existing capabilities toencompass additional types of knowledge, and recon-figuring existing knowledge into new types of capabil-ity.

The ease with which existing capabilities can beextended to encompass new knowledge depends heav-ily upon the characteristics of knowledge with regard tocommunicability. If new knowledge is explicit, or iftacit knowledge can be articulated in explicit form,then integrating new knowledge does not pose majordifficulties. In designing its 777 passenger plane, Boe-ing was able to greatly extend its knowledge of elec-tronics and new materials through an advanced CADsystem which provided a common language for special-ists across widely different knowledge areas and differ-ent companies to communicate and integrate. By con-trast. General Motors' upgrading of its manufacturingcapability to encompass the knowledge embodied inToyota's system of lean production was a slow andpainftil process because much of that knowledge wastacit and the routines for its integration were deeplyembedded with Toyota's history and culture.

The reconfiguration of existing knowledge throughnew patterns of integration is more complex, but maybe even more important in creating competitive advan-tage. Such knowledge reconfiguration is centra! toAbernathy and Clark's (1985) concept of ""architecturalinnovation". Subsequent research by Henderson andClark (1990) and Henderson and Cockburn (1995)identifies the critical role of "architectural knowledge"—the "integration of knowledge across disciplinaryand organizational boundaries within the firm"

(Henderson 1995, p. 3)—in driving such innovation.Her studies of pharmaceuticals and the semiconductorphotolithographic alignment equipment industry pro-vide strong support for the role of broad-scope knowl-edge integration in supporting superior performance.

Such architectural innovations are concerned notonly with product and process innovations, but alsowith strategic innovations which reconfigure knowledgeinto new approaches to competing. Such "new-gamestrategies" (Buaron 1981) are not specific to technol-ogy-based industries. Baden-Fuller and Stopford (1994,Chapter 3) show that strategic innovation is fundamen-tal to creating competitive advantage in mature busi-ness environments. In fashion clothing for example,Benetton and The Limited have created "quick-response capability" through innovative approaches tovalue-chain reconfiguration (Richardson 1996).

Most examples of firms' reconfiguring knowledgeinto architectural innovations (EMI's CT scanner, thePolaroid instant camera, the Apple Macintosh,Pilkington's float glass process, Lanier's "virtualreality") and strategic innovations (Nucor in steel,Benetton in apparel, Starbuck's in coffee houses) pointto these innovations as isolated successes rather thanevidence of flexible capabilities which have the capac-ity to continuously and repeatedly reconfigure knowl-edge in new patterns of interaction. Given the difficul-ties inherent in integrating tacit knowledge and depen-dence of such integration upon routines and communi-cation patterns developed over time, establishing orga-nizational arrangements needed to achieve the "flexi-ble integration" proposed by Henderson (1995) and"meta-flexibility" proposed by Volberda (1996) repre-sents a formidable challenge to management. Continu-ous innovation in dynamically-competitive environ-ments (e.g., Rubbermaid in plastic housewares, 3M inadhesive and thin-film products, Sony in consumerelectronics. Motorola in communication products)tends to be the result of the deployment and extensionof a continuing core of capabilities rather than theconstant creation of new capabilities. Achieving flexi-ble integration, either through continually integratingnew tacit knowledge or through constantly reconfigur-ing existing knowledge, is likely to impose substantialcosts in terms of reducing the efficiency of knowledgeintegration. The implication is that radical, discontinu-ous change in industry environments (such as the microrevolution in computing and the possible displacementof internal combustion engines by electric motors inautos) is likely to be accompanied by the decline ofestablished market leaders. The noteworthy feature ofIBM's performance during the 1980s and 1990s is not

382 SciENCE/Vol. 7, No. 4, July-August 1996

ROBERT M. GRANT Dynamicatfy-competitiveEndronments

SO much its decline during the 1990s, but its remark-able success in microcomputers during the 1980s.

