CHAPTER 42 The Dynamic Capabilities of Firms*cmap.upb.edu.co › rid=1T319V6NG-29P7SD1-1RM › Teece y...The Dynamic Capabilities of Firms 197 specific assets. Each approach asks different,

CHAPTER 42

The Dynamic Capabilities of Firms*David Teece' and Gary Pisano^

' Institute of Management, Innovation and Organization, Haas School of Business, Univer-sity of Califomia, Berkeley, CA, USA

^Graduate School of Business, Harvard University, Boston, MA, USA

An expanded paradigm is needed to explain how competitive advantage is gained and held.Firms resorting to 'resource-based strategy' attempt to accumulate valuable technology assetsand employ an aggressive intellectual property stance. However, winners in the globalmarketplace have been firms demonstrating timely responsiveness and rapid and flexibleproduct innovation, along with the management capability to effectively coordinate and rede-ploy intemal and external competences. This source of competitive advantage, 'dynamiccapabilities', emphasizes two aspects. First, it refers to the shifting character of the environ-ment; second, it emphasizes the key role of strategic management in appropriately adapting,integrating, and re-configuring intemal and extemal organizational skills, resources, and func-tional competences toward a changing environment. Only recently have researchers begun tofocus on the specifics of developing firm-specific capabilities and the manner in which com-petences are renewed to respond to shifts in the business environment. The dynamic capabili-ties approach provides a coherent framework to integrate existing conceptual and empiricalknowledge, and facilitate prescription. This chapter argues that the competitive advantage offirms stems fi-om dynamic capabilities rooted in high performance routines operating insidethe firm, embedded in the firm's processes, and conditioned by its history. It offers dynamiccapabilities as an emerging paradigm of the modem business firm that draws on multipledisciplines and advances, with the help of industry studies in the USA and elsewhere.

Keywords: Competitive Advantage; Dynamic Capabilities; Intellectual Property; Tech-nology

1 Introduction

The global competitive battles in high technology industries such as semiconduc-tors, infonnation services, and software have demonstrated the need for an ex-panded paradigm to understand how competitive advantage is gained and held.Well-known companies like IBM, Texas Instruments, Phillips, and others appearto have followed a 'resource-based strategy' of accumulating valuable technologyassets, often guarded by an aggressive intellectual property stance. However, thisstrategy is often not enough to support a significant competitive advantage. Win-

' This chapter is reprinted by permission of the Oxford University Press from Industrial andCorporate Change, Vol,3, No, 3.

196 David Teece and Gary Pisano

ners in the global marketplace have been firms that can demonstrate timely re-sponsiveness and rapid and flexible product innovation, coupled with the man-agement capability to effectively coordinate and redeploy intemal and extemalcompetences. Not surprisingly, industry observers have remarked that companiescan accumulate a large stock of valuable technology assets and still not have manyuseful capabilities.

We refer to this source of competitive advantage as 'dynamic capabilities' toemphasize two key aspects which were not the main focus of attention in previousstrategy perspectives. The term 'dynamic' refers to the shifting character of theenvironment; certain strategic responses are required when time-to-market andtiming is critical, the and extemal organizational skills, resources, and functionalcompetences toward changing pace of innovation is accelerating, and the nature offuture competition and markets is difficult to determine. The term 'capabilities'emphasizes the key role of strategic management in appropriately adapting,integrating, and re-configuring intemal environment.

The notion that competitive advantage requires both the exploitation of existingintemal and extemal firm-specific capabilities and developing of new ones is par-tially developed in Penrose (1959), Teece (1982), and Wemerfelt (1984), How-ever, only recently have researchers begun to focus on the specifics of how someorganizations first develop firm-specific capabihties and how they renew compe-tences to respond to shifts in the business environment (e.g., Iansiti and Clark(1995), Henderson (1994)). These issues are intimately tied to the firm's businessprocesses, market positions, and expansion paths. Several writers have offeredinsights and evidence on how firms can develop their capability to adapt and evencapitalize on rapidly changing environments (see Hayes et al. (1988), Prahaladand Hamel (1990), Dierickx and Cook (1989), Chandler (1990), and Teece(1993)). The dynamic capabilities approach provides a coherent framework thatcan both integrate existing conceptual and empirical knowledge, and facilitate pre-scription. In doing so, it builds on the theoretical foundations provided bySchumpeter (1934), Penrose (1959), Williamson (1975, 1985), Bamey (1986),Nelson and Winter (1982), Teece (1988), and Teece et al. (1994).

2 Toward a Dynamic Capabilities Framework

2.1 Markets and Strategic Capabilities

Different approaches to strategy view sources of wealth creation and the essenceof the strategic problem faced by firms differently. The competitive forcesframework sees the strategic problem in terms of market entry, entry deterrence,and positioning; game-theoretic models view the strategic problem as one of theinteraction between rivals with certain expectations about how each other will be-have;' resource-based perspectives have focused on the exploitation of firm-

' In sequential move games, each player looks ahead and anticipates his rivals' future re-sponses in order to reason hack and decide on an action, i,e,, look forward, reason hack-ward.

The Dynamic Capabilities of Firms 197

specific assets. Each approach asks different, often complementary, questions. Akey step in building a conceptual framework related to dynamic capabilities is toidentify the foundations upon which distinctive and difficult-to-replicate advan-tages can be built.

A useful way to vector in on the strategic elements of the business enterprise isto first identify what is not strategic. To be strategic, a capability must be honedto a user need (so that there are customers), unique (so that the products/servicesproduced can be priced without too much regard to competition), and difficult toreplicate (so that profits will not be competed away). Accordingly, any assets orentity that is homogeneous and can be bought and sold at an established price can-not be all that strategic (Bamey, 1986). What is it, then, about firms that under-grids competitive advantage?

