10. teece pisano dynamic capabilities

Strategic Management Journal, Vol. 18:7, 509–533 (1997)
DYNAMIC CAPABILITIES AND STRATEGIC
MANAGEMENT
DAVID J. TEECE1
*, GARY PISANO2
and AMY SHUEN3
1
Haas School of Business, University of California, Berkeley, California, U.S.A.
2
Graduate School of Business Administration, Harvard University, Boston, Massa-
chusetts, U.S.A.
3
School of Business, San Jose State University, San Jose, California, U.S.A.
The dynamic capabilities framework analyzes the sources and methods of wealth creation and
capture by private enterprise firms operating in environments of rapid technological change.
The competitive advantage of firms is seen as resting on distinctive processes (ways of
coordinating and combining), shaped by the firm’s (specific) asset positions (such as the firm’s
portfolio of difficult-to-trade knowledge assets and complementary assets), and the evolution
path(s) it has adopted or inherited. The importance of path dependencies is amplified where
conditions of increasing returns exist. Whether and how a firm’s competitive advantage is
eroded depends on the stability of market demand, and the ease of replicability (expanding
internally) and imitatability (replication by competitors). If correct, the framework suggests
that private wealth creation in regimes of rapid technological change depends in large measure
on honing internal technological, organizational, and managerial processes inside the firm. In
short, identifying new opportunities and organizing effectively and efficiently to embrace them
are generally more fundamental to private wealth creation than is strategizing, if by strategizing
one means engaging in business conduct that keeps competitors off balance, raises rival’s
costs, and excludes new entrants. © 1997 by John Wiley & Sons, Ltd.
INTRODUCTION respect to assisting in the understanding of how
and why certain firms build competitive advan-
tage in regimes of rapid change. Our approach isThe fundamental question in the field of strategic
management is how firms achieve and sustain especially relevant in a Schumpeterian world of
innovation-based competition, price/performancecompetitive advantage.1
We confront this question
here by developing the dynamic capabilities rivalry, increasing returns, and the ‘creative
destruction’ of existing competences. Theapproach, which endeavors to analyze the sources
of wealth creation and capture by firms. The approach endeavors to explain firm-level success
and failure. We are interested in both building adevelopment of this framework flows from a
recognition by the authors that strategic theory is better theory of firm performance, as well as
informing managerial practice.replete with analyses of firm-level strategies for
sustaining and safeguarding extant competitive In order to position our analysis in a manner
that displays similarities and differences withadvantage, but has performed less well with
existing approaches, we begin by briefly
reviewing accepted frameworks for strategic man-
Key words: competences; capabilities; innovation;
agement. We endeavor to expose implicit assump-strategy; path dependency; knowledge assets
tions, and identify competitive circumstances
*Correspondence to: David J. Teece, Institute of Management, where each paradigm might display some relative
Innovation and Organization, Haas School of Business, Uni- advantage as both a useful descriptive and norma-
versity of California, Berkeley, CA 94720–1930, U.S.A.
tive theory of competitive strategy. While numer-1
For a review of the fundamental questions in the field of
strategy, see Rumelt, Schendel, and Teece (1994). ous theories have been advanced over the past
CCC 0143–2095/97/070509–25$17.50 Received 17 April 1991
© 1997 by John Wiley & Sons, Ltd. Final revision received 4 March 1997

510 D. J. Teece, G. Pisano and A. Shuen
two decades about the sources of competitive recognizes but does not attempt to explain the
nature of the isolating mechanisms that enableadvantage, many cluster around just a few loosely
structured frameworks or paradigms. In this paper entrepreneurial rents and competitive advantage
to be sustained.we attempt to identify three existing paradigms
and describe aspects of an emerging new para- Another component of the efficiency-based
approach is developed in this paper. Rudimentarydigm that we label dynamic capabilities.
The dominant paradigm in the field during efforts are made to identify the dimensions of
firm-specific capabilities that can be sources ofthe 1980s was the competitive forces approach
developed by Porter (1980). This approach, advantage, and to explain how combinations of
competences and resources can be developed,rooted in the structure–conduct–performance
paradigm of industrial organization (Mason, 1949; deployed, and protected. We refer to this as the
‘dynamic capabilities’ approach in order to stressBain, 1959), emphasizes the actions a firm can
take to create defensible positions against com- exploiting existing internal and external firm-
specific competences to address changingpetitive forces. A second approach, referred to as
a strategic conflict approach (e.g., Shapiro, 1989), environments. Elements of the approach can be
found in Schumpeter (1942), Penrose (1959),is closely related to the first in its focus on
product market imperfections, entry deterrence, Nelson and Winter (1982), Prahalad and Hamel
(1990), Teece (1976, 1986a, 1986b, 1988) andand strategic interaction. The strategic conflict
approach uses the tools of game theory and thus in Hayes, Wheelwright, and Clark (1988):
Because this approach emphasizes the develop-implicitly views competitive outcomes as a func-
tion of the effectiveness with which firms keep ment of management capabilities, and difficult-
to-imitate combinations of organizational, func-their rivals off balance through strategic invest-
ments, pricing strategies, signaling, and the con- tional and technological skills, it integrates and
draws upon research in such areas as the manage-trol of information. Both the competitive forces
and the strategic conflict approaches appear to ment of R&D, product and process development,
technology transfer, intellectual property, manu-share the view that rents flow from privileged
product market positions. facturing, human resources, and organizational
learning. Because these fields are often viewedAnother distinct class of approaches empha-
sizes building competitive advantage through cap- as outside the traditional boundaries of strategy,
much of this research has not been incorporatedturing entrepreneurial rents stemming from funda-
mental firm-level efficiency advantages. These into existing economic approaches to strategy
issues. As a result, dynamic capabilities can beapproaches have their roots in a much older
discussion of corporate strengths and weaknesses; seen as an emerging and potentially integrative
approach to understanding the newer sources ofthey have taken on new life as evidence suggests
that firms build enduring advantages only through competitive advantage.
We suggest that the dynamic capabilitiesefficiency and effectiveness, and as developments
in organizational economics and the study of approach is promising both in terms of future
research potential and as an aid to managementtechnological and organizational change become
applied to strategy questions. One strand of this endeavoring to gain competitive advantage in
increasingly demanding environments. To illus-literature, often referred to as the ‘resource-based
perspective,’ emphasizes firm-specific capabilities trate the essential elements of the dynamic capa-
bilities approach, the sections that follow compareand assets and the existence of isolating mech-
anisms as the fundamental determinants of firm and contrast this approach to other models of
strategy. Each section highlights the strategicperformance (Penrose, 1959; Rumelt, 1984;
Teece, 1984; Wernerfelt, 1984).2
This perspective
how. Over time, these assets may expand beyond the point
of profitable reinvestment in a firm’s traditional market.2
Of these authors, Rumelt may have been the first to self-
consciously apply a resource perspective to the field of strat- Accordingly, the firm may consider deploying its intangible
assets in different product or geographical markets, where theegy. Rumelt (1984: 561) notes that the strategic firm ‘is
characterized by a bundle of linked and idiosyncratic resources expected returns are higher, if efficient transfer modes exist.’
Wernerfelt (1984) was early to recognize that this approachand resource conversion activities.’ Similarly, Teece (1984:
95) notes: ‘Successful firms possess one or more forms of was at odds with product market approaches and might consti-
tute a distinct paradigm of strategy.intangible assets, such as technological or managerial know-

Dynamic Capabilities 511
insights provided by each approach as well as (1980). Competitive strategies are often aimed at
altering the firm’s position in the industry vis-a`-the different competitive circumstances in which
it might be most appropriate. Needless to say, vis competitors and suppliers. Industry structure
plays a central role in determining and limitingthese approaches are in many ways complemen-
tary and a full understanding of firm-level, com- strategic action.
Some industries or subsectors of industriespetitive advantage requires an appreciation of all
four approaches and more. become more ‘attractive’ because they have struc-
tural impediments to competitive forces (e.g.,
entry barriers) that allow firms better oppor-
tunities for creating sustainable competitiveMODELS OF STRATEGY
EMPHASIZING THE EXPLOITATION advantages. Rents are created largely at the indus-
try or subsector level rather than at the firm level.OF MARKET POWER
While there is some recognition given to firm-
Competitive forces
specific assets, differences among firms relate
primarily to scale. This approach to strategyThe dominant paradigm in strategy at least during
the 1980s was the competitive forces approach. reflects its incubation inside the field of industrial
organization and in particular the industrial struc-Pioneered by Porter (1980), the competitive
forces approach views the essence of competitive ture school of Mason and Bain3
(Teece, 1984).
strategy formulation as ‘relating a company to its
environment . . . [T]he key aspect of the firm’s
Strategic conflict
environment is the industry or industries in which
it competes.’ Industry structure strongly influ- The publication of Carl Shapiro’s 1989 article,
confidently titled ‘The Theory of Businessences the competitive rules of the game as well
as the strategies potentially available to firms. Strategy,’ announced the emergence of a new
approach to business strategy, if not strategicIn the competitive forces model, five industry-
level forces—entry barriers, threat of substitution, management. This approach utilizes the tools of
game theory to analyze the nature of competitivebargaining power of buyers, bargaining power
of suppliers, and rivalry among industry interaction between rival firms. The main thrust
of work in this tradition is to reveal how a firmincumbents—determine the inherent profit poten-
tial of an industry or subsegment of an industry. can influence the behavior and actions of rival
firms and thus the market environment.4
The approach can be used to help the firm find
a position in an industry from which it can Examples of such moves are investment in
capacity (Dixit, 1980), R&D (Gilbert and New-best defend itself against competitive forces or
influence them in its favor (Porter, 1980: 4). berry, 1982), and advertising (Schmalensee,
1983). To be effective, these strategic movesThis ‘five-forces’ framework provides a sys-
tematic way of thinking about how competitive require irreversible commitments.5
The moves in
question will have no effect if they can beforces work at the industry level and how these
forces determine the profitability of different costlessly undone. A key idea is that by manipu-
lating the market environment, a firm may beindustries and industry segments. The competitive
forces framework also contains a number of able to increase its profits.
underlying assumptions about the sources of com-
petition and the nature of the strategy process. 3
In competitive environments characterized by sustainable and
stable mobility and structural barriers, these forces mayTo facilitate comparisons with other approaches,
become the determinants of industry-level profitability. How-we highlight several distinctive characteristics of
ever, competitive advantage is more complex to ascertain in
the framework. environments of rapid technological change where specific
assets owned by heterogeneous firms can be expected to playEconomic rents in the competitive forces
a larger role in explaining rents.framework are monopoly rents (Teece, 1984). 4
The market environment is all factors that influence market
Firms in an industry earn rents when they are outcomes (prices, quantities, profits) including the beliefs of
customers and of rivals, the number of potential technologiessomehow able to impede the competitive forces
employed, and the costs or speed with which a rival can(in either factor markets or product markets)
enter the industry.
which tend to drive economic returns to zero. 5
For an excellent discussion of committed competition in
multiple contexts, see Ghemawat (1991).Available strategies are described in Porter