Internal Versus External Integration:The Case for NetworksThe need for flexibility in organizational capabilitiesposes complex issues with regard to firm boundariesand choices between internal and external knowledgeintegration. In common with other types of transac-tions, there are three basic alternatives for knowledgetransfer and integration: internalization within the firm,market contracts, and relational contacts (which inmultiple form create firm networks). Given uncertain-ties over appropriability and valuation, market con-tracts are typically inefficient means for transferringknowledge. In Demsetz's (1991) analysis, market trans-actions are only efficient in transferring knowledgewhen that knowledge is embodied within a product.Such transfer of product-embodied knowledge acrossmarkets is efficient when the effective utilization of theproduct by buyers is not dependent upon the buyersneeding access to the knowledge embodied within theproduct. Thus, within the context of fashion apparel(Richardson 1996), Benetton does not need to inte-grate knowledge of the application of computer scienceto computer-aided design into its design capability, if itcan purchase CAD software adequate to its needs. Onthe other hand, expertise in fashion design is tacit, andit cannot be embodied into expert-system software.Thus, Benetton cannot purchase fashion design knowl-edge packaged into software, neither can Benetton relyupon purchasing individual fashion designs from inde-pendent designers because of the need for garmentdesign to integrate multiple knowledge bases: fashiondesign flair, Benetton's own market knowledge, andmanufacturing expertise. The implication is that Benet-ton is likely to require internalization of at least someof fashion design capability. Similarly, in the case ofthe regional Bell companies" expansion into wirelesscommunication. If knowledge concerning wirelessswitching and siting of cellular bases is not capable ofembodiment within marketable products and services,then these companies will, ultimately, be required toextend their capabilities to embody such knowledge(Smith 1996).

Relational contracts, either in individual strategicalliances or broader interfirm networks, are an inter-mediate solution justified by a number of intermediatesituations. For example, explicit knowledge which isnot embodied in specific products cannot be efficientlytransferred through market contracts, but diffusion of

ORGANIZATION SCIENCE/VOL 7, No. 4, July-August 1996

its sources or uncertainty over its applicability to thefirm's products may not justify the internalization of itsproducers within the firm. Networks, either of firms orof individuals, may be well-suited to the transfer andintegration of such knowledge. Thus, in biotechnology,social networks of scientists provide a powerful vehiclefor the transfer of scientific knowledge since suchnetworks provide the reputational assets and the re-peated-game characteristics necessary to avoid theinefficiencies associated with market exchanges(Liebeskind et al. 1996).

Interfirm collaboration through relational contractsis also likely to provide efficient mechanisms for knowl-edge integration where there is a lack of perfect corre-spondence between the knowledge base of the firmand its set of products. The scope of a firm may bedefined in terms of its range of knowledge or its rangeof products. Where the boundaries of both knowledgeand products correspond perfectly, not only are firmboundaries unambiguous, but knowledge resources arefully utilized. Where a perfect correspondence doesnot exist, or where uncertainty exists over the linkagesbetween knowledge and products, then two conse-quences follow:

(a) ambiguity is created over the optimal boundariesof the firm;

(b) internal provision of the full range of specializedknowledge needed for a particular set of products mustresult in the inefficient exploitation of at least some ofthat specialist knowledge.

In such circumstances, interfirm collaboration canincrease the efficiency with which specialized knowl-edge is utilized. A consequence of hypercompetition isuncertainty over links between knowledge inputs andproduct outputs. In biotechnology, new knowledge mayhave applications in "human health, crop productionand protection, chemical feedstock production andprocessing, food processing, and waste management"(Liebeskind et al. 1996). As a result, "these sources oftechnological and competitive uncertainty make it ex-tremely difficult to determine which scientific knowl-edge is potentially valuable and which is not" (ibid).While my analysis points to the superiority of intrafirmrelationships in integrating knowledge, the importanceof networks in sourcing biotechnological knowledgesuggests that the inefficiencies of interorganizationalrelationships are outweighed by the flexibility advan-tages associated with a wider set of knowledge-productlinkages.