To answer this, one must first make some fundamental distinctions betweenmarkets and internal organization (firms). The essence of the firm, as Coase(1937) pointed out, is that it displaces market organization. It does so in the mainbecause inside the firm one can organize certain types of economic activity inways one cannot using markets. This is not only because of transaction costs, asWilliamson (1975, 1985) has emphasized, but aJso because there are many typesof arrangements where injecting high powered (market-like) incentives might wellbe quite destructive of the cooperative activity and leaming. Indeed, the essenceof internal organization is that it is a domain of unleveraged or low-powered in-centives. By unleveraged we mean that rewards are determined at the group ororganization level, not primarily at the individual level, in an effort to encourageteam behavior, not individual behavior, in order to accomplish certain tasks well.Inside an organization, exchange cannot take place in the same manner that it canoutside and organization, not just because it might be destructive to provide high-powered individual incentives, but also because it is difficult if not impossible totightly calibrate individual contributions to a joint effort. Hence, contrary to Ar-row's (1969) view of firms as quasi markets, and the task of management to injectmarkets into firms, we recognize the inherent limits and possible counterproduc-tive results of attempting to fashion firms into clusters of internal markets. In par-ticular, leaming and intemal technology transfer may well be jeopardized.

Indeed, what is distinctive about firms is that they are domains for organizingactivity in a non-market-like fashion. Accordingly, as we discuss what is distinc-tive about firms, we stress competences/capabilities which are ways of organizingand getting things done which cannot be accomplished by using the price systemto coordinate activity. The very essence of capabilities/competences is that theycannot be readily assembled through markets (Teece, 1982, 1986a; Kogut andZander, 1992). If the ability to assemble competences using markets is what ismeant by the firm as a nexus of contracts (Fama, 1980), then we unequivocallystate that the firm about which we theorize cannot be usefully modeled as a nexusof contracts. By contracts we are referring to a transaction undergirded by a legalagreement, or some other arrangement which clearly spells out rights, rewards,and responsibilities. Moreover, the firm as a nexus of contracts suggests a seriesof bilateral contracts orchestrated by a coordinator, where our view of the firm isthat the organization takes place in a more multilateral fashion, with pattems ofbehavior and leaming being orchestrated in a much more decentralized fashion.


The key point, however, is that the properties of intemal organization cannot bereplicated by a portfolio of business units amalgamated through formal contracts,as the distinctive elements of intemal organization simply cannot be replicated inthe market.^ That is, entrepreneurial activity cannot lead to the immediate replica-tion of unique organization skills through simply entering a market and piercingthe parts together overnight. Replication takes time, and the replication of bestpractice may be illusive. Indeed, firm capabilities need to be understood not interms of balance sheet items, but mainly in terms of the organizational stmcturesand managerial processes which support productive activity. By construction, thefirm's balance sheet contains items that can be valued, at least at original marketprices (cost). It is necessarily the case, therefore, that the balance sheet is a poorshadow of a firm's distinctive competence.^ That which is distinctive cannot bebought and sold short of buying the firm itself, or one or more of its subunits.

There are many dimensions to the business firm that must be understood if oneis to grasp firm-level distinctive competences/capabilities. In this chapter wemerely identify several classes of factors that will help determine a firm's dynamiccapabilities. We organize these into three categories: processes, positions, andpaths.

2.2 Processes, Positions, and Paths

We advance the argument that the strategic dimensions of the firm are its manage-rial and organizational processes, its present position, and the paths available to it.By managerial and organizational processes we refer to the way things are done inthe firm, or what might be referred to as its 'routines', or pattems of current prac-tice and leaming. By position we refer to its current endowment of technologyand intellectual property, as well as its customer base and upstream relations witlisuppliers (we also recognize its strategic alliances with competitors.). By paths werefer to the strategic altematives available to the firm, and the attractiveness of theopportunities that lie ahead. Our focus throughout is on asset stmctures for whichno ready market exists, as these are the only assets of strategic interests. A finalsection focuses on replication and imitation, as it is these phenomena that deter-mine how readily a competence or capability can be closed by competitors, andtherefore the durability of its advantage.

The firm's processes and positions collectively encompass its capabilities orcompetences. A hierarchy of competences/capabilities ought be recognized, assome competences may be on the factory fioor, some in the R&D labs, some inthe executive suites, and some in the way everything is integrated. A difficult-to-replicate or difficult-to-imitate competence/capability can be considered a distinc-tive competence. As indicated, the key feature of distinctive competence and ca-

^ As we note in Teece et al. (1994), the conglomerate offers few, if any, efficiencies be-cause there is little provided by the conglomerate form that shareholders cannot obtain forthemselves simply by holding a diversified portfolio of stocks.' Owners' equity may reflect, in part, certain historic capabilities. Recently, some scholarshave begun to attempt to measure organizational capability using financial statement data.See Baldwin and Clark (1991) and Lev and Sougiannis (1996).


pabilities is that there is not a market for them, except possibly through the marketfor business units'* or corporate control. Hence, competences and capabilities areintriguing assets as they typically must be built because they cannot be bought.Dynamic capabilities are the subset of the competences/capabilities which allowthe firm to create new products and processes, and respond to changing marketcircumstances.

2.3 Organizational and Managerial Processes

2.3.1 Integration

While the price system supposedly coordinates the economy, managers coordinateor integrate activity inside the firm. How efficiently and effectively intemal coor-dination or integration is achieved is very important (Aoki, 1990). Indeed, RonaldCoase, author of the pathbreaking 1937 article "The Nature of the Firm," whichfocused on the costs of organizational coordination inside the firm as compared toacross the market, half a century later has identified as critical the understandingof "why the costs of organizing particular activities differ among firms" (Coase,1988; p. 47). We argue that a firm's distinctive ability needs to be understood as arefiection of distinctive organizational or coordinative capabilities. This form ofintegration (i.e., inside business units) is different from the integration betweenbusiness units; they could be viable on a stand-alone basis (extemal integration).For a useful taxonomy, see Iansiti and Clark (1995).

How efficiently and effectively extemal coordination or integration is achievedis also very important. Increasingly, strategic advantage requires the integration ofextemal activities and technologies. The growing literature on strategic alliances,the virtual corporation, and buyer-supplier relations and technology collaborationevidences the importance of extemal integration and sourcing. Amy Shuen (1994)examines the gains and hazards of the technology make-versus-buy decision andsupplier co-development.