This literature, together with the contestability which the strategic conflict literature is relevant
to strategic management. Firms that have aliterature (Baumol, Panzar, and Willig, 1982), has
led to a greater appreciation of the role of sunk tremendous cost or other competitive advantage
vis-a`-vis their rivals ought not be transfixed bycosts, as opposed to fixed costs, in determining
competitive outcomes. Strategic moves can also the moves and countermoves of their rivals. Their
competitive fortunes will swing more on totalbe designed to influence rivals’ behavior through
signaling. Strategic signaling has been examined demand conditions, not on how competitors
deploy and redeploy their competitive assets. Putin a number of contexts, including predatory
pricing (Kreps and Wilson, 1982a, 1982b) and differently, when there are gross asymmetries in
competitive advantage between firms, the resultslimit pricing (Milgrom and Roberts, 1982a,
1982b). More recent treatments have emphasized of game-theoretic analysis are likely to be obvious
and uninteresting. The stronger competitor willthe role of commitment and reputation (e.g.,
Ghemawat, 1991) and the benefits of firms simul- generally advance, even if disadvantaged by cer-
tain information asymmetries. To be sure, incum-taneously pursuing competition and cooperation6
(Brandenburger and Nalebuff, 1995, 1996). bent firms can be undone by new entrants with
a dramatic cost advantage, but no ‘gaming’ willIn many instances, game theory formalizes
long-standing intuitive arguments about various overturn that outcome. On the other hand, if
firms’ competitive positions are more delicatelytypes of business behavior (e.g., predatory pric-
ing, patent races), though in some instances it has balanced, as with Coke and Pepsi, and United
Airlines and American Airlines, then strategicinduced a substantial change in the conventional
wisdom. But by rationalizing observed behavior conflict is of interest to competitive outcomes.
Needless to say, there are many such circum-by reference to suitably designed games, in
explaining everything these models also explain stances, but they are rare in industries where
there is rapid technological change and fast-shift-nothing, as they do not generate testable predic-
tions (Sutton, 1992). Many specific game- ing market circumstances.
In short, where competitors do not have deep-theoretic models admit multiple equilibrium, and
a wide range of choice exists as to the design of seated competitive advantages, the moves and
countermoves of competitors can often be use-the appropriate game form to be used. Unfortu-
nately, the results often depend on the precise fully formulated in game-theoretic terms. How-
ever, we doubt that game theory can comprehen-specification chosen. The equilibrium in models
of strategic behavior crucially depends on what sively illuminate how Chrysler should compete
against Toyota and Honda, or how United Air-one rival believes another rival will do in a
particular situation. Thus the qualitative features lines can best respond to Southwest Airlines since
Southwest’s advantage is built on organizationalof the results may depend on the way price
competition is modeled (e.g., Bertrand or attributes which United cannot readily replicate.8
Indeed, the entrepreneurial side of strategy—howCournot) or on the presence or absence of stra-
tegic asymmetries such as first-mover advantages. significant new rent streams are created and
protected—is largely ignored by the game-The analysis of strategic moves using game
theory can be thought of as ‘dynamic’ in the theoretic approach.9
Accordingly, we find that
the approach, while important, is most relevantsense that multiperiod analyses can be pursued
both intuitively and formally. However, we use
the term ‘dynamic’ in this paper in a different
sense, referring to situations where there is rapid
8
Thus even in the air transport industry game-theoretic formu-change in technology and market forces, and
lations by no means capture all the relevant dimensions of
‘feedback’ effects on firms.7
competitive rivalry. United Airlines’ and United Express’s
difficulties in competing with Southwest Airlines because ofWe have a particular view of the contexts in
United’s inability to fully replicate Southwest’s operation
capabilities is documented in Gittel (1995).
9
Important exceptions can be found in Brandenburger and
Nalebuff (1996) such as their emphasis on the role of com-6
Competition and cooperation have also been analyzed ouside
of this tradition. See, for example, Teece (1992) and Link, plements. However, these insights do not flow uniquely from
game theory and can be found in the organizational economicsTeece and Finan (1996).
7
Accordingly, both approaches are dynamic, but in very literature (e.g., Teece, 1986a, 1986b; de Figueiredo and
Teece, 1996).different senses.

when competitors are closely matched10
and the that may deter entry and raise prices above long-
run costs, but because they have markedly lowerpopulation of relevant competitors and the iden-
tity of their strategic alternatives can be readily costs, or offer markedly higher quality or product
performance. This approach focuses on the rentsascertained. Nevertheless, coupled with other
approaches it can sometimes yield powerful accruing to the owners of scarce firm-specific
resources rather than the economic profits frominsights.
However, this research has an orientation that product market positioning.12
Competitive advan-
tage lies ‘upstream’ of product markets and restswe are concerned about in terms of the implicit
framing of strategic issues. Rents, from a game- on the firm’s idiosyncratic and difficult-to-
imitate resources.13
theoretic perspective, are ultimately a result of
managers’ intellectual ability to ‘play the game.’ One can find the resources approach suggested
by the earlier preanalytic strategy literature. AThe adage of the strategist steeped in this
approach is ‘do unto others before they do unto leading text of the 1960s (Learned et al., 1969)
noted that ‘the capability of an organization is itsyou.’ We worry that fascination with strategic
moves and Machiavellian tricks will distract man- demonstrated and potential ability to accomplish
against the opposition of circumstance or compe-agers from seeking to build more enduring
sources of competitive advantage. The approach tition, whatever it sets out to do. Every organiza-
tion has actual and potential strengths and weak-unfortunately ignores competition as a process
involving the development, accumulation, combi- nesses; it is important to try to determine what
they are and to distinguish one from the other.’nation, and protection of unique skills and capa-
bilities. Since strategic interactions are what Thus what a firm can do is not just a function
of the opportunities it confronts; it also dependsreceive focal attention, the impression one might
receive from this literature is that success in the on what resources the organization can muster.
Learned et al. proposed that the real key to amarketplace is the result of sophisticated plays
and counterplays, when this is generally not the company’s success or even to its future develop-
ment lies in its ability to find or create ‘a com-case at all.11
In what follows, we suggest that building a petence that is truly distinctive.’14
This literature
also recognized the constraints on firm behaviordynamic view of the business enterprise—
something missing from the two approaches we and, in particular, noted that one should not
assume that management ‘can rise to anyhave so far identified—enhances the probability
of establishing an acceptable descriptive theory occasion.’ These insights do appear to keenly
anticipate the resource-based approach that hasof strategy that can assist practitioners in the
building of long-run advantage and competitive since emerged, but they did not provide a theory
or systematic framework for analyzing businessflexibility. Below, we discuss first the resource-
based perspective and then an extension we call strategies. Indeed, Andrews (1987: 46) noted that
‘much of what is intuitive in this process isthe dynamic capabilities approach.
yet to be identified.’ Unfortunately, the academic
literature on capabilities stalled for a couple of
decades.MODELS OF STRATEGY
EMPHASIZING EFFICIENCY New impetus has been given to the resource-
based approach by recent theoretical develop-
Resource-based perspective
ments in organizational economics and in the
theory of strategy, as well as by a growingThe resource-based approach sees firms with
superior systems and structures being profitable
not because they engage in strategic investments
12
In the language of economics, rents flow from unique firm-
specific assets that cannot readily be replicated, rather than
from tactics which deter entry and keep competitors off10
When closely matched in an aggregate sense, they may
nevertheless display asymmetries which game theorists can balance. In short, rents are Ricardian.
13
Teece (1982: 46) saw the firm as having ‘a variety of endanalyze.
11
The strategic conflict literature also tends to focus prac- products which it can produce with its organizational tech-
nology.’titioners on product market positioning rather than on
developing the unique assets which make possible superior 14
Elsewhere Andrews (1987: 47) defined a distinctive com-
petence as what an organization can do particularly well.product market positions (Dierickx and Cool, 1989).

body of anecdotal and empirical literature15
that process.18
Quite simply, firms lack the organiza-
tional capacity to develop new competenceshighlights the importance of firm-specific factors
in explaining firm performance. Cool and Schen- quickly (Dierickx and Cool, 1989). Secondly,
some assets are simply not readily tradeable, fordel (1988) have shown that there are systematic
and significant performance differences among example, tacit know-how (Teece, 1976, 1980)
and reputation (Dierickx and Cool, 1989). Thus,firms which belong to the same strategic group
within the U.S. pharmaceutical industry. Rumelt resource endowments cannot equilibrate through
factor input markets. Finally, even when an asset(1991) has shown that intraindustry differences
in profits are greater than interindustry differences can be purchased, firms may stand to gain little by
doing so. As Barney (1986) points out, unless ain profits, strongly suggesting the importance of
firm-specific factors and the relative unimportance firm is lucky, possesses superior information, or
both, the price it pays in a competitive factorof industry effects.16
Jacobsen (1988) and Hansen
and Wernerfelt (1989) made similar findings. market will fully capitalize the rents from the asset.
Given that in the resources perspective firmsA comparison of the resource-based approach
and the competitive forces approach (discussed possess heterogeneous and sticky resource
bundles, the entry decision process suggested byearlier in the paper) in terms of their implications
for the strategy process is revealing. From the this approach is as follows: (1) identify your
firm’s unique resources; (2) decide in which mar-first perspective, an entry decision looks roughly
as follows: (1) pick an industry (based on its kets those resources can earn the highest rents;
and (3) decide whether the rents from those assets‘structural attractiveness’); (2) choose an entry
strategy based on conjectures about competitors’ are most effectively utilized by (a) integrating
into related market(s), (b) selling the relevantrational strategies; (3) if not already possessed,
acquire or otherwise obtain the requisite assets to intermediate output to related firms, or (c) selling
the assets themselves to a firm in related busi-compete in the market. From this perspective, the
process of identifying and developing the requi- nesses (Teece, 1980, 1982).
The resource-based perspective puts both verti-site assets is not particularly problematic. The
process involves nothing more than choosing cal integration and diversification into a new stra-
tegic light. Both can be viewed as ways of captur-rationally among a well-defined set of investment
alternatives. If assets are not already owned, they ing rents on scarce, firm-specific assets whose
services are difficult to sell in intermediate mar-can be bought. The resource-based perspective is
strongly at odds with this conceptualization. kets (Penrose, 1959; Williamson, 1975; Teece,
1980, 1982, 1986a, 1986b; Wernerfelt, 1984).From the resource-based perspective, firms are
heterogeneous with respect to their resources/ Empirical work on the relationship between per-
formance and diversification by Wernerfelt andcapabilities/endowments. Further, resource endow-
ments are ‘sticky:’ at least in the short run, firms Montgomery (1988) provides evidence for this
proposition. It is evident that the resource-basedare to some degree stuck with what they have
and may have to live with what they lack.17
This perspective focuses on strategies for exploiting
existing firm-specific assets.stickiness arises for three reasons. First, business
development is viewed as an extremely complex However, the resource-based perspective also
invites consideration of managerial strategies for
developing new capabilities (Wernerfelt, 1984).15
Studies of the automobile and other industries displayed
Indeed, if control over scarce resources is thedifferences in organization which often underlay differences
amongst firms. See, for example, Womack, Jones, and Roos, source of economic profits, then it follows that
1991; Hayes and Clark, 1985; Barney, Spender and Reve, such issues as skill acquisition, the management
1994; Clark and Fujimoto, 1991; Henderson and Cockburn,
of knowledge and know-how (Shuen, 1994), and1994; Nelson, 1991; Levinthal and Myatt, 1994.
16
Using FTC line of business data, Rumelt showed that stable learning become fundamental strategic issues. It
industry effects account for only 8 percent of the variance in is in this second dimension, encompassing skill
business unit returns. Furthermore, only about 40 percent of
acquisition, learning, and accumulation of organi-the dispersion in industry returns is due to stable industry
effects. zational and intangible or ‘invisible’ assets (Itami
17
In this regard, this approach has much in common with
recent work on organizational ecology (e.g., Freeman and
Boeker, 1984) and also on commitment (Ghemawat, 1991:
17–25). 18
Capability development, however, is not really analyzed.