A final consideration concerns the speed with whichnew capabilities can be built and extended. Even ifrelational contracts are imperfect vehicles for integrat-

383

ROBERT M. GRANT Dynamicaliy-competitiveEnuironments

ing knowledge, a critical concern is that they canpermit knowledge to be transferred and integratedwith a comparatively short time. If competitive advan-tage in dynamic market settings is critical dependentupon establishing first-mover advantage then the criti-cal merit of firm networks is in providing speed ofaccess to new knowledge. Such considerations provedto be critically important both in biotechnology{Liebeskind et al. 1996, and in telecommunications(Smith 1996). In fashion apparel where the need toaccess new knowledge was less apparent, firm networksdid not provide any clear advantage over vertical inte-gration (Richardson 1996).

Similar considerations explain the establishment ofthe Nordvest Forum regional learning network(Hanssen-Bauer and Snow 1996). Although such in-terorganizational contacts have limited potential forintegrating knowledge across companies, such a net-work permits fuller utilization of knowledge by permit-ting firms to share knowledge that has applicationoutside of each firm's product set. Second, it encour-ages investments in knowledge acquisition in the faceof uncertainty over knowledge-product linkages.

Summary and ConclusionI have established that knowledge is the preeminentresource of the firm, and that organizational capabilityinvolves the integration of multiple knowledge bases.The resulting theory of organizational capability pro-vides a more cogent description of firm competenceand analyzes more precisely than hitherto the relation-ship of organizational capability to competitive advan-tage in markets where market leadership and power iscontinually undermined by competition and externalchange. I show that the processes through which firmsintegrate specialized knowledge are fundamental totheir ability to create and sustain competitive advan-tage. Figure 2 summarizes this theory of organizationalcapability and its implications for competitive advan-tage in hypercompetitive environments.

While making some progress in integrating priorresearch on organizational learning and organizationalresources and capabilities, much remains to be done atboth the empirical and the theoretical level, especiallyin relation to understanding the organizational pro-cesses through which knowledge is integrated. For ex-ample, while organizational routines are generally rec-ognized as important mechanisms for coordinationwithin firms, with a few notable exceptions (e.g.Pentland 1992, Pentland and Rueter 1994), detailedstudy of the operation of organizational routines is

limited. Further progress is critically dependent uponcloser observation of the processes through which tacitknowledge is transferred and integrated.

Despite its limited achievements so far, this analysisoffers considerable potential—especially in buildingbridges between strategic management and organiza-tion theory and design. Conventional notions of organi-zational structure rest heavily upon concepts such asdivision of labor, unity of command, and grouping ofsimilar tasks. The view of the firm as an integrator ofknowledge provides a rather different perspective onthe functions of organization structure. The analysiscan also offer insight into many current developmentsin management practice. Cross-functional product de-velopment teams, TQM, and organizational changeprograms such as GE's "workout" can be viewed asattempts to change organizational structure and pro-cesses to achieve better integration across broad spec-tra of specialized knowledge. The trend towards "em-powerment" takes account of the nature of knowledgeacquisition and storage in firms:—if each employeepossesses unique specialized knowledge and if eachemployee has access to only part of every other em-ployee's knowledge base— t̂hen top-down decisionmaking must be a highly inefficient means of knowl-edge integration. The task is to devise decision pro-cesses that permit integration ofthe specialized knowl-edge held throughout the organization—not just in theboardroom, but on the shop floor as well.

The paper offers little solace to managers grapplingwith the uncertainties and demands of hypercompeti-tive environments. While establishing that, under con-ditions of intense and dynamic competition, internalcapabilities provide a more stable basis for strategythan market positioning, my analysis points to thedifficulties inherent in achieving the dynamic capabili-ties which for many are the "solution" to the problemof sustaining competitive advantage under conditionsof hypercompetition. Volberda (1996) identifies thesedynamic capabilities with "the repertoire of fiexibility-increasing capabilities that management possesses."But, if such capabilities depend upon integration acrossa broad span of largely-tacit knowledge, then a firm'sstrategic fiexibility is limited by two factors: first, itsrepertoire of capabilities is unlikely to extend far be-yond those currently deployed within existing businessactivities; second, the time horizon and uncertaintyassociated with creating new capabilities. The "fiexibieintegration" and network relationships 1 propose asresponses to this problem identify what is required, butoffer little guidance as to the management actionsneeded to achieve flexibility in knowledge integration.