Tbere is some field-based empirical research that provides support for the no-tion that the way production is organized by management inside the fnm is thesource of differences in firms' competence in various domains. For example,Garvin's (1988) study of eighteen room air conditioning plants reveals that qualityperformance was not related to either capital investment or the degree of automa-tion of the facilities. Instead, quality performance was driven by special organiza-tional routines. These included routines for gathering and processing information,for linking customer experiences with engineering design choices, and forcoordinating factories and component suppliers. Garvin (1998) provides atypology of organizational processes.

The work of Clark and Fujimoto (1991) on project development in the automo-bile industry also illustrates the role played by coordinative routines. Their studyreveals a significant degree of variation in how different firms coordinate the vari-ous activities required to bring a new model from concept to market. These dif-ferences in coordinative routines and capabilities seem to have a significant im-

•* Such competences may unravel if the subunit is separated from the parent.


pact on such performance variables as development cost, development lead times,and quality. Furthermore, they tended to find significant firm-level differences incoordination routines and these differences seem to have persisted for a long time.This suggests that routines related to coordination are firm-specific in nature.

Also, the notion that competence/capability is embedded in distinct ways of co-ordinating and combining helps to explain how and why change can have devas-tating impacts on incumbent firms' abilities to compete in a market. Hendersonand Clark (1990), for example, have shown that incumbents in the photolitho-graphic equipment industry were devastated by innovations that had major im-pacts on the configuration of systems. They attributed these difficulties to the factthat systems-level or 'architectural' innovations often require new routines to inte-grate and coordinate engineering tasks. These findings and others suggest thatproductive systems display high interdependency, and that it may not be possibleto change one level without changing others. This appears to be tme with respectto the 'lean production' model (Womack et al., 1991) which has now transformedthe Taylor or Ford model of manufacturing organization in the automobile indus-try. Details appear in the appendix to this chapter.

Lean production requires distinctive shop fioor practices and processes as wellas distinctive higher order managerial processes. Put differently, organizationalprocesses often display high levels of coherence, and when they do, replicationmay be difficult because it requires systemic changes throughout the organizationand also among interorganizational linkages which might be very hard to effectu-ate. Put differently, partial imitation or replication of a successful model mayyield zero benefits.

The notion that there is a certain rationality or coherence to processes and sys-tems is not quite the same concept as corporate culture, as we understand the lat-ter. Corporate culture refers to the values and benefits that employees hold; cul-ture can be a de facto govemance system as it mediates the behavior of individualsand economizes on more formal administrative methods. Rationality or coherencenotions are more akin to the Nelson and Winter (1982) notion of organizationalroutines. However, the routines concept is a little too amorphous to properly cap-ture the congmence amongst processes and between processes and incentives thatwe have in mind. Consider a professional service organization like an accountingfirm. If it is to have relatively high-powered incentives that reward individual per-formance, then it must build organizational processes that channel individual be-havior; if it has weak or low-powered incentives, it must find symbolic ways torecognize the high performers, and it must use altemative methods to build effortand enthusiasm.

What one may think of as styles of organization in fact contain necessary, notdiscretionary, elements to achieve performance. Recognizing the congmences andcomplementarities among processes, and between processes and incentives, iscritical to the understanding of organizational capabiiities. In particular, they canhelp us explain why architectural and radical innovations are so often introducedinto an industry by new entrants. The incumbents develop distinctive organiza-tional processes that cannot support the new technology, despite certain overtsimilarities between the old and the new. The frequent failure of incumbents tointroduce new technologies can thus be seen as a consequence of the mismatchthat may exist between the set of organizational processes needed to support the


conventional product/service and the requirements of the new. Radical organiza-tional re-engineering will usually be required to support the new product, whichmay well do better embedded in a separate subsidiary where a new set of coherentorganizational processes can be fashioned (Abemathy and Clark, 1985).

2.3.2 Learning

Perhaps even more important than integration is leaming. Leaming is a processby which repetition and experimentation enable tasks to be performed better andmore quickly and new production opportunities to be identified (for a useful re-view and contribution, see Levitt and March (1988)). In the context of the firm, ifnot more generally, leaming has several key characteristics.

First, leaming involves organizational as well as individual skills (Mahoney,1995). While individual skills are of relevance, their value depends on their em-ployment, in particular organizational settings. Leaming processes are intrinsicallysocial and collective and occur not only through the imitation and emulation ofindividuals, as with teacher-student or master-apprentice, but also because of jointcontributions to the understanding of complex problems. Leaming requires com-mon codes of communication and coordinated search procedures.

Second, the organizational knowledge generated by such activity resides in newpattems of activity, in 'routines', or a new logic of organization. As indicated ear-iier, routines are patters of interactions that represent successful solutions to par-ticular problems. These pattems of interaction are resident in group behavior,although certain subroutines may be resident in individual behavior.

The concept of dynamic capabilities as a coordinative management processopens the door to the potential for interorganizational leaming. Researchers (Dozand Shuen, 1989; Mody, 1990) have pointed out that collaborations and partner-ships can be vehicles for new organizational leaming, helping firms to recognizedysfunctional routines, and preventing strategic blindspots.

2.3.3 Reconfiguration and Transformation

In rapidly changing environments, there is obviously value in the ability to sensethe need to reconfigure the firm's asset structure, and to accomplish the necessaryintemal and extemal transformation (Amit and Schoemaker, 1993; Langlois,1997). This requires constant surveillance of markets and technologies and thewillingness to adopt best practices. In this regard, benchmarking is of considerablevalue as an organized process for accomplishing such ends (Camp, 1989). In dy-namic environments, narcissistic organizations are likely to be impaired. The ca-pacity to reconfigure and transform is itself a leamed organizational skill. Themore frequently practiced, the more easily accomplished.

Change is costly and so firms must develop processes to minimize low pay-offchange. The ability to calibrate the requirements for change and to effectuate thenecessary adjustments would appear to depend on the ability to scan the environ-ment, to evaluate markets and competitors, and to quickly accomplish reconfigura-tion and transformation ahead of competition. Decentralization and local auton-


omy assists these processes. Firms that have honed these capabilities are some-times referred to as 'high fiex.'