and Roehl, 1987), that we believe lies the greatest are influenced by past choices. At any given point
in time, firms must follow a certain trajectory orpotential for contributions to strategy.
path of competence development. This path not
only defines what choices are open to the firm
The dynamic capabilities approach: Overview
today, but it also puts bounds around what its
internal repertoire is likely to be in the future.The global competitive battles in high-technology
industries such as semiconductors, information Thus, firms, at various points in time, make long-
term, quasi-irreversible commitments to certainservices, and software have demonstrated the need
for an expanded paradigm to understand how domains of competence.19
The notion that competitive advantage requirescompetitive advantage is achieved. Well-known
companies like IBM, Texas Instruments, Philips, both the exploitation of existing internal and
external firm-specific capabilities, and developingand others appear to have followed a ‘resource-
based strategy’ of accumulating valuable tech- new ones is partially developed in Penrose
(1959), Teece (1982), and Wernerfelt (1984).nology assets, often guarded by an aggressive
intellectual property stance. However, this strat- However, only recently have researchers begun
to focus on the specifics of how some organiza-egy is often not enough to support a significant
competitive advantage. Winners in the global tions first develop firm-specific capabilities and
how they renew competences to respond to shiftsmarketplace have been firms that can demonstrate
timely responsiveness and rapid and flexible prod- in the business environment.20
These issues are
intimately tied to the firm’s business processes,uct innovation, coupled with the management
capability to effectively coordinate and redeploy market positions, and expansion paths. Several
writers have recently offered insights and evi-internal and external competences. Not surpris-
ingly, industry observers have remarked that com- dence on how firms can develop their capability
to adapt and even capitalize on rapidly changingpanies can accumulate a large stock of valuable
technology assets and still not have many use- environments.21
The dynamic capabilities
approach seeks to provide a coherent frameworkful capabilities.
We refer to this ability to achieve new forms which can both integrate existing conceptual and
empirical knowledge, and facilitate prescription.of competitive advantage as ‘dynamic capabili-
ties’ to emphasize two key aspects that were not In doing so, it builds upon the theoretical foun-
dations provided by Schumpeter (1934), Penrosethe main focus of attention in previous strategy
perspectives. The term ‘dynamic’ refers to the (1959), Williamson (1975, 1985), Barney (1986),
Nelson and Winter (1982), Teece (1988), andcapacity to renew competences so as to achieve
congruence with the changing business environ- Teece et al. (1994).
ment; certain innovative responses are required
when time-to-market and timing are critical, the
rate of technological change is rapid, and the TOWARD A DYNAMIC CAPABILITIES
FRAMEWORKnature of future competition and markets difficult
to determine. The term ‘capabilities’ emphasizes
Terminology
the key role of strategic management in appropri-
ately adapting, integrating, and reconfiguring In order to facilitate theory development and
intellectual dialogue, some acceptable definitionsinternal and external organizational skills,
resources, and functional competences to match are desirable. We propose the following.
the requirements of a changing environment.
One aspect of the strategic problem facing 19
Deciding, under significant uncertainty about future states
an innovating firm in a world of Schumpeterian of the world, which long-term paths to commit to and when
to change paths is the central strategic problem confrontingcompetition is to identify difficult-to-imitate
the firm. In this regard, the work of Ghemawat (1991) isinternal and external competences most likely to
highly germane to the dynamic capabilities approach to
support valuable products and services. Thus, as strategy.
20
See, for example, Iansiti and Clark (1994) and Hendersonargued by Dierickx and Cool (1989), choices
(1994).about how much to spend (invest) on different 21
See Hayes et al. (1988), Prahalad and Hamel (1990),
possible areas are central to the firm’s strategy. Dierickx and Cool (1989), Chandler (1990), and Teece
(1993).However, choices about domains of competence

to which a core competence is distinctive depends
Factors of production
on how well endowed the firm is relative to its
competitors, and on how difficult it is for com-These are ‘undifferentiated’ inputs available in
disaggregate form in factor markets. By undiffer- petitors to replicate its competences.
entiated we mean that they lack a firm-specific
component. Land, unskilled labor, and capital are
Dynamic capabilities
typical examples. Some factors may be available
for the taking, such as public knowledge. In the We define dynamic capabilities as the firm’s
ability to integrate, build, and reconfigure internallanguage of Arrow, such resources must be ‘non-
fugitive.’22
Property rights are usually well and external competences to address rapidly
changing environments. Dynamic capabilities thusdefined for factors of production.
reflect an organization’s ability to achieve new
and innovative forms of competitive advantage
Resources23
given path dependencies and market positions
(Leonard-Barton, 1992).Resources are firm-specific assets that are difficult
if not impossible to imitate. Trade secrets and
certain specialized production facilities and engi-
Products
neering experience are examples. Such assets are
difficult to transfer among firms because of trans- End products are the final goods and services
produced by the firm based on utilizing the com-actions costs and transfer costs, and because the
assets may contain tacit knowledge. petences that it possesses. The performance
(price, quality, etc.) of a firm’s products relative
to its competitors at any point in time will depend
Organizational routines/competences
upon its competences (which over time depend
on its capabilities).When firm-specific assets are assembled in inte-
grated clusters spanning individuals and groups
so that they enable distinctive activities to be
Markets and strategic capabilities
performed, these activities constitute organiza-
tional routines and processes. Examples include Different approaches to strategy view sources of
wealth creation and the essence of the strategicquality, miniaturization, and systems integration.
Such competences are typically viable across mul- problem faced by firms differently. The competi-
tive forces framework sees the strategic problemtiple product lines, and may extend outside the
firm to embrace alliance partners. in terms of industry structure, entry deterrence,
and positioning; game-theoretic models view the
strategic problem as one of interaction between
Core competences
rivals with certain expectations about how each
other will behave;25
resource-based perspectivesWe define those competences that define a firm’s
fundamental business as core. Core competences have focused on the exploitation of firm-specific
assets. Each approach asks different, often com-must accordingly be derived by looking across
the range of a firm’s (and its competitors) prod- plementary questions. A key step in building a
conceptual framework related to dynamic capa-ucts and services.24
The value of core com-
petences can be enhanced by combination with bilities is to identify the foundations upon which
distinctive and difficult-to-replicate advantagesthe appropriate complementary assets. The degree
can be built, maintained, and enhanced.
A useful way to vector in on the strategic22
Arrow (1996) defines fugitive resources as ones that can
move cheaply amongst individuals and firms. elements of the business enterprise is first to
23
We do not like the term ‘resource’ and believe it is identify what is not strategic. To be strategic, a
misleading. We prefer to use the term firm-specific asset. We
use it here to try and maintain links to the literature on the
resource-based approach which we believe is important.
24
Thus Eastman Kodak’s core competence might be con-
sidered imaging, IBM’s might be considered integrated data 25
In sequential move games, each player looks ahead and
anticipates his rival’s future responses in order to reason backprocessing and service, and Motorola’s untethered communi-
cations. and decide action, i.e., look forward, reason backward.

capability must be honed to a user need26
(so to coordinate activity.28
The very essence of most
capabilities/competences is that they cannot bethere is a source of revenues), unique (so that
the products/services produced can be priced readily assembled through markets (Teece, 1982,
1986a; Zander and Kogut, 1995). If the abilitywithout too much regard to competition) and
difficult to replicate (so profits will not be com- to assemble competences using markets is what
is meant by the firm as a nexus of contractspeted away). Accordingly, any assets or entity
which are homogeneous and can be bought and (Fama, 1980), then we unequivocally state that
the firm about which we theorize cannot be use-sold at an established price cannot be all that
strategic (Barney, 1986). What is it, then, about fully modeled as a nexus of contracts. By ‘con-
tract’ we are referring to a transaction undergirdedfirms which undergirds competitive advantage?
To answer this, one must first make some by a legal agreement, or some other arrangement
which clearly spells out rights, rewards, andfundamental distinctions between markets and
internal organization (firms). The essence of the responsibilities. Moreover, the firm as a nexus of
contracts suggests a series of bilateral contractsfirm, as Coase (1937) pointed out, is that it
displaces market organization. It does so in the orchestrated by a coordinator. Our view of the
firm is that the organization takes place in a moremain because inside the firms one can organize
certain types of economic activity in ways one multilateral fashion, with patterns of behavior
and learning being orchestrated in a much morecannot using markets. This is not only because
of transaction costs, as Williamson (1975, 1985) decentralized fashion, but with a viable head-
quarters operation.emphasized, but also because there are many
types of arrangements where injecting high-pow- The key point, however, is that the properties
of internal organization cannot be replicated byered (market like) incentives might well be quite
destructive of cooperative activity and learning.27
a portfolio of business units amalgamated just
through formal contracts as many distinctiveInside an organization, exchange cannot take
place in the same manner that it can outside an elements of internal organization simply cannot
be replicated in the market.29
That is, entrepre-organization, not just because it might be destruc-
tive to provide high-powered individual incen- neurial activity cannot lead to the immediate rep-
lication of unique organizational skills throughtives, but because it is difficult if not impossible
to tightly calibrate individual contribution to a simply entering a market and piecing the parts
together overnight. Replication takes time, andjoint effort. Hence, contrary to Arrow’s (1969)
view of firms as quasi markets, and the task the replication of best practice may be illusive.
Indeed, firm capabilities need to be understoodof management to inject markets into firms, we
recognize the inherent limits and possible counter- not in terms of balance sheet items, but mainly
in terms of the organizational structures andproductive results of attempting to fashion firms
into simply clusters of internal markets. In parti- managerial processes which support productive
activity. By construction, the firm’s balance sheetcular, learning and internal technology transfer
may well be jeopardized. contains items that can be valued, at least at
original market prices (cost). It is necessarily theIndeed, what is distinctive about firms is that
they are domains for organizing activity in a case, therefore, that the balance sheet is a poor
shadow of a firm’s distinctive competences.30
nonmarket-like fashion. Accordingly, as we dis-
cuss what is distinctive about firms, we stress
competences/capabilities which are ways of 28
We see the problem of market contracting as a matter of
coordination as much as we see it a problem of opportunismorganizing and getting things done which cannot
in the fact of contractual hazards. In this sense, we arebe accomplished merely by using the price system
consonant with both Richardson (1960) and Williamson
(1975, 1985).
29
As we note in Teece et al. (1994), the conglomerate offers
few if any efficiencies because there is little provided by26
Needless to say, users need not be the current customers
of the enterprise. Thus a capability can be the basis for the conglomerate form that shareholders cannot obtain for
themselves simply by holding a diversified portfolio of stocks.diversification into new product markets.
27
Indeed, the essence of internal organization is that it is a 30
Owners’ equity may reflect, in part, certain historic capabili-
ties. Recently, some scholars have begun to attempt to meas-domain of unleveraged or low-powered incentives. By unlever-
aged we mean that rewards are determined at the group or ure organizational capability using financial statement data.
See Baldwin and Clark (1991) and Lev and Sougiannisorganization level, not primarily at the individual level, in an
effort to encourage team behavior, not individual behavior. (1992).