384 ORGANIZATION ScifeNCE/Vol. 7, No. 4, July-August 1996

KOBERl' M. CRANT Dynamically-competitive Environments

Figure 2 Summary: The Knowledge-based Theory of Organizational Capability

ASSUMPTiONS I

• Knowledge is the principal productive resource ofthe firm.*OI the two main types of knowledge, explicit and tacit, the latter is especially important due to its limited transferability.* Tacit knowledge is acquired by and stored within individuals in highly specialized form.• Production requires a wide array of knowledge.

PROPOSITIONS

1. The nature of the firm. The fundamental role of the firm is the integration of individu^s' specialist knowledge. Organizationalcapabilities are the manifestation of this knowledge integration. .

2. Capability and structure. The capabilities of the firm are hierarchically structured according to the scope of knowledge which theyintegrate. Effectiveness in creating and managing broad-scope capabilities requires correspondence between the scope ofknowledge and the structures needed for managing such integration.

3. Integration mechanistns. Two primary mechanisms exist to integrate knowledge; direction and routine. Reliance upon directionincreases with complexity of ihe activity, the number of locations in which the activity is performed, and the stringency ofperformance specifii::ations. The advantage of routine in integrating tacit knowledge is in economizing upon communication andpermitting flexible responses to changing circumstances.

A. Capability and competitive advantage.(A) The competitive advantage conferred by an organizational capability depends, in part, upon the efficiency of knowledge

integration wnich is a function ol: (a) the level of common knowledge among organizationai members; (b) frequency and variabilityof the activity; (c) a structure vt^ich economizes on communication {e.g., through some form of modularity),

(B) An organizational capability's potential for establishing and sustaining competitive advantage increases with the span ofknowledge integrated. ' •, "

(C) Sustaining competitive advantage under conditions of dynamic competition requires continuous innovation which requires flexibleintegration through either (a) extending existing capabilities to encompass new knowledge, or (b) reconfiguring existingKnowledge within new patterns of integration. Since efficient integration of tacit knowledge requires experience through repitition,achieving flexible integration represents a formidable management challenge.

IMPLICATION: FIRM NETWORKS UNDER HYPERCOMPETITION

Firm networks based upon relational contracts are an efficient and effective basis for accessing knowiedge:

- where knowledge can be expressed in explicit form;- where there is a lack of perfect correspondence between the knowledge domain and product domain of individual firms, or

uncertainty over the product-knowledge linkages;- where speed in extending the knowledge base of the firm is critical in creating competitive advantage.

Acknowledgements1 am grateful to Jon Hanssen-Bauer. Duane Helleloid, Arie Lxwin,Julia Liebeskind, Steve Postrcl, James Richardson. Ann Smith, andHenk Volberda for helpful comments and suggestions.

Endnotes'The pan ol national income attributable to knowledge may becalculated as wages and salaries over and above that which would beearned by unskilled manual labor, plus royalties and license fees. Totliis a major part ol profit can be added, since profit is a return to the

ORGANIZATION SciENCE/Voi. 7, No. 4, July-August 1996

resources owned by the firm, a major part of which comprise orembody the knowledge of people. International differences in livingstandards and productivity are mainly due to differences in humancapital, Denison's research into international differences in growthrates found thai, in the case of Britain, advances in knowledgeaccounted for 46 percent increases in real national income perperson employed between 1950 and 1960 (Denison iy68).^A key distinction between an administrative hierarchy and the hier-archy of capabilities is that, in the administrative hierarchy, the spanof control can remain constant throughout the hierarchy. In the

385

ROBERT M. GRANT Dynamicalfy-competitiue Environments

hierarchy of capabilities, the fact that each layer of capabilities

cannot directly integrate the preceding layer of capabiiities and mustretum to the base in terms of integrating individual's knowledge,

means that the span of integration increases as one ascends thehierarchy.