2.4 Positions

The strategic posture of a firm is determined not only by its leaming processes andby the coherence of its intemal and extemal processes and incentives, but also byits location at any point in time with respect to its business assets. By businessassets we do not mean its plant and equipment unless they are specialized; ratherwe mean its difficult-to-trade knowledge assets and assets complementary tothem, as well as its reputational and relational assets. These will determine itsmarket share and profitability at any point in time.

2.4.1 Technological Assets

While there is an emerging market for know-how (Teece, 1981) much technologydoes not enter it. This is either because the firm is unwilling to sell it' or becauseof difficulties in transacting in the market for know-how (Teece, 1980). A firm'stechnological assets may or may not be protected by the standard instmments ofintellectual property law. Either way, the ownership protection and utilization oftechnological assets are clearly key differentiators among firms. Likewise forcomplementary assets.

2.4.2 Complementary Assets

Technological innovations require the use of certain related assets to produce anddeliver new products and services. Prior commercialization activities require andenable firms to build such complementarities (Teece, 1986b). Such capabilitiesand assets, while necessary for the firm's established activities, may have otheruses as well. Such assets typically lie downstream. New products and processescan either enhance or destroy the value of such assets (Tushman et al., 1986).Thus, the development of computers enhanced the value of IBM's direct salesforce in office products, while disk brakes rendered useless much of the auto in-dustries' investment in dmm brakes.

2.4.3 Financial Assets

In the short mn, a firm's cash position and degree of leverage may have strategicimplications. While there is nothing more fungible than cash, it cannot always beraised from extemal markets without the dissemination of considerable informa-tion to potential investors. Accordingly, what a firm can do in short order is oftena function of its balance sheet. In the longer mn, that ought not to be so, as cashfiow ought to be more determinative.

' Managers often evoke the 'crown jevvfels' metaphor. That is, if the technology is released,the kingdom will be lost.


2.4.4 Locational Assets

Geography matters too. Uniqueness in certain businesses can stem from loca-tional assets that are non-tradable (e.g., positioning of a refinery in a certain geo-graphic market). While real estate markets are well developed, land use and envi-ronmental restrictions often make locational assets non-tradable, and hence maybe the source of difficult-to-replicate advantages which manifest themselves inlower transport costs, superior convenience, and the like.

2.5 Paths

2.5.1 Path Dependencies

Where a firm can go is a function of its current position and the paths ahead. It is,of course, also shaped by tbe path bebind. In standard economics textbooks, firmshave an infinite range of technologies from which they can choose and marketsthey can occupy. Changes in product or factor prices will be responded to instan-taneously, with technologies moving in and out according to value maximizationcriteria. Only in the short mn are irreversibilities recognized. Fixed costs - suchas equipment and overheads - cause firms to price below fully amortized costs butnever constrain future investment choices. 'Bygones are bygones.' Path depend-encies are simply not recognized.

The notion of path dependencies recognizes that 'history matters'. Bygones arerarely bygones, despite the predications of rational actor theory. Thus a firm'sprevious investments and its repertoire of routines (its 'history') constrains its fu-ture behavior. Leonard-Barton (1992) notes that an organization's core capabilitiescan just as easily create 'core rigidities.' This follows because leaming tends to belocal. Tbat is, opportunities for leaming will be 'close in' to previous activitiesand thus will be transaction and production specific (Teece, 1988). This is be-cause leaming is often a process of trial, feedback, and evaluation. If too manyparameters are changed simultaneously, the ability of firms to conduct meaningfulnatural quasi experiments is attenuated. If many aspects of a firm's leaming envi-ronment change simultaneously, the ability to ascertain cause-effect relationshipsis confounded because cogtiitive structures will not be formed and rates of leam-ing diminish as a result. One implication is that many investments are muchlonger term than is commonly thought.

2.5.2 Technological Opportunities

The concept of path dependencies can be given forward meaning through the con-sideration of an industry's technological opportunities. It is well recognized thathow far and how fast a particular area of industrial activity can proceed is in partdue to the technological opportunities that lie before it. Such opportunities areusually a lagged function of foment and diversity in basic science, and the rapiditywith which new scientific breakthroughs are being made.

However, technological opportunities may not be completely exogenous to in-dustry, not only because some firms have the capacity to engage in or at least sup-port basic research, but also because technological opportunities are often fed byinnovative activity itself. Moreover, the recognition of such opportunities is af-


fected by the organizational stmctures that link the institutions engaging in basicresearch (primarily the university) to the business enterprise. Hence, the existenceof technological opportunities can be quite firm specific.

Important for our purposes is the rate and direction in which relevant scientificfrontiers are being rolled back. Firms engaging in R&D may find the path deadahead closed off, although breakthroughs in related areas may be sufficiently closeto be attractive. Likewise, if the path dead ahead is extremely attractive, theremay be no incentive for firms to shift the allocation of resources away from tradi-tional pursuits. The depth and width of technological opportunities in theneighborhood of a firm's prior research activities thus are likely to impact a firm'soptions with respect to both the amount and level of R&D activity that it can jus-tify. In addition, a firm's past experience conditions the altematives that man-agement is able to perceive. Thus, not only do firms in the same industry face'menus' with different costs associated with particular technological choices, theyalso are looking at menus containing different choices. This is a critical element inNelson and Winter's (1982) view of firms and technical change.

2.6 Assessment

The assessment of a firm's strategic capability at any point in time is presentedhere as a function of the firm's processes, positions, and paths. What it can do andwhere it can go is thus heavily constrained by the typography of its processes, po-sitions, and paths. Each component of this capability framework needs to be ana-lyzed in a strategic audit.

We submit that if one can identify each of these components and understandtheir interrelationships, one can at least predict the performance of the firm undervarious assumptions about changes in the extemal environment. One can alsoevaluate the richness of the menu of new opportunities from which the firm mayselect, and its likely performance in a changing environment.