That which is distinctive cannot be bought and focuses on replication and imitation, as it is these
phenomena which determine how readily a com-sold short of buying the firm itself, or one or
more of its subunits. petence or capability can be cloned by competi-
tors, and therefore distinctiveness of its com-There are many dimensions of the business
firm that must be understood if one is to grasp petences and the durability of its advantage.
The firm’s processes and positions collectivelyfirm-level distinctive competences/capabilities. In
this paper we merely identify several classes of encompass its competences and capabilities. A
hierarchy of competences/capabilities ought to befactors that will help determine a firm’s distinc-
tive competence and dynamic capabilities. We recognized, as some competences may be on the
factory floor, some in the R&D labs, some in theorganize these in three categories: processes, po-
sitions, and paths. The essence of competences executive suites, and some in the way everything
is integrated. A difficult-to-replicate or difficult-and capabilities is embedded in organizational
processes of one kind or another. But the content to-imitate competence was defined earlier as a
distinctive competence. As indicated, the key fea-of these processes and the opportunities they
afford for developing competitive advantage at ture of distinctive competence is that there is not
a market for it, except possibly through the mar-any point in time are shaped significantly by the
assets the firm possesses (internal and market) ket for business units. Hence competences and
capabilities are intriguing assets as they typicallyand by the evolutionary path it has
adopted/inherited. Hence organizational processes, must be built because they cannot be bought.
shaped by the firm’s asset positions and molded
by its evolutionary and co-evolutionary paths,
Organizational and managerial processes
explain the essence of the firm’s dynamic capa-
bilities and its competitive advantage. Organizational processes have three roles:
coordination/integration (a static concept); learn-
ing (a dynamic concept); and reconfiguration (a
Processes, positions, and paths
transformational concept). We discuss each in
turn.We thus advance the argument that the competi-
tive advantage of firms lies with its managerial
and organizational processes, shaped by its Coordination/integration. While the price sys-
tem supposedly coordinates the economy,32
man-(specific) asset position, and the paths available
to it.31
By managerial and organizational proc- agers coordinate or integrate activity inside the
firm. How efficiently and effectively internalesses, we refer to the way things are done in the
firm, or what might be referred to as its routines, coordination or integration is achieved is very
important (Aoki, 1990).33
Likewise for externalor patterns of current practice and learning. By
position we refer to its current specific endow- coordination.34
Increasingly, strategic advantage
requires the integration of external activities andments of technology, intellectual property, com-
plementary assets, customer base, and its external technologies. The growing literature on strategic
relations with suppliers and complementors. By
paths we refer to the strategic alternatives avail-
32
The coordinative properties of markets depend on pricesable to the firm, and the presence or absence of
being “sufficient” upon which to base resource allocationincreasing returns and attendant path depen-
decisions.
dencies. 33
Indeed, Ronald Coase, author of the pathbreaking 1937
article ‘The nature of the firm,’ which focused on the costsOur focus throughout is on asset structures for
of organizational coordination inside the firm as compared towhich no ready market exists, as these are the
across the market, half a century later has identified as critical
only assets of strategic interest. A final section the understanding of ‘why the costs of organizing particular
activities differs among firms’ (Coase, 1988: 47). We argue
that a firm’s distinctive ability needs to be understood as a
reflection of distinctive organizational or coordinative capabili-31
We are implicitly saying that fixed assets, like plant and
equipment which can be purchased off-the-shelf by all industry ties. This form of integration (i.e., inside business units) is
different from the integration between business units; theyparticipants, cannot be the source of a firm’s competitive
advantage. In asmuch as financial balance sheets typically could be viable on a stand-alone basis (external integration).
For a useful taxonomy, see Iansiti and Clark (1994).reflect such assets, we point out that the assets that matter
for competitive advantage are rarely reflected in the balance 34
Shuen (1994) examines the gains and hazards of the tech-
nology make-vs.-buy decision and supplier codevelopment.sheet, while those that do not are.

alliances, the virtual corporation, and buyer– gest that productive systems display high interde-
pendency, and that it may not be possible tosupplier relations and technology collaboration
evidences the importance of external integration change one level without changing others. This
appears to be true with respect to the ‘leanand sourcing.
There is some field-based empirical research production’ model (Womack et al., 1991) which
has now transformed the Taylor or Ford modelthat provides support for the notion that the way
production is organized by management inside of manufacturing organization in the automobile
industry.36
Lean production requires distinctivethe firm is the source of differences in firms’
competence in various domains. For example, shop floor practices and processes as well as
distinctive higher-order managerial processes. PutGarvin’s (1988) study of 18 room air-condition-
ing plants reveals that quality performance was differently, organizational processes often display
high levels of coherence, and when they do,not related to either capital investment or the
degree of automation of the facilities. Instead, replication may be difficult because it requires
systemic changes throughout the organization andquality performance was driven by special organi-
zational routines. These included routines for also among interorganizational linkages, which
might be very hard to effectuate. Put differently,gathering and processing information, for linking
customer experiences with engineering design partial imitation or replication of a successful
model may yield zero benefits.37
choices, and for coordinating factories and
component suppliers.35
The work of Clark and
Fujimoto (1991) on project development in the 36
Fujimoto (1994: 18–20) describes key elements as they
automobile industry also illustrates the role played existed in the Japanese auto industry as follows: ‘The typical
volume production system of effective Japanese makers ofby coordinative routines. Their study reveals a
the 1980s (e.g., Toyota) consists of various intertwinedsignificant degree of variation in how different
elements that might lead to competitive advantages. Just-in-
firms coordinate the various activities required to Time (JIT), Jidoka (automatic defect detection and machine
stop), Total Quality Control (TQC), and continuous improve-bring a new model from concept to market. These
ment (Kaizen) are often pointed out as its core subsystems.differences in coordinative routines and capabili-
The elements of such a system include inventory reduction
ties seem to have a significant impact on such mechanisms by Kanban system; levelization of production
volume and product mix (heijunka); reduction of ‘muda’performance variables as development cost, devel-
(non-value adding activities), ‘mura’ (uneven pace ofopment lead times, and quality. Furthermore,
production) and muri (excessive workload); production plans
Clark and Fujimoto tended to find significant based on dealers’ order volume (genyo seisan); reduction of
die set-up time and lot size in stamping operation; mixedfirm-level differences in coordination routines and
model assembly; piece-by-piece transfer of parts betweenthese differences seemed to have persisted for a
machines (ikko-nagashi); flexible task assignment for volume
long time. This suggests that routines related to changes and productivity improvement (shojinka); multi-task
job assignment along the process flow (takotei-mochi); U-coordination are firm-specific in nature.
shape machine layout that facilitates flexible and multipleAlso, the notion that competence/capability is
task assignment, on-the-spot inspection by direct workers
embedded in distinct ways of coordinating and (tsukurikomi); fool-proof prevention of defects (poka-yoke);
real-time feedback of production troubles (andon); assemblycombining helps to explain how and why seem-
line stop cord; emphasis on cleanliness, order and disciplineingly minor technological changes can have
on the shop floor (5-S); frequent revision of standard operating
devastating impacts on incumbent firms’ abilities procedures by supervisors; quality control circles; standardized
tools for quality improvement (e.g., 7 tools for QC, QCto compete in a market. Henderson and Clark
story); worker involvement in preventive maintenance (Total(1990), for example, have shown that in-
Productive Maintenance); low cost automation or semi-auto-
cumbments in the photolithographic equipment mation with just-enough functions); reduction of process steps
for saving of tools and dies, and so on. The human-resourceindustry were sequentially devasted by seemingly
management factors that back up the above elements includeminor innovations that, nevertheless, had major
stable employment of core workers (with temporary workers
impacts on how systems had to be configured. in the periphery); long-term training of multi-skilled (multi-
task) workers; wage system based in part on skill accumu-They attribute these difficulties to the fact that
lation; internal promotion to shop floor supervisors; coopera-systems-level or ‘architectural’ innovations often
tive relationships with labor unions; inclusion of production
require new routines to integrate and coordinate supervisors in union members; generally egalitarian policies
for corporate welfare, communication and worker motivation.engineering tasks. These findings and others sug-
Parts procurement policies are also pointed out often as a
source of the competitive advantage.
37
For a theoretical argument along these lines, see Milgrom35
Garvin (1994 ) provides a typology of organizational
processes. and Roberts (1990).

The notion that there is a certain rationality or which repetition and experimentation enable tasks
to be performed better and quicker. It also enablescoherence to processes and systems is not quite
the same concept as corporate culture, as we new production opportunities to be identified.39
In the context of the firm, if not more generally,understand the latter. Corporate culture refers to
the values and beliefs that employees hold; cul- learning has several key characteristics. First,
learning involves organizational as well as indi-ture can be a de facto governance system as it
mediates the behavior of individuals and econo- vidual skills.40
While individual skills are of rel-
evance, their value depends upon their employ-mizes on more formal administrative methods.
Rationality or coherence notions are more akin ment, in particular organizational settings.
Learning processes are intrinsically social andto the Nelson and Winter (1982) notion of organi-
zational routines. However, the routines concept collective and occur not only through the imi-
tation and emulation of individuals, as withis a little too amorphous to properly capture
the congruence amongst processes and between teacher–student or master–apprentice, but also
because of joint contributions to the understand-processes and incentives that we have in mind.
Consider a professional service organization like ing of complex problems.41
Learning requires
common codes of communication and coordinatedan accounting firm. If it is to have relatively
high-powered incentives that reward individual search procedures. Second, the organizational
knowledge generated by such activity resides inperformance, then it must build organizational
processes that channel individual behavior; if it new patterns of activity, in ‘routines,’ or a new
logic of organization. As indicated earlier, rou-has weak or low-powered incentives, it must find
symbolic ways to recognize the high performers, tines are patterns of interactions that represent
successful solutions to particular problems. Theseand it must use alternative methods to build effort
and enthusiasm. What one may think of as styles patterns of interaction are resident in group
behavior, though certain subroutines may be resi-of organization in fact contain necessary, not
discretionary, elements to achieve performance. dent in individual behavior. The concept of
dynamic capabilities as a coordinative manage-Recognizing the congruences and complemen-
tarities among processes, and between processes ment process opens the door to the potential
for interorganizational learning. Researchers (Dozand incentives, is critical to the understanding of
organizational capabilities. In particular, they can and Shuen, 1990; Mody, 1993) have pointed out
that collaborations and partnerships can be ahelp us explain why architectural and radical
innovations are so often introduced into an indus- vehicle for new organizational learning, helping
firms to recognize dysfunctional routines, andtry by new entrants. The incumbents develop
distinctive organizational processes that cannot preventing strategic blindspots.
support the new technology, despite certain overt
similarities between the old and the new. The Reconfiguration and transformation. In rapidly
changing environments, there is obviously valuefrequent failure of incumbents to introduce new
technologies can thus be seen as a consequence in the ability to sense the need to reconfigure
the firm’s asset structure, and to accomplish theof the mismatch that so often exists between the
set of organizational processes needed to support necessary internal and external transformation
(Amit and Schoemaker, 1993; Langlois, 1994).the conventional product/service and the require-
ments of the new. Radical organizational re- This requires constant surveillance of markets and
technologies and the willingness to adopt bestengineering will usually be required to support
the new product, which may well do better practice. In this regard, benchmarking is of con-
embedded in a separate subsidiary where a new
set of coherent organizatonal processes can be 39
For a useful review and contribution, see Levitt and
March (1988).fashioned.38
40
Levinthal and March, 1993. Mahoney (1992) and Mahoney
and Pandian (1995) suggest that both resources and mental
Learning. Perhaps even more important than models are intertwined in firm-level learning.
41
There is a large literature on learning, although only aintegration is learning. Learning is a process by
small fraction of it deals with organizational learning. Relevant
contributors include Levitt and March (1988), Argyris and
Schon (1978), Levinthal and March (1981), Nelson and
Winter (1982), and Leonard-Barton (1995).38
See Abernathy and Clark (1985).