ReferencesAbernathy. W. J. and K. B. Clark (1985), "Innovation: Mapping the

Winds of Creative Destruction", Research Policy, 14, 3-22.

Argyris, C. and D. A. Schon (1978). Organizational Leaming, Read-

ing, MA: Addison-Wesley.

Baden-Fuller, C. and J. M. Stopford (1994), Rejuvenating the Mature

Enterprise. Boston, MA: Harvard Business Schooi Press,Barney, J. (1986), "Strategic Factor Markets: Expectations, Luck and

Business Strategy," Management Science, 32, 1231-1241.

Buaron, R. (1981), "New-game Strategies", McKinsey Quarterfy, Fall,24-40,

Clark, K. B. and T. Fujimoto (1991), Product Development Perfor-

mance, Boston, MA: Harvard Business School Press,

Demsetz, H. (1991), "The Theory of the Firm Revisited." in O. E,

Williamson and S. Winter (Eds.), The Nature of the Firm, New

York: Oxford University Press, 159-178.

Denison, E. F. (1968), "Economic Growth," in R. E. Caves (Ed.),

Britain's Economic Prospects, Washington: Brookings Institution.

Dierickx. I. and K. Cool (1989), "Asset Stock Accumulation and

Sustainability of Competitive Advantage," Management Science,

35,1504-1513.

Garfinkei, H. (1967), Studies in Ethnomethodology, Englewood Cliffs,

NJ: Prentice-Hall.

Grant, R. M. (1991), "The Resource-based Theory of Competitive

Advantage: Implications for Strategy Formulation", Califomia

Management Review, 33, 3, 114-135.

Hanssen-Bauer, J. and C. C. Snow (1996), "Responding to Hyper-

competition: The Structure and Processes of a Regional Learn-ing Network Organization," Organization Science, 1, 4, 413-427.

Henderson, R. and K. Clark (199(1). "Architectural innovation: The

Reconfiguration of Existing Product Technologies and the Fail-

ure of Established Firms", Administrative Science Quarterly, 35,

9-31.

Henderson, R. and 1. Cockburn (1995), "Measuring Competence?Exploring Firm Effects in Pharmaceutical Research." Strategic

Management Joumal, 15, winter, 63-84.

Imai, K., I. Nonaka, and H. Takeuchi (1985). "Managing the NewProduct Development Process: How Japanese Companies Learn

and Unlearn," in K. Clark, R. Hayes, and C. Lorenz (Eds.), The

Uneasy Alliance, Boston, MA: Harvard Business School Press.

Jacobsen, R. (1992), "The 'Austrian' School of Strategy," Academy of

Management Review, 17, 782-805.Kogut. B. and U. Zander (1992), "Knowledge ofthe Firm, Combina-

tive Capabilities, and the Replication of Technology," O/ganira-tion Studies. 3, 383-397.

Levin, R. C, A. K. Klevorick, R. R. Nelson, and S. G. Winter (1987),"Appropriating the Returns from Industrial Research and De-

velopment," Brooking.^ Papers on Economic Activity. 3, 783-820.Levitt, B. and J. G. March (1988). "Organizational Leaming," An-

nual Review of Sociology, 14, 319-340.

Levitt, T. (1960), "Marketing Myopia." Harvard Business Review,

July-August. 24-47.

Liebeskind, J. P., A. Oliver, L. Zucker, and M. Brewer (1996),"Sociat Networks, Learning, and Flexibility: Sourcing Scientific

Knowledge in New Biotechnology Firms," Organization Science,

1, 4. 428-443.

Lippman, S. and R. Rumelt (1982), "Uncertain Imitability; An Anal-

ysis of Interfirm Differences in Efficiency Under Uncertainty,"Bell Joumal of Economics, 13, 418-438.

March, J. and H. Simon (1958), Organizations, New York: Wiley.