The parameters we have identified for determining performance are radicallydifferent from those in the standard textbook theory of the firm, and in the com-petitive forces and strategic confiict approach to strategy (in both, the firm is stilllargely a black box; certainly, little or no attention is given to processes, positions,and paths). Moreover, the agency theoretic view of the firm as a nexus of contractswould put no weight on processes, oppositions, and paths. While agency ap-proaches to the firm may recognize that opportunism and shirking may limit whata firm can do, they do not recognize the opportunities and constraints imposed byprocesses, positions, and paths. Moreover, the firm in our conceptualization ismuch more than the sum of its parts - or a team tied together by contracts (see Al-chian and Demsetz (1972)). Indeed, to some extent individuals can be moved inand out of organizations and, so long as the intemal processes and stmctures re-main in place, performance will not be necessarily impaired. A shift in the envi-ronment is a far more serious threat to the firm than is the loss of key individuals,as individuals can be replaced more readily than organizations can be transformed.Furthermore, the dynamic capabilities view of the firm would suggest that thebehavior and performance of particular firms may be quite hard to replicate, evenif its coherence and rationality are observable. This matter and related issuesinvolving replication and imitation are taken up in the section that follows.


2.7 Replicability and Imitability of Organizational Processesand Positions

Thus far, we have argued that the capabihties of a firm rest on processes, posi-tions, and paths. However, distinctive organizational capabilities can providecompetitive advantage and generate rents if they are based on a collection of rou-tines, skills, and complementary assets that are difficult to imitate (see Dierickxand Cook (1989) for a discussion of the characteristics of assets that make them asource of rents). A particular set of routines can lose their value if they support acompetence that no longer matters in the marketplace, or if they can be readilyreplicated or emulated by competitors. Imitation occurs when firms discover andsimply copy a firm's organizational routines and procedures. Emulation occurswhen firms discover alternative ways of achieving the same functionality. Thereis ample evidence that a given type of competence (e.g., quality) can be supportedby different routines and combinations of skills. For example, the Garvin (1988)and Clark and Fujimoto (1991) studies both indicate that there was no one 'for-mula' for achieving either high quality or high product development performance.

2.7.1 Replication

Replication involves transferring or redeploying competences from one economicsetting to another. Because productive knowledge is embodied, this cannot be ac-complished by simply transmitting infonnation. Only in those instances where allrelevant knowledge if fully codified and understood can replication be collapsedinto a simple problem of information transfer. Too often, the contextual depend-ence of original performance is poorly appreciated, so unless firms have replicatedtheir systems of productive knowledge on many prior occasions, the act of replica-tion is likely to be difficult (Teece, 1976). Indeed, replication and transfer are of-ten impossible absent the transfer of people, though this can be minimized if in-vestments are made to convert tacit knowledge to codified knowledge. Often,however, this is simply not possible.

In short, organizational capabilities, and the routines on which they rest, arenormally rather difficult to replicate. See Gabriel Szulanski's (1993) discussion ofthe intra-firm transfer of best practice. He quotes a senior vice-president of Xeroxas saying "you can see a high performance factory or office, but it just doesn'tspread. I don't know why." Szulanski also discusses the role of benchmarking infacilitating the transfer of best practices.

Even understanding what all the relevant routines are that support a particularcompetence may not be transparent. Indeed, Lippman and Rumelt (1992) haveargued that some sources of competitive advantage are so complex that the firmitself, let alone its competitors, does not understand them; if this is so, it is our be-lief that the firm's advantage is likely to fade, as luck does run out. As Nelson andWinter (1982) and Teece (1982) have explained, many organizational routines arequite tacit in nature. Imitation can also be hindered by the fact that few routinesare stand-alone; coherence may require that a change in one set of routines in onepart of the firm (e.g., production) requires changes in some other part (e.g., R&D).

Some routines and competences seem to be attributable to local or regionalforces that shape firms' capabilities at early stages in their lives. Porter (1990), for


example, shows that differences in local product markets, local factor markets, andinstitutions play an important role in shaping competitive capabilities. Differencesalso exist within populations of firms from the same country. Various studies forthe automobile industry, for example, show that not all Japanese automobile com-panies are top performers in terms of quality, productivity, or product develop-ment (see, for example, Clark and Fujimoto, 1991). The role of firm-specific his-tory has been highlighted as a critical factor explaining such firm-level (as op-posed to regional- or national-level) differences (Nelson and Winter, 1982). Rep-lication in a different context may thus be rather difficult.

At least two types of strategic value fiow from replication. One is the ability tosupport geographic and product line expansion. To the extent that the capabilitiesin question are relevant to customer needs elsewhere, replication can confervalue.' Another is that the ability to replicate also indicates that the firm has thefoundations in place for learning and improvement. Empirical evidence supportsthe notion that the understanding of processes, both in production and in manage-ment, is the key to process improvement (Hayes et al, 1988). In short, an organi-zation cannot improve that which it does not understand. Deep processunderstanding is often required to accomplish codification. Indeed, if knowledgeis highly tacit, it indicates that underlying structures are not well understood,which limits leaming because scientific and engineering principles cannot be assystematically applied. Instead, leaming is confined to proceeding through trialand error, and the leverage that might otherwise come from the application ofscientific theory is denied.

2.7.2 Imitation

Imitation is simply replication performed by a competitor. If self-replication isdifficult, imitation is likely to be even harder. In competitive markets, it is theease of imitation that determines the sustainability of competitive advantage. Easyimitation implies the rapid dissipation of rents.

Factors that make replication difficult also make imitation difficult. Thus, themore tacit the firm's productive knowledge, the harder it is to replicate by the firmitself or its competitors. When the tacit component is high, imitation may well beimpossible, absent the hiring away of key individuals and the transfer of key or-ganizational processes.