siderable value as an organized process for Prior commercialization activities require and
enable firms to build such complementaritiesaccomplishing such ends (Camp, 1989). In
dynamic environments, narcissistic organizations (Teece, 1986b). Such capabilities and assets,
while necessary for the firm’s established activi-are likely to be impaired. The capacity to recon-
figure and transform is itself a learned organiza- ties, may have other uses as well. These assets
typically lie downstream. New products and proc-tional skill. The more frequently practiced, the
easier accomplished. esses either can enhance or destroy the value of
such assets (Tushman, Newman, and Romanelli,Change is costly and so firms must develop
processes to minimize low pay-off change. The 1986). Thus the development of computers
enhanced the value of IBM’s direct sales forceability to calibrate the requirements for change
and to effectuate the necessary adjustments would in office products, while disk brakes rendered
useless much of the auto industry’s investmentappear to depend on the ability to scan the
environment, to evaluate markets and competitors, in drum brakes.
and to quickly accomplish reconfiguration and
transformation ahead of competition. Decentrali- Financial assets. In the short run, a firm’s cash
position and degree of leverage may have stra-zation and local autonomy assist these processes.
Firms that have honed these capabilities are tegic implications. While there is nothing more
fungible than cash, it cannot always be raisedsometimes referred to as ‘high-flex’.
from external markets without the dissemination
of considerable information to potential investors.
Positions
Accordingly, what a firm can do in short order
is often a function of its balance sheet. In theThe strategic posture of a firm is determined not
only by its learning processes and by the coher- longer run, that ought not be so, as cash flow
ought be more determinative.ence of its internal and external processes and
incentives, but also by its specific assets. By
specific assets we mean for example its special- Reputational assets. Firms, like individuals,
have reputations. Reputations often summarize aized plant and equipment. These include its
difficult-to-trade knowledge assets and assets good deal of information about firms and shape
the responses of customers, suppliers, and com-complementary to them, as well as its reputational
and relational assets. Such assets determine its petitors. It is sometimes difficult to disentangle
reputation from the firm’s current asset and mar-competitive advantage at any point in time. We
identify several illustrative classes. ket position. However, in our view, reputational
assets are best viewed as an intangible asset that
enables firms to achieve various goals in theTechnological assets. While there is an emerg-
ing market for know-how (Teece, 1981), much market. Its main value is external, since what is
critical about reputation is that it is a kind oftechnology does not enter it. This is either
because the firm is unwilling to sell it42
or summary statistic about the firm’s current assets
and position, and its likely future behavior.because of difficulties in transacting in the market
for know-how (Teece, 1980). A firm’s techno- Because there is generally a strong asymmetry
between what is known inside the firm and whatlogical assets may or may not be protected by
the standard instruments of intellectual property is known externally, reputations may sometimes
be more salient than the true state of affairs, inlaw. Either way, the ownership protection and
utilization of technological assets are clearly key the sense that external actors must respond to
what they know rather than what is knowable.differentiators among firms. Likewise for com-
plementary assets.
Structural assets. The formal and informal
structure of organizations and their external link-Complementary assets. Technological inno-
vations require the use of certain related assets ages have an important bearing on the rate and
direction of innovation, and how competencesto produce and deliver new products and services.
and capabilities co-evolve (Argyres, 1995; Teece,
1996). The degree of hierarchy and the level of42
Managers often evoke the ‘crown jewels’ metaphor. That
is, if the technology is released, the kingdom will be lost. vertical and lateral integration are elements of

firm-specific structure. Distinctive governance Organizational boundaries. An important
dimension of ‘position’ is the location of a firm’smodes can be recognized (e.g., multiproduct, inte-
grated firms; high ‘flex’ firms; virtual corpora- boundaries. Put differently, the degree of inte-
gration (vertical, lateral, and horizontal) is oftions; conglomerates), and these modes support
different types of innovation to a greater or lesser quite some significance. Boundaries are not only
significant with respect to the technological anddegree. For instance, virtual structures work well
when innovation is autonomous; integrated struc- complementary assets contained within, but also
with respect to the nature of the coordination thattures work better for systemic innovations.
can be achieved internally as compared to through
markets. When specific assets or poorly protectedInstitutional assets. Environments cannot be
defined in terms of markets alone. While public intellectual capital are at issue, pure market
arrangements expose the parties to recontractingpolicies are usually recognized as important in
constraining what firms can do, there is a ten- hazards or appropriability hazards. In such cir-
cumstances, hierarchical control structures maydency, particularly by economists, to see these
as acting through markets or through incentives. work better than pure arms-length contracts.44
However, institutions themselves are a critical
element of the business environment. Regulatory
Paths
systems, as well as intellectual property regimes,
tort laws, and antitrust laws, are also part of the
environment. So is the system of higher education Path dependencies. Where a firm can go is a
function of its current position and the pathsand national culture. There are significant national
differences here, which is just one of the reasons ahead. Its current position is often shaped by
the path it has traveled. In standard economicsgeographic location matters (Nelson, 1994). Such
assets may not be entirely firm specific; firms of textbooks, firms have an infinite range of technol-
ogies from which they can choose and marketsdifferent national and regional origin may have
quite different institutional assets to call upon they can occupy. Changes in product or factor
prices will be responded to instantaneously, withbecause their institutional/policy settings are so
different. technologies moving in and out according to
value maximization criteria. Only in the short run
are irreversibilities recognized. Fixed costs—suchMarket (structure) assets. Product market po-
sition matters, but it is often not at all determina- as equipment and overheads—cause firms to price
below fully amortized costs but never constraintive of the fundamental position of the enterprise
in its external environment. Part of the problem future investment choices. ‘Bygones are bygones.’
Path dependencies are simply not recognized.lies in defining the market in which a firm com-
petes in a way that gives economic meaning. This is a major limitation of microeconomic
theory.More importantly, market position in regimes of
rapid technological change is often extremely The notion of path dependencies recognizes
that ‘history matters.’ Bygones are rarelyfragile. This is in part because time moves on a
different clock in such environments.43
Moreover, bygones, despite the predictions of rational actor
theory. Thus a firm’s previous investments andthe link between market share and innovation has
long been broken, if it ever existed (Teece, 1996).
All of this is to suggest that product market
44
Williamson (1996: 102–103) has observed, failures of coor-position, while important, is too often overplayed.
dination may arise because ‘parties that bear a long termStrategy should be formulated with regard to the
bilateral dependency relationship to one another must recog-
more fundamental aspects of firm performance, nize that incomplete contracts require gap filling and some-
times get out of alignment. Although it is always in thewhich we believe are rooted in competences and
collective interest of autonomous parties to fill gaps, correctcapabilities and shaped by positions and paths.
errors, and affect efficient realignments, it is also the case
that the distribution of the resulting gains is indeterminate.
Self-interested bargaining predictably obtains. Such bargaining
is itself costly. The main costs, however, are that transactions43
For instance, an Internet year might well be thought of as
equivalent to 10 years on many industry clocks, because as are maladapted to the environment during the bargaining
interval. Also, the prospect of ex post bargaining invites exmuch change occurs in the Internet business in a year that
occurs in say the auto industry in a decade. ante prepositioning of an inefficient kind.’

its repertoire of routines (its ‘history’) constrain gically through technology-sponsoring activities.46
The first type of competition is not unlike biologi-its future behavior.45
This follows because learn-
ing tends to be local. That is, opportunities for cal competition amongst species, although it can
be sharpened by managerial activities thatlearning will be ‘close in’ to previous activities
and thus will be transaction and production spe- enhance the performance of products and proc-
esses. The reality is that companies with the bestcific (Teece, 1988). This is because learning is
often a process of trial, feedback, and evaluation. products will not always win, as chance events
may cause ‘lock-in’ on inferior technologiesIf too many parameters are changed simul-
taneously, the ability of firms to conduct mean- (Arthur, 1983) and may even in special cases
generate switching costs for consumers. However,ingful natural quasi experiments is attenuated. If
many aspects of a firm’s learning environment while switching costs may favor the incumbent,
in regimes of rapid technological change switch-change simultaneously, the ability to ascertain
cause–effect relationships is confounded because ing costs can become quickly swamped by
switching benefits. Put differently, new productscognitive structures will not be formed and rates
of learning diminish as a result. One implication employing different standards often appear with
alacrity in market environments experiencingis that many investments are much longer term
than is commonly thought. rapid technological change, and incumbents can
be readily challenged by superior products andThe importance of path dependencies is ampli-
fied where conditions of increasing returns to services that yield switching benefits. Thus the
degree to which switching costs cause ‘lock-in’adoption exist. This is a demand-side phenom-
enon, and it tends to make technologies and is a function of factors such as user learning,
rapidity of technological change, and the amountproducts embodying those technologies more
attractive the more they are adopted. Attractive- of ferment in the competitive environment.
ness flows from the greater adoption of the prod-
uct amongst users, which in turn enables them to Technological opportunities. The concept of
path dependencies is given forward meaningbecome more developed and hence more useful.
Increasing returns to adoption has many sources through the consideration of an industry’s techno-
logical opportunities. It is well recognized thatincluding network externalities (Katz and Shapiro,
1985), the presence of complementary assets how far and how fast a particular area of indus-
trial activity can proceed is in part due to the(Teece, 1986b) and supporting infrastructure
(Nelson, 1996), learning by using (Rosenberg, technological opportunities that lie before it. Such
opportunities are usually a lagged function of1982), and scale economies in production and
distribution. Competition between and amongst foment and diversity in basic science, and the
rapidity with which new scientific breakthroughstechnologies is shaped by increasing returns.
Early leads won by good luck or special circum- are being made.
However, technological opportunities may notstances (Arthur, 1983) can become amplified by
increasing returns. This is not to suggest that be completely exogenous to industry, not only
because some firms have the capacity to engagefirst movers necessarily win. Because increasing
returns have multiple sources, the prior posi- in or at least support basic research, but also
because technological opportunities are often fedtioning of firms can affect their capacity to exploit
increasing returns. Thus, in Mitchell’s (1989) by innovative activity itself. Moreover, the recog-
nition of such opportunities is affected by thestudy of medical diagnostic imaging, firms
already controlling the relevant complementary
assets could in theory start last and finish first. 46
Because of huge uncertainties, it may be extremely difficult
In the presence of increasing returns, firms can to determine viable strategies early on. Since the rules of the
game and the identity of the players will be revealed onlycompete passively, or they may compete strate-
after the market has begun to evolve, the pay-off is likely to
lie with building and maintaining organizational capabilities
that support flexibility. For example, Microsoft’s recent about-
face and vigorous pursuit of Internet business once the Net-
Scape phenomenon became apparent is impressive, not so
much because it perceived the need to change strategy, but
because of its organizational capacity to effectuate a stra-45
For further development, see Bercovitz, de Figueiredo, and
Teece, 1996. tegic shift.