Nelson, R. and S. Winter (1982), An Evolutionary Theory of Eco-

nomic Change. Cambridge: Belknap.Nonaka, 1. (1990), "Redundant, Overlapping Organization: A

Japanese Approach to Managing the Innovation Process," Cali-

fomia Management Review, 32, Spring, 27-38.

(1994), "A Dynamic Theory of Organizational Knowledge

C.rtz\\or\," Organization Science, S, 1, 14-37.Organization Science (1991), Special Issue, "Organizational Learning:

Papers in Honor of (and by) James G. March,"2. 1, 1-163.

Pentland, B. T. (1992), "Organizing Moves in Software Support,"

Administrative Science Quarterfy, 37, 527-548.

and H. H. Rueter (1994), 'Organizational Routines as Gram-

mars oi Aziion" Administrative Science Quarterfy, 39, 484-510.

Polanyi, M. (1966), The Tacit Dimension, New York: Anchor Day.

Porter, M. E. (1991), The Competitive Advantage of Nations. New

York: Free Press.

(1992), "Towards a Dynamic Theory of Strategy," Strategic

Management Joumat, 12. Winter Special Issue. 95-118.

Prahalad, C. K. and G. Hamel (1990), "The Core Competences of

the Coiporalion" Harvard Business Review. May-June, 79-91.Ouinn, J. B. (1992), Intelligent Enterprise, New York: Free Press.

Richardson. J. (1996), "Vertical Integration and Rapid Response in

Fashion Apparel," Organization Science, 7, 4, 400-412.

Rockart, J. F. and J. E. Short (1989), "IT in the 1990s: ManagingOrganizational Interdependence," Sloan Management Review, 30,

2, 17-33.Rumelt. R. P. (1991). "How Much Does Industry Matter?", Stmtegic

Management Joumal, 12, 167-185.

Schumpeter, J. A. (1934), Vie Theory of Economic Development,

Cambridge, MA: Harvard University Press.Simon, H. A. (1973), "Applying Information Technology to Organi-

zation Design," Public Administration Review, lOfi. 467-482.

(1991). "Bounded Rationality and Organizational Learning,"Organization Science, 2, 125-134.

Smith, A. and C. ZcithamI (1996), "Baby Bells, Garbage Cans, andHypercompetition," Ofgtiniza/ion Sciencel, 4, 388-399.

Spender, J-C. (1992), "Limits to Leaming from the West: HowWestern Management Advice May Prove Limited in EasternEurope," Intemational Executive, 34, 5, September/October,

389-410.Starbuck, W. H. (1992), "Leaming by Knowledge-intensive Firms,"

Joumal of Management Studies, 29, 713-739.Thompson. J. D. (1967), Organizations in Action, New York:

McGraw-Hill.Van de Ven, A. H. and R. Drazin (1985), "The Concept of Fit in

Contingency Theory." in L. L. Cummins and B. Staw (Eds.),

386 ORGANIZATION ScrENCE/Vol. 7, No. 4, July-August 1996

ROBERT M. GRANT Dynamically-competitive Environments

Re.wanh in Organizational Behavior, 1, Greenwich, CY\ JAIPress, 333-365.

Volberda, H. W. (19%), "Towards the Flexible Form: How to

Remain Vital in Hypercompetitive Environments," Organization

Science, 1, 4, 359-374.

Wheelwright, S. C. and K. B. Clark (1992), Revolutionizing Product

Detielopment, New York: Tree Press.Williamson, O. E. (1975), Markets and Hierarchies, Englewood Cliffs,

NJ: Prentice-Hall.

(1981), "The Modem Corporation: Origins, Evolution, M-

?,," Joumal of Economic Literature. 19, 1537-1568.

Winter, S. G. (1987), "Knowledge and Competence as StrategicAssets," in D. Teece (Ed.), The Competitive Challenge, Cam-bridge, MA: Ballinger.

Zucker, L. (1987), "Institutional Theories of Organizations,"

Review of Sociology, 13, 443-464.

.Accepted by Arie Y. Lewin; received June 1994. This paper has been with the author for two revisions.

ORGANIZATION SCIENCE/VOI. 7, No. 4, July-August 1996