However, another set of barriers impeded imitation of certain capabilities inadvanced industrial countries. This is the system of intellectual property rights,such as patents, trade secrets, and trademarks, and even trade dress, which refersto the 'look and feel' of a retail establishment (e.g., the distinctive marketing andpresentation style of The Nature Company). Intellectual property protection is ofincreasing importance in the USA, as since 1982 the legal system has adopted amore pro-patent posture. Similar trends are evident outside the USA. Besides the

Needless to say, there are many examples of firms replicating their capabilities inappro-priately by applying extant routines to circumstances where they may not be applicable, e.g.Nestl6's transfer of developed country marketing methods for infant formula to the thirdworld (Hartley, 1989). A key strategic need is for firms to screen capabilities for their ap-plicability to new environments.


patent system, several other factors cause there to be a difference between replica-tion costs and imitation costs.

The observability of the technology of the organization is one such importantfactor. Whereas vistas into product technology can be obtained through strategiessuch as reverse engineering; this is not the case for process technology, as a firmneed not expose its process technology to the outside in order to benefit from it.An interesting but important exception to this can be found in second sourcing. Inthe microprocessor business, until the introduction of the 386 chip, Intel and mostother merchant semi producers were encouraged by large customers like IBM toprovide second sources (i.e., to license and share their proprietary process tech-nology with competitors like AMD and NEC). The microprocessor developers didso to assure customers that they had sufficient manufacturing capability to meetdemand at all times. Firms with product technology, on the other hand, confrontthe unfortunate circumstances that they must expose what they have got in orderto profit from the technology. Secrets are thus more protectable if there is no needto expose them in contexts where competitors can leam about them.

One should not, however, overestimate the overall importance of intellectualproperty protection; yet it presents a formidable imitation barrier in certain par-ticular contexts. Intellectual property protection is not uniform across products,processes, and technologies, and is best thought of as an island in a sea of opencompetition. If one is not able to place the fi^its of one's investment, ingenuity,or creativity on one or more of the islands, then one indeed is at sea.

We use the term 'appropriability regimes' to describe the ease of imitation.Appropriability is a function both of the ease of replication and the efficacy of in-tellectual property rights as a barrier to imitation. Appropriability is strong when atechnology is both inherently difficult to replicate and the intellectual propertysystem provides legal barriers to imitation. When it is inherently easy to replicateand intellectual property protection is either unavailable or ineffectual, then ap-propriability is weak. Intermediate conditions also exist (see Figure 1).

;hts

Pro

•a

lect

unt

el .2H

Inherent Replicability

Easy

WEAK

MODERATE

Hard

MODERATE

STRONG

Figure 1. Appropriability Regimes


2.8 Strategic Issues from a Dynamic Capabilities Perspective

The dynamic capabilities approach views competition in Schumpeterian terms.This means, at one level, that firms compete on the basis of product design, prod-uct quality, process efficiency, and other attributes. However, in a Schumpeterianworld, firms are constantly seeking to create 'new combinations', and rivals arecontinuously attempting to improve their competences or to imitate the compe-tence of their most qualified competitors. Rivahy to develop new competences orto improve existing ones is critical in a Schumpeterian world. Such processesdrive creative destruction. Differences in firms' capabilities to improve their dis-tinctive competences or to develop new distinctive domains of competence play acritical role in shaping long-term competitive outcomes.

The strategic problem facing an innovating firm in a world of Schumpeteriancompetition is to decide upon and develop difficult-to-imitate processes and pathsmost likely to support valuable products and services. Thus, as argued byDierickx and Cool (1989), choices about how much to spend (invest) on differentpossible areas are central to the firm's strategy. However, choices about domainsof competence are infiuenced by past choices. At any given point in time, firmsmust follow a certain trajectory or path of competence development. This path notonly defines what choices are open to the firm today, but it also puts boundariesaround what its repertoire is likely to be in the future. Thus, firms, at variouspoints in time, make long-term, quasi-irreversible commitments to certain do-mains of competence. Deciding, under significant uncertainty about future statesof the world, which long-term paths to commit to and when to change paths is thecentral strategic problem confronting the firm. In this regard, the work of Ghe-mawat (1991) is highly germane to the dynamic capabilities approach to strategy.

3 Conclusion

We posit that the competitive advantage of firms stem from dynamic capabilitiesrooted in high performance routines operating inside the firm, embedded in thefirm's processes, and conditioned by its history. Because of imperfect factor mar-kets, or more precisely the non-tradability of 'soft' assets like values, culture, andorganizational experience, these capabilities generally cannot be bought; theymust be built. This may take years - possibly decades. In some cases, as whenthe competence is protected by patents, imitation by a competitor is illegal as ameans to access the technology. The capabilities approach accordingly sees defi-nite limits on strategic options, at least in the short run. Competitive success oc-curs in part because of processes and structures already established and experienceobtained in earlier periods.

The notion that competitive success arises from the continuous development,exploitation, and protection of firm-specific assets, while not the dominant view inindustrial organization, nevertheless has a long tradition going back at least toSchumpeter. Schumpeters, in his Theory of Economic Development (1934), saweconomic development as consisting of a process where entrepreneurs dipped intoa stream of technical opportunities ostensibly made for reasons independent of


particular markets and brought those innovations to market. The successful inno-vator achieved a monopoly in a particular market through bringing to marketsomething which was quite unique, only to have that monopoly successfully whit-tled away by the entry (swarming) of imitators. The dynamic capabilities ap-proach is a descendant of the Schumpeterian. However, it emphasizes organiza-tional processes inside the firm more than Schumpeter ever did; and it is not just apositive theory of industrial change. It can also offer prescription because of itsfirm-level orientation, and it looks inside firms to help explain market processes.

Because it is hard to transform organizational processes, the dynamic capabili-ties approach sees value augmenting strategic change as being difficult and costly.Moreover, it can generally occtir only incrementally. Because capabilities cannoteasily be bought and must be built,' opportunities for growth from diversificationare thus likely to be limited, lying 'close in' to the firm's existing lines of product(Rumelt, 1974; Teece et al, 1994). In attempting to explicate competitive advan-tage, the dynamic capabilities emphasis on the firm's intemal processes, assets andmarket positions, the path along which it has traveled, and the paths that lie ahead.The framework also exphcitly takes into account replicability and imitability.