organizational structures that link the institutions approaches to the firm and to strategy.49
More-
over, the agency theoretic view of the firm as aengaging in basic research (primarily the
university) to the business enterprise. Hence, the nexus of contracts would put no weight on proc-
esses, positions, and paths. While agencyexistence of technological opportunities can be
quite firm specific. approaches to the firm may recognize that oppor-
tunism and shirking may limit what a firm canImportant for our purposes is the rate and
direction in which relevant scientific frontiers are do, they do not recognize the opportunities and
constraints imposed by processes, positions, andbeing rolled back. Firms engaging in R&D may
find the path dead ahead closed off, though break- paths.
Moreover, the firm in our conceptualization isthroughs in related areas may be sufficiently close
to be attractive. Likewise, if the path dead ahead much more than the sum of its parts—or a team
tied together by contracts.50
Indeed, to someis extremely attractive, there may be no incentive
for firms to shift the allocation of resources away extent individuals can be moved in and out of
organizations and, so long as the internal proc-from traditional pursuits. The depth and width of
technological opportunities in the neighborhood esses and structures remain in place, performance
will not necessarily be impaired. A shift in theof a firm’s prior research activities thus are likely
to impact a firm’s options with respect to both environment is a far more serious threat to the
firm than is the loss of key individuals, as individ-the amount and level of R&D activity that it
can justify. In addition, a firm’s past experience uals can be replaced more readily than organiza-
tions can be transformed. Furthermore, theconditions the alternatives management is able to
perceive. Thus, not only do firms in the same dynamic capabilities view of the firm would sug-
gest that the behavior and performance of parti-industry face ‘menus’ with different costs associa-
ted with particular technological choices, they cular firms may be quite hard to replicate, even
if its coherence and rationality are observable.also are looking at menus containing different
choices.47
This matter and related issues involving repli-
cation and imitation are taken up in the section
that follows.
Assessment
The essence of a firm’s competence and dynamic
Replicability and imitatability of
capabilities is presented here as being resident in
organizational processes and positions
the firm’s organizational processes, that are in
turn shaped by the firm’s assets (positions) and its Thus far, we have argued that the competences
and capabilities (and hence competitive advantage)evolutionary path. Its evolutionary path, despite
managerial hubris that might suggest otherwise, of a firm rest fundamentally on processes, shaped
by positions and paths. However, competencesis often rather narrow.48
What the firm can do
and where it can go are thus rather constrained can provide competitive advantage and generate
rents only if they are based on a collection ofby its positions and paths. Its competitors are
likewise constrained. Rents (profits) thus tend to routines, skills, and complementary assets that are
difficult to imitate.51
A particular set of routinesflow not just from the asset structure of the firm
and, as we shall see, the degree of its imitability, can lose their value if they support a competence
which no longer matters in the marketplace, orbut also by the firm’s ability to reconfigure and
transform. if they can be readily replicated or emulated by
competitors. Imitation occurs when firms discoverThe parameters we have identified for de-
termining performance are quite different from and simply copy a firm’s organizational routines
and procedures. Emulation occurs when firmsthose in the standard textbook theory of the firm,
and in the competitive forces and strategic conflict
49
In both the firm is still largely a black box. Certainly, little
or no attention is given to processes, positions, and paths.47
This is a critical element in Nelson and Winter’s (1982)
view of firms and technical change. 50
See Alchian and Demsetz (1972).
51
We call such competences distinctive. See also Dierickx48
We also recognize that the processes, positions, and paths of
customers also matter. See our discussion above on increasing and Cool (1989) for a discussion of the characteristics of
assets which make them a source of rents.returns, including customer learning and network externalities.

discover alternative ways of achieving the same Some routines and competences seem to be
attributable to local or regional forces that shapefunctionality.52
firms’ capabilities at early stages in their lives.
Porter (1990), for example, shows that differences
Replication
in local product markets, local factor markets,
and institutions play an important role in shapingTo understand imitation, one must first understand
replication. Replication involves transferring or competitive capabilities. Differences also exist
within populations of firms from the same coun-redeploying competences from one concrete eco-
nomic setting to another. Since productive knowl- try. Various studies of the automobile industry,
for example, show that not all Japanese auto-edge is embodied, this cannot be accomplished
by simply transmitting information. Only in those mobile companies are top performers in terms of
quality, productivity, or product developmentinstances where all relevant knowledge is fully
codified and understood can replication be col- (see, for example, Clark and Fujimoto, 1991).
The role of firm-specific history has been high-lapsed into a simple problem of information trans-
fer. Too often, the contextual dependence of ori- lighted as a critical factor explaining such firm-
level (as opposed to regional or national-level)ginal performance is poorly appreciated, so unless
firms have replicated their systems of productive differences (Nelson and Winter, 1982). Repli-
cation in a different context may thus be ratherknowledge on many prior occasions, the act of
replication is likely to be difficult (Teece, 1976). difficult.
At least two types of strategic value flowIndeed, replication and transfer are often impos-
sible absent the transfer of people, though this from replication. One is the ability to support
geographic and product line expansion. To thecan be minimized if investments are made to
convert tacit knowledge to codified knowledge. extent that the capabilities in question are rel-
evant to customer needs elsewhere, replicationOften, however, this is simply not possible.
In short, competences and capabilities, and the can confer value.55
Another is that the ability
to replicate also indicates that the firm has theroutines upon which they rest, are normally rather
difficult to replicate.53
Even understanding what foundations in place for learning and improve-
ment. Considerable empirical evidence supportsall the relevant routines are that support a parti-
cular competence may not be transparent. Indeed, the notion that the understanding of processes,
both in production and in management, is theLippman and Rumelt (1992) have argued that
some sources of competitive advantage are so key to process improvement. In short, an
organization cannot improve that which it doescomplex that the firm itself, let alone its competi-
tors, does not understand them.54
As Nelson and not understand. Deep process understanding is
often required to accomplish codification.Winter (1982) and Teece (1982) have explained,
many organizational routines are quite tacit in Indeed, if knowledge is highly tacit, it indicates
that underlying structures are not well under-nature. Imitation can also be hindered by the fact
few routines are ‘stand-alone;’ coherence may stood, which limits learning because scientific
and engineering principles cannot be as system-require that a change in one set of routines in
one part of the firm (e.g., production) requires atically applied.56
Instead, learning is confined
to proceeding through trial and error, and thechanges in some other part (e.g., R&D).
55
Needless to say, there are many examples of firms rep-52
There is ample evidence that a given type of competence
(e.g., quality) can be supported by different routines and licating their capabilities inappropriately by applying extant
routines to circumstances where they may not be applicable,combinations of skills. For example, the Garvin (1988) and
Clark and Fujimoto (1991) studies both indicate that there e.g., Nestle’s transfer of developed-country marketing methods
for infant formula to the Third World (Hartley, 1989). A keywas no one ‘formula’ for achieving either high quality or
high product development performance. strategic need is for firms to screen capabilities for their
applicability to new environments.53
See Szulanski’s (1995) discussion of the intrafirm transfer
of best practice. He quotes a senior vice president of Xerox 56
Different approaches to learning are required depending on
the depth of knowledge. Where knowledge is less articulatedas saying ‘you can see a high performance factory or office,
but it just doesn’t spread. I don’t know why.’ Szulanski also and structured, trial and error and learning-by-doing are
necessary, whereas in mature environments where the underly-discusses the role of benchmarking in facilitating the transfer
of best practice. ing engineering science is better understood, organizations
can undertake more deductive approaches or what Pisano54
If so, it is our belief that the firm’s advantage is likely to
fade, as luck does run out. (1994) refers to as ‘learning-before-doing.’

leverage that might otherwise come from the are thus more protectable if there is no need to
expose them in contexts where competitors canapplication of scientific theory is denied.
learn about them.
One should not, however, overestimate the
Imitation
overall importance of intellectual property protec-
tion; yet it presents a formidable imitation barrierImitation is simply replication performed by a
competitor. If self-replication is difficult, imitation in certain particular contexts. Intellectual property
protection is not uniform across products, proc-is likely to be harder. In competitive markets,
it is the ease of imitation that determines the esses, and technologies, and is best thought of as
islands in a sea of open competition. If one issustainability of competitive advantage. Easy imi-
tation implies the rapid dissipation of rents. not able to place the fruits of one’s investment,
ingenuity, or creativity on one or more of theFactors that make replication difficult also
make imitation difficult. Thus, the more tacit the islands, then one indeed is at sea.
We use the term appropriability regimes tofirm’s productive knowledge, the harder it is to
replicate by the firm itself or its competitors. describe the ease of imitation. Appropriability
is a function both of the ease of replication andWhen the tacit component is high, imitation may
well be impossible, absent the hiring away of the efficacy of intellectual property rights as a
barrier to imitation. Appropriability is strongkey individuals and the transfers of key organiza-
tion processes. when a technology is both inherently difficult
to replicate and the intellectual property systemHowever, another set of barriers impedes imi-
tation of certain capabilities in advanced industrial provides legal barriers to imitation. When it
is inherently easy to replicate and intellectualcountries. This is the system of intellectual pro-
perty rights, such as patents, trade secrets, and property protection is either unavailable or inef-
fectual, then appropriability is weak. Intermedi-trademarks, and even trade dress.57
Intellectual
property protection is of increasing importance in ate conditions also exist.
the United States, as since 1982 the legal system
has adopted a more pro-patent posture. Similar
trends are evident outside the United States. CONCLUSION
Besides the patent system, several other factors
cause there to be a difference between replication The four paradigms discussed above are quite
different, though the first two have much incosts and imitation costs. The observability of
the technology or the organization is one such common with each other (strategizing) as do the
last two (economizing). But are these paradigmsimportant factor. Whereas vistas into product
technology can be obtained through strategies complementary or competitive? According to
some authors, ‘the resource perspective com-such as reverse engineering, this is not the case
for process technology, as a firm need not expose plements the industry analysis framework’ (Amit
and Schoemaker, 1993: 35). While this isits process technology to the outside in order to
benefit from it.58
Firms with product technology, undoubtedly true, we think that in several
important respects the perspectives are also com-on the other hand, confront the unfortunate cir-
cumstances that they must expose what they have petitive. While this should be recognized, it is
not to suggest that there is only one frameworkgot in order to profit from the technology. Secrets
that has value. Indeed, complex problems are
likely to benefit from insights obtained from all
57
Trade dress refers to the ‘look and feel’ of a retail establish- of the paradigms we have identified plus more.
ment, e.g., the distinctive marketing and presentation style of
The trick is to work out which frameworks areThe Nature Company.
58
An interesting but important exception to this can be found appropriate for the problem at hand. Slavish
in second sourcing. In the microprocessor business, until the adherence to one class to the neglect of all
introduction of the 386 chip, Intel and most other merchant
others is likely to generate strategic blindspots.semi producers were encouraged by large customers like IBM
to provide second sources, i.e., to license and share their The tools themselves then generate strategic
proprietary process technology with competitors like AMD vulnerability. We now explore these issues
and NEC. The microprocessor developers did so to assure
further. Table 1 summarizes some similaritiescustomers that they had sufficient manufacturing capability to
meet demand at all times. and differences.

DynamicCapabilities527
Table 1. Paradigms of strategy: Salient characteristics
Paradigm Intellectual Representative Nature Rationality Fundamental Short-run Role of Focal
roots authors of rents assumptions units of capacity for industrial concern
addressing of managers analysis strategic structure
strategic reorientation
management
questions
(1) Attenuating Mason, Porter (1980) Chamberlinean Rational Industries, High Exogenous Structural
competitive Bain firms, conditions and
forces products competitor
positioning
(2) Strategic Machiavelli, Ghemawat (1986) Chamberlinean Hyper-rational Firms, Often Endogenous Strategic
conflict Schelling, Shapiro (1989) products infinite interactions
Cournot, Brandenburger and
Nash, Nalebuff (1995)
Harsanyi,
Shapiro
(3) Resource-based Penrose, Rumelt (1984) Ricardian Rational Resources Low Endogenous Asset
perspectives Selznick, Chandler (1966) fungibility
Christensen, Wernerfelt (1984)
Andrews Teece (1980, 1982)
(4) Dynamic Schumpeter, Dosi, Teece, and Schumpeterian Rational Processes, Low Endogenous Asset
capabilities Nelson, Winter (1989) positions, accumulation,
perspective Winter, Prahalad and paths replicability
Teece Hamel (1990) and
Hayes and inimitability
Wheelwright (1984)
Dierickx and
Cool (1989)
Porter (1990)

in special circumstances, too much ‘strategizing’Efficiency vs. market power
can lead firms to underinvest in core competences
and neglect dynamic capabilities, and thus harmThe competitive forces and strategic conflict
approaches generally see profits as stemming long-term competitiveness.
from strategizing—that is, from limitations on
competition which firms achieve through raising
Normative implications
rivals’ costs and exclusionary behavior (Teece,
1984). The competitive forces approach in The field of strategic management is avowedly
normative. It seeks to guide those aspects ofparticular leads one to see concentrated industries
as being attractive—market positions can be general management that have material effects on
the survival and success of the business enter-shielded behind entry barriers, and rivals costs
can be raised. It also suggests that the sources prise. Unless these various approaches differ in
terms of the framework and heuristics they offerof competitive advantage lie at the level of the
industry, or possibly groups within an industry. management, then the discourse we have gone
through is of limited immediate value. In thisIn text book presentations, there is almost no
attention at all devoted to discovering, creating, paper, we have already alluded to the fact that
the capabilities approach tends to steer managersand commercializing new sources of value.
The dynamic capabilities and resources toward creating distinctive and difficult-to-imitate
advantages and avoiding games with customersapproaches clearly have a different orientation.
They see competitive advantage stemming from and competitors. We now survey possible differ-
ences, recognizing that the paradigms are still inhigh-performance routines operating ‘inside the
firm,’ shaped by processes and positions. Path their infancy and cannot confidently support
strong normative conclusions.dependencies (including increasing returns) and
technological opportunities mark the road ahead.
Because of imperfect factor markets, or more
Unit of analysis and analytic focus
precisely the nontradability of ‘soft’ assets like
values, culture, and organizational experience, Because in the capabilities and the resources
framework business opportunities flow from adistinctive competences and capabilities generally
cannot be acquired; they must be built. This firm’s unique processes, strategy analysis must
be situational.61
This is also true with the strategicsometimes takes years—possibly decades. In
some cases, as when the competence is protected conflict approach. There is no algorithm for
creating wealth for the entire industry. Prescrip-by patents, replication by a competitor is ineffec-
tual as a means to access the technology. The tions they apply to industries or groups of firms
at best suggest overall direction, and may indicatecapabilities approach accordingly sees definite
limits on strategic options, at least in the short errors to be avoided. In contrast, the competitive
forces approach is not particularly firm specific;run. Competitive success occurs in part because
of policies pursued and experience and efficiency it is industry and group specific.
obtained in earlier periods.
Competitive success can undoubtedly flow
Strategic change
from both strategizing and economizing,59
but
along with Williamson (1991) we believe that The competitive forces and the strategic conflict
approach, since they pay little attention to skills,‘economizing is more fundamental than strategiz-
ing . . . . or put differently, that economy is the know-how, and path dependency, tend to see
best strategy.’60
Indeed, we suggest that, except
Williamson’s as it embraces more than efficient contract59
Phillips (1971) and Demsetz (1974) also made the case that
market concentration resulted from the competitive success design and the minimization of transactions costs. We also
address production and organizational economies, and theof more efficient firms, and not from entry barriers and
restrictive practices. distinctive ways that things are accomplished inside the busi-
ness enterprise.60
We concur with Williamson that economizing and strategiz-
ing are not mutually exclusive. Strategic ploys can be used 61
On this point, the strategic conflict and the resources and
capabilities are congruent. However, the aspects of ‘situation’to disguise inefficiencies and to promote economizing out-
comes, as with pricing with reference to learning curve costs. that matter are dramatically different, as described earlier in
this paper.Our view of economizing is perhaps more expansive than

strategic choice occurring with relative facility. cal subfield in its industry, following a techno-
logical discontinuity. Additionally, the interactionThe capabiliies approach sees value augmenting
strategic change as being difficult and costly. between specialized assets such as firm-specific
capabilities and rivalry had the greatest influenceMoreover, it can generally only occur incremen-
tally. Capabilities cannot easily be bought; they on entry timing.
must be built. From the capabilities perspective,
strategy involves choosing among and committing
Diversification
to long-term paths or trajectories of competence
development. Related diversification—that is, diversification that
builds upon or extends existing capabilities—isIn this regard, we speculate that the dominance
of competitive forces and the strategic conflict about the only form of diversification that a
resources/capabilities framework is likely to viewapproaches in the United States may have some-
thing to do with observed differences in strategic as meritorious (Rumelt, 1974; Teece, 1980, 1982;
Teece et al., 1994). Such diversification will beapproaches adopted by some U.S. and some for-
eign firms. Hayes (1985) has noted that American justifiable when the firms’ traditional markets
decline.62
The strategic conflict approach is likelycompanies tend to favor ‘strategic leaps’ while,
in contrast, Japanese and German companies tend to be a little more permissive; acquisitions that
raise rivals’ costs or enable firms to effectuateto favor incremental, but rapid, improvements.
exclusive arrangements are likely to be seen as
efficacious in certain circumstances.
Entry strategies
Here the resources and the capabilities approaches
Focus and specialization
suggest that entry decisions must be made with
reference to the competences and capabilities Focus needs to be defined in terms of distinctive
competences or capability, not products. Productswhich new entrants have, relative to the compe-
tition. Whereas the other approaches tell you little are the manifestation of competences, as com-
petences can be molded into a variety of products.about where to look to find likely entrants, the
capabilities approach identifies likely entrants. Product market specialization and decentalization
configured around product markets may causeRelatedly, whereas the entry deterrence approach
suggests an unconstrained search for new business firms to neglect the development of core com-
petences and dynamic capabilities, to the extentopportunities, the capabilities approach suggests
that such opportunities lie close in to one’s exist- to which competences require accessing assets
across divisions.ing business. As Richard Rumelt has explained
it in conversation, ‘the capabilities approach sug- The capabilities approach places emphasis on
the internal processes that a firm utilizes, as wellgests that if a firm looks inside itself, and at its
market environment, sooner or later it will find as how they are deployed and how they will
evolve. The approach has the benefit of indicatinga business opportunity.’
that competitive advantage is not just a function
of how one plays the game; it is also a function
Entry timing
of the ‘assets’ one has to play with, and how
these assets can be deployed and redeployed inWhereas the strategic conflict approach tells little
abut where to look to find likely entrants, the a changing market.
resources and the capabilities approach identifies
likely entrants and their timing of entry. Brittain
and Freeman (1980) using population ecology
methodologies argued that an organization is
62
Cantwell shows that the technological competence of firmsquick to expand when there is a significant over-
persists over time, gradually evolving through firm-specific
lap between its core capabilities and those needed learning. He shows that technological diversification has been
greater for chemicals and pharmaceuticals than for electricalto survive in a new market. Recent research
and electronic-related fields., and he offers as an explanation(Mitchell, 1989) showed that the more industry-
the greater straight-ahead opportunities in electrical and elec-
specialized assets or capabilities a firm possesses, tronic fields than in chemicals and pharmaceuticals. See
Cantwell (1993).the more likely it is to enter an emerging techni-