We offer dynamic capabiiities as an emerging paradigm of the modem businessfirm. It is an eclectic paradigm drawing from multiple disciplines, and advancingwith the help of industry studies in the USA and elsewhere. There are, of course,a wide variety of theories of the firm, each sometimes highlighting a different as-pect. Thus, transaction cost economics highlights boundaries, agency theory high-lights incentives and control, and the production function highlights the role offixed factors. It appears that the dynamic capabilities approach is seeking attentionby promising to explain matters such as the limits of diversification, the feasibilityof 'converting' firms from military to civilian purposes, the adaptability of somefirms and the intransigence of others, etc. Perhaps a decade from now we will beable to assess whether the promise has been honored, and whether as a conse-quence the fields of industrial organization and business strategy can help us sol-idly come to grips with the challenges of our times.

Acknowledgements

We would like to thank Amy Shuen for useful comments. This introduction drawsfrom D. J. Teece, G. Pisano, and A. Shuen, 'Dynamic Capabilities and StrategicManagement,' CCC Working paper #94-9, University of Califomia, Berkeley(August, 1994).

' Robert Hayes (1985) has noted that American companies tend to favor 'strategic leaps',while, in contrast, Japanese and German companies tend to favor incremental, but rapid,improvements. If this is correct, it seems to indicate that the Japanese and German manag-ers more fully recognize the validity of the dynamic capabilities framework than do theirAmerican counterparts.


Appendix

Fujimoto (1994, pp. 18-20) describes key elements as they existed in the Japaneseauto industry as follows: "The typical volume production system of effectiveJapanese makers of the 198O's (e.g., Toyota) consists of various intertwined ele-ments that might lead to competitive advantages. Just-in-Time (JIT), Jidoka(automatic defect detection and machine stop). Total Quality Control (TQC), andcontinuous improvement (Kaizen) are often pointed out as its core subsystems.The elements of such a system include inventory reduction mechanisms by Kan-ban system; levelization of production volume and product mix (heijunka); reduc-tion of 'muda' (non-value adding activities), 'mura' (uneven pace of production)and muri (excessive workload); production plans based on dealers' order volume(genyo seisan); reduction of die set-up time and lot size in stamping operation;mixed model assembly; piece by piece transfer of parts between machines (ikko-nagashi), fiexible task assignment for volume changes and productivity improve-ment (shojinka); multi-task job assignment along the process fiow (takorei-mochi); U-shape machine layout that facilitates fiexible and multiple task assign-ment, on-the-spot inspection by direct workers (tsukurikomi); fool-proof preven-tion of defects (poka-yoke); real-time feedback of production troubles (andon);assembly line stop cord; emphasis on cleanliness, order, and discipline on the shopfioor (5-S); frequent revision of standard operating procedures by supervisors;quality control circles; standardized tools for quality improvement (e.g., 7 tools forQC, QC story); workers involvement in preventive maintenance (Total ProductiveMaintenance); low cost automation or semi-automation with just-enough func-tions; reduction of process steps for saving of tools and dies, and so on. The hu-man-resource management factors that back up the above elements include stableemployment of core workers (with temporary workers in the periphery); long-termtraining of multi-skilled (multi-task) workers; wage system based in part on skillaccumulation; intemal promotion to shop fioor supervisors; cooperative relation-ships with labor unions; inclusion of production supervisors in union members;generally egalitarian policies for corporate welfare, communication and workermotivation. Parts procurement policies are also pointed out often as a source ofdie competitive advantage; relatively high ratio of parts out-sourcing; multi-layerhierarchy of suppliers; long-term relations with suppliers; relatively small numberof technologically capable suppliers at the first tier; subassembly functions of thefirst-tier parts markers; detail-engineering capability of the first tier makers (de-sign-in, back box parts); competition based on long-term capability of design andimprovements rather than bidding; pressures for continuous reduction of partsprice; elimination of incoming parts inspection; plant inspection and technical as-sistance by auto makers, and so on."

References

Abemathy, W. J. and K. Clark, "Innovation: Mapping the Winds of Creative De-stmction," Research Policy, 14, 1985, 3-22.


Alchian, A. A. and H. Demsetz, "Production, Information Costs, and EconomicOrganization," American Economic Review, 62, 1972, 777-795.

Amit, R. and P. Schoemaker, "Strategic Assets and Organizational Rent," Strate-gic Management Joumal, 14, 8, 1993, 33-46.

Aoki, M., "The Participatory Generation of Information Rents and the Theory ofthe Firm," in Aoki, M.; et al. (eds.). The Firm as a Nexus of Treaties, London:Sage, 1990.

Arrow, K., "The Organization of Economic Activity: Issues Pertinent to theChoice of Market vs. Nonmarket Allocation," in The Analysis and Evaluationof Public Expenditures: The PPB System, 1, US Joint Economic Committee,91" Session, Washington: US Govemment Printing Office, 1969, 59-73.

Baldwin, C. and K. Clark, "Capabilities and Capital Investment: New Perspectiveson Capital Budgeting," Harvard Business School, Working Paper 92-004, 1991.

Bamey, J. B., "Strategic Factor Markets: Expectations, Luck, and Business Strat-egy," Management 5dence, 32,1986, 1231-1241.

Camp, R., Benchmarking: The Search for Industry Best Practice That Lead to Su-perior Performanc, New York: Quality Resources, 1989.

Chandler, A. D., Jr., Scale and Scope: The Dynamics of Industrial Competition,Cambridge, MA: Harvard University Press, 1990.

Clark, K. and T. Fujimoto, Product Development Performance: Strategy, Organi-zation and Management in the World Auto Industries, Cambridge, MA: Har-vard Business School Press, 1991.

Coase, R., "The Nature of the Firm," Economica, November, 1937, 386-495.

Coase, R., "Lecture on the Nature of the Firm, in," Joumal of Law, Economicsand Organization, 4, 1988, 33-47.

Dierickx, I. and K. Cool, "Asset Stock Accumulation and Sustainability of Com-petitive Advantage," Management Science, 35, 1989, 1504-1511.

Doz, Y. and A. Shuen, "From Intent to Outcome: A Process Framework for Part-nerships," INSEAD Working Paper, 1990.

Fama, E. F., "Agency Problems and the Theory of the Firm," Joumal of PoliticalEconomy, 88, 1980, 288-307.

Fujimoto, T., "Reinterpreting the Resource-Capability View of the Firm: A Caseof the Development-Production System of the Japanese Automakers," Draftworking paper. Faculty of Economics, University of Tokyo, May, 1994.

Garvin, D., Managing Quality, New York: The Free Press, 1988.

Garvin, D., "The Processes of Organization and Management," Sloan Manage-ment Review, 39,4, 1998, 33-50.

Ghemawat, P., Commitment: The Dynamics of Strategy, New York: The FreePress, 1991.

Hartley, R. F., Marketing Mistakes, New York: John Wiley, 1989.


Hayes, R., "Strategic Planning: Forward in Reverse," Harvard Business Review,November - December, 1985, 111-119.

Hayes, R., S. Wheelwright, and K. Clark, Dynamic Manufacturing: Creating theLeaming Organization, New York: The Free Press, 1988.

Henderson, R. M., "The Evolution of Integrative Capability: Innovation in Car-diovascular Dmg Discovery," Industrial and Corporate Change, 3, 1995, 607-630.

Henderson, R. M. and K. B. Clark, "Architectural Innovation: The Reconfigura-tion of Existing Product Technologies and the Failure to Established Firms,"Administrative Science Quarterly, 35, 1990, 9-30.

Iansiti, M. and K. B. Clark, "Integration and Dynamic Capability: Evidence fromProduct Development in Automobiles and Mainframe Computers," Industrialand Corporate Change, 3, 1995, 557-605.

Kogut, I. and U. Zander, "Knowledge of the Firm, Combinative Capabilities, andReplication of Technology," Organization Science, 3, 3, 1992, 383-397.

Langlois, R., "Cognition and Capabilities: Opportunities Seized and Missed in theHistory of the Computer Industry," in Gamd, R..; Nayyar, P. and Shapira, Z.(eds.). Technological Innovation: Oversights and Foresights, New York: Cam-bridge University Press, 1997.

Leonard-Barton, D., "Core Capabilities and Core Rigidities: A Paradox in Manag-ing New Product Development," Strategic Management Joumal, 13, 1992,111-125.

Lev, B. and T. Sougiannis, "The Capitalization, Amortization and Value-Relevance of R&D," Joumal of Accounting and Economics, 21, Febmary,1996,107-138.

Levitt, B. and J. March, "Organizational Leaming," Annual Review of Sociology,14,1988, 319-340.

Lippman, S. A. and R. P. Rumelt, "Demand Uncertainty and Investment in Indus-try-Specific Capital," Industrial and Corporate Change, 1, 1992, 235-262.

Mahoney, J., "The Management of Resources and the Resources of Management,"Joumal of Business Research, 33, 1995, 91-101.

Mody, A., "Leaming through Alliances," Working Paper, The World Bank, Wash-ington, DC, September 6, 1990.

Nelson, R. and S. Winter, An Evolutionary Theory of Economic Change, Cam-bridge MA: Harvard University Press, 1982.

Penrose, E., The Theory of the Growth of the Firm, London: Basil Blackwell, 1959.

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

Prahalad, C. K. and G. Hamel, "The Core Competence of the Corporation," Har-vard Business Review, May-June, 1990, 79-91.

Rumelt, R. P., Strategy, Structure, and Economic Performance, Cambridge, MA:Harvard University Press, 1974.


Schumpeter, J. A., Theory of Economic Development, Cambridge, MA: HarvardUniversity Press, 1934.

Shuen, A., "Technology Sourcing and Leaming Strategies in the SemiconductorIndustry," Unpublished PhD dissertation. University of Califomia, Berkeley,1994.

Szulanski, G., "Intrafirm Transfer of Best Practice, Appropriate Capabilities, Or-ganizational Barriers to Appropriation," Working Paper, INSEAD, March,1993.

Teece, D. J., The Multinational Corporation and the Resource Cost of Interna-tional Technology Transfer, Cambridge: Ballinger, 1976.

Teece, D. J., "Economics of Scope and the Scope of an Enterprise," Joumal ofEconomic Behavior and Organization, 1, 1980,223-247.

Teece, D. J., "Towards and Economic Theory of the Multiproduct Firm," Joumalof Economic Behavior and Organization, 3, 1982, 39-63.

Teece, D. J., "Transactions Cost Economics and the Multinational Enterprise,"Joumal of Economic Behavior and Organization, 7, 1986a, 21-45.

Teece, D. J., "Profiting from Technological Innovation," Research Policy, 15,1986b.

Teece, D. J., "Technological Cbange and the Nature of the Firm," in G. Dosi, G.;et al. (eds.). Technical Change and Economic Theory, London: Pinter, 1988.

Teece, D. J., "The Dynamics of Industrial Capitalism: Perspectives on AlfredChandler's Scale and Scope (1990)," Joumal of Economic Literature, 31,1993.

Teece, D. J., R. Rumelt, G. Dosi, and S. Winter, "Understanding Corporate Co-herence: Theory and Evidence," Joumal of Economic Behavior and Organiza-tion, 23, 1994, 1-30.

Tushman, M. L., W. H. Newman, and E. Romanelli, "Convergence and Upheaval:Managing the Unsteady Pace of Organizational Evolution," Califomia Man-agement Review, 29, 1986, 29-44.

Wemerfelt, B., "A Resource-Based View of the Firm," Strategic ManagementyoMma/,5, 1984, 171-180.

Williamson, O. E., Markets and Hierarchies, New York: The Free Press, 1975.

Williamson, O. E., The Economic Institutions of Capitalism, New York: The FreePress, 1985.

Womack, J., D. Jones, and D. Roos, The Machine That Changed the World, NewYork: Harper-Perennial, 1991.

Documents

CHAPTER 42 The Dynamic Capabilities of Firms*cmap.upb.edu.co › rid=1T319V6NG-29P7SD1-1RM › Teece y...The Dynamic Capabilities of Firms 197 specific assets. Each approach asks different,