Alchian, A. A. and H. Demsetz (1972). ‘Production,Future directions
information costs, and economic organization’,
American Economic Review, 62, pp. 777–795.
We have merely sketched an outline for a Amit, R. and P. Schoemaker (1993). ‘Strategic assets
dynamic capabilities approach. Further theoretical and organizational rent’, Strategic Management
Journal 14(1), pp. 33–46.work is needed to tighten the framework, and
Andrews, K. (1987). The Concept of Corporateempirical research is critical to helping us under-
Strategy (3rd ed.). Dow Jones-Irwin, Homewood, IL.stand how firms get to be good, how they some-
Aoki, M. (1990). ‘The participatory generation of infor-
times stay that way, why and how they improve, mation rents and the theory of the firm’. In M.
and why they sometimes decline.63
Researchers Aoki, B. Gustafsson and O. E. Williamson (eds.),
The Firm as a Nexus of Treaties. Sage, London,in the field of strategy need to join forces with
pp. 26–52.researchers in the fields of innovation, manufac-
Argyres, N. (1995). ‘Technology strategy, governanceturing, and organizational behavior and business
structure and interdivisional coordination’, Journal
history if they are to unlock the riddles that lie of Economic Behavior and Organization, 28,
behind corporate as well as national competitive pp. 337–358.
Argyris, C. and D. Schon (1978). Organizationaladvantage. There could hardly be a more
Learning. Addison-Wesley, Reading, MA.ambitious research agenda in the social sciences
Arrow, K. (1969). ‘The organization of economictoday.
activity: Issues pertinent to the choice of market vs.
nonmarket allocation’. In The Analysis and Evalu-
ation of Public Expenditures: The PPB System, 1.
U.S. Joint Economic Committee, 91st Session. U.S.ACKNOWLEDGEMENTS
Government Printing Office, Washington, DC,
pp. 59–73.Research for this paper was aided by support
Arrow, K. (1996) ‘Technical information and industrial
from the Alfred P. Sloan Foundation through the structure’, Industrial and Corporate Change, 5(2),
Consortium on Competitiveness and Cooperation pp. 645–652.
Arthur, W. B. (1983). ‘Competing technologies andat the University of California, Berkeley. The
lock-in by historical events: The dynamics of allo-authors are grateful for helpful comments from
cation under increasing returns’, working paper WP-two anonymous referees, as well as from Raffi
83-90, International Institute for Applied Systems
Amit, Jay Barney, Joseph Bower, Henry Ches- Analysis, Laxenburg, Austria.
brough, Giovanni Dosi, Sumantra Goshal, Pankaj Bain, J. S. (1959). Industrial Organization. Wiley,
New York.Ghemawat, Connie Helfat, Rebecca Henderson,
Baldwin, C. and K. Clark (1991). ‘Capabilities andDan Levinthal, Richard Nelson, Margie Peteraf,
capital investment: New perspectives on capitalRichard Rosenbloom, Richard Rumelt, Carl
budgeting’, Harvard Business School working paper
Shapiro, Oliver Williamson, and Sidney Winter. #92–004.
Useful feedback was obtained from workshops at Barney, J. B. (1986). ‘Strategic factor markets: Expec-
tations, luck, and business strategy’, Managementthe Haas School of Business, the Wharton School,
Science 32(10), pp. 1231–1241.the Kellogg School (Northwestern), the Harvard
Barney, J. B., J.-C. Spender and T. Reve (1994).Business School, and the International Institute
Crafoord Lectures, Vol. 6. Chartwell-Bratt, Bromley,
of Applied Systems Analysis (IIASA) in Vienna, U.K. and Lund University Press, Lund, Sweden.
the London School of Economics, and the London Baumol, W., J. Panzar and R. Willig (1982). Contest-
able Markets and the Theory of Industry Structure.Business School.
Harcourt Brace Jovanovich, New York.
Bercovitz, J. E. L., J. M. de Figueiredo and D. J. Teece
(1996). ‘Firm capabilities and managerial decision-REFERENCES
making: A theory of innovation biases’. In R. Garud,
P. Nayyar and Z. Shapira (eds), Innovation: Over-Abernathy, W. J. and K. Clark (1985). ‘Innovation:
sights and Foresights. Cambridge University Press,Mapping the winds of creative destruction’,
Cambridge, U.K. pp. 233–259.Research Policy, 14, pp. 3–22.
Brandenburger, A. M. and B. J. Nalebuff (1996). Co-
opetition. Doubleday, New York.
63
For a gallant start, see Miyazaki (1995) and McGrath Brandenburger, A. M. and B. J. Nalebuff (1995). ‘The
et al. (1996). Chandler’s (1990) work on scale and scope, right game: Use game theory to shape strategy’,
summarized in Teece (1993), provides some historical support Harvard Business Review, 73(4), pp. 57–71.
for the capabilities approach. Other relevant studies can be Brittain, J. and J. Freeman (1980). ‘Organizational
found in a special issue of Industrial and Corporate Change
proliferation and density-dependent selection’. In J.3(3), 1994, that was devoted to dynamic capabilities.
R. Kimberly and R. Miles (eds.), The Organizational

Life Cycle. Jossey-Bass, San Francisco, CA, Ghemawat, P. (1986). ‘Sustainable advantage’, Harvard
Business Review, 64(5), pp. 53–58.pp. 291–338.
Camp, R. (1989). Benchmarking: The Search for Indus- Ghemawat, P. (1991). Commitment: The Dynamics of
Strategy. Free Press, New York.try Best practices that Lead to Superior Perform-
ance. Quality Press, Milwaukee, WI. Gilbert, R. J. and D. M. G. Newberry (1982). ‘Preemp-
tive patenting and the persistence of monopoly’,Cantwell, J. (1993). ‘Corporate technological speciali-
zation in international industries’. In M. Casson American Economic Review, 72, pp. 514–526.
Gittell, J. H. (1995). ‘Cross functional coordination,and J. Creedy (eds.), Industrial Concentration and
Economic Inequality. Edward Elgar, Aldershot, control and human resource systems: Evidence from
the airline industry’, unpublished Ph.D. thesis, Mas-pp. 216–232.
Chandler, A.D., Jr. (1966). Strategy and Structure. sachusetts Institute of Technology.
Hansen, G. S. and B. Wernerfelt (1989). ‘DeterminantsDoubleday, Anchor Books Edition, New York.
Chandler, A. D., Jr. (1990). Scale and Scope: The of firm performance: The relative importance of
economic and organizational factors’, Strategic Man-Dynamics of Industrial Competition. Harvard Uni-
versity Press, Cambridge, MA. agement Journal, 10(5), pp. 399–411.
Hartley, R. F. (1989). Marketing Mistakes. Wiley,Clark, K. and T. Fujimoto (1991). Product Develop-
ment Performance: Strategy, Organization and Man- New York.
Hayes, R. (1985). ‘Strategic planning: Forward inagement in the World Auto Industries. Harvard
Buiness School Press, Cambridge, MA. reverse’, Harvard Business Review, 63(6),
pp. 111–119.Coase, R. (1937). ‘The nature of the firm’, Economica,
4, pp. 386–405. Hayes, R. and K. Clark (1985). ‘Exploring the sources
of productivity differences at the factory level’. InCoase, R. (1988). ‘Lecture on the Nature of the Firm,
III’, Journal of Law, Economics and Organization, K. Clark, R. H. Hayes and C. Lorenz (eds.), The
Uneasy Alliance: Managing the Productivity–4, pp. 33–47.
Cool, K. and D. Schendel (1988). ‘Performance differ- Technology Dilemma. Harvard Business School
Press, Boston, MA, pp. 151–188.ences among strategic group members’, Strategic
Management Journal, 9(3), pp. 207–223. Hayes, R. and S. Wheelwright (1984). Restoring our
Competitive Edge: Competing Through Manufactur-de Figueiredo, J. M. and D. J. Teece (1996). ‘Mitigat-
ing procurement hazards in the context of inno- ing. Wiley, New York.
Hayes, R., S. Wheelwright and K. Clark (1988).vation’, Industrial and Corporate Change, 5(2),
pp. 537–559. Dynamic Manufacturing: Creating the Learning
Organization. Free Press, New York.Demsetz, H. (1974). ‘Two systems of belief about
monopoly’. In H. Goldschmid, M. Mann and J. F. Henderson, R. M. (1994). ‘The evolution of integrative
capability: Innovation in cardiovascular drug dis-Weston (eds.), Industrial Concentration: The New
Learning. Little, Brown, Boston, MA, pp. 161–184. covery’, Industrial and Corporate Change, 3(3),
pp. 607–630.Dierickx, I. and K. Cool (1989). ‘Asset stock accumu-
lation and sustainability of competitive advantage’, Henderson, R. M. and K. B. Clark (1990). ‘Architec-
tural innovation: The reconfiguration of existingManagement Science, 35(12), pp. 1504–1511.
Dixit, A. (1980). ‘The role of investment in entry product technologies and the failure of established
firms’, Administrative Science Quarterly, 35,deterrence’, Economic Journal, 90, pp. 95–106.
Dosi, G., D. J. Teece and S. Winter (1989). ‘Toward a pp. 9–30.
Henderson, R. M. and I. Cockburn (1994). ‘Measuringtheory of corporate coherence: Preliminary remarks’,
unpublished paper, Center for Research in Manage- competence? Exploring firm effects in pharmaceut-
ical research, Strategic Management Journal, Sum-ment, University of California at Berkeley.
Doz, Y. and A. Shuen (1990). ‘From intent to outcome: mer Special Issue, 15, pp. 63–84.
Iansiti, M. and K. B. Clark (1994). ‘Integration andA process framework for partnerships’, INSEAD
working paper. dynamic capability: Evidence from product develop-
ment in automobiles and mainframe computers’,Fama, E. F. (1980). ‘Agency problems and the theory
of the firm’, Journal of Political Economy, 88, Industrial and Corporate Change, 3(3), pp. 557–
605.pp. 288–307.
Freeman, J. and W. Boeker (1984). ‘The ecological Itami, H. and T. W. Roehl (1987). Mobilizing Invisible
Assets. Harvard University Press, Cambridge, MA.analysis of business strategy’. In G. Carroll and D.
Vogel (eds.), Strategy and Organization. Pitman, Jacobsen, R. (1988). ‘The persistence of abnormal
returns’, Strategic Management Journal, 9(5),Boston, MA, pp. 64–77.
Fujimoto, T. (1994). ‘Reinterpreting the resource-capa- pp. 415–430.
Katz, M. and C. Shapiro (1985). ‘Network externalities,bility view of the firm: A case of the development-
production systems of the Japanese automakers’, competition and compatibility’, American Economic
Review, 75, pp. 424–440.draft working paper, Faculty of Economics, Univer-
sity of Tokyo. Kreps, D. M. and R. Wilson (1982a). ‘Sequential equi-
libria’, Econometrica, 50, pp. 863–894.Garvin, D. (1988). Managing Quality. Free Press,
New York. Kreps, D. M. and R. Wilson (1982b). ‘Reputation and
imperfect information’, Journal of Economic Theory,Garvin, D. (1994). ‘The processes of organization and
management’, Harvard Business School working 27, pp. 253–279.
Langlois, R. (1994). ‘Cognition and capabilities:paper #94–084.

Opportunities seized and missed in the history of Mitchell, W. (1989). ‘Whether and when? Probability
and timing of incumbents’ entry into emergingthe computer industry’, working paper, University
of Connecticut. Presented at the conference on industrial subfields’, Administrative Science Quar-
terly, 34, pp. 208–230.Technological Oversights and Foresights, Stern
School of Business, New York University, 11–12 Miyazaki, K. (1995). Building Competences in the
Firm: Lessons from Japanese and European Optoe-March 1994.
Learned, E., C. Christensen, K. Andrews and W. Guth lectronics. St. Martins Press, New York.
Mody, A. (1993). ‘Learning through alliances’, Journal(1969). Business Policy: Text and Cases. Irwin,
Homewood, IL. of Economic Behavior and Organization, 20(2),
pp. 151–170.Leonard-Barton, D. (1992). ‘Core capabilities and core
rigidities: A paradox in managing new product Nelson, R. R. (1991). ‘Why do firms differ, and how
does it matter?’ Strategic Management Journal,development’, Strategic Management Journal, Sum-
mer Special Issue, 13, pp. 111–125. Winter Special Issue, 12, pp. 61–74.
Nelson, R. R. (1994). ‘The co-evolution of technology,Leonard-Barton, D. (1995). Wellsprings of Knowledge.
Harvard Business School Press, Boston, MA. industrial structure, and supporting institutions’,
Industrial and Corporate Change, 3(1), pp. 47–63.Lev, B. and T. Sougiannis (1992). ‘The capitalization,
amortization and value-relevance of R&D’, unpub- Nelson, R. (1996). ‘The evolution of competitive or
comparative advantage: A preliminary report on alished manuscript, University of California, Berke-
ley, and University of Illinois, Urbana–Champaign. study’, WP-96-21, International Institute for Applied
Systems Analysis, Laxemberg, Austria.Levinthal, D. and J. March (1981). ‘A model of adap-
tive organizational search’, Journal of Economic Nelson, R. and S. Winter (1982). An Evolutionary
Theory of Economic change. Harvard UniversityBehavior and Organization, 2, pp. 307–333.
Levinthal, D. A. and J. G. March (1993). ‘The myopia Press, Cambridge, MA.
Penrose, E. (1959). The Theory of the Growth of theof learning’, Strategic Management Journal, Winter
Special Issue, 14, pp. 95–112. Firm. Basil Blackwell, London.
Phillips, A. C. (1971). Technology and Market Struc-Levinthal, D. and J. Myatt (1994). ‘Co-evolution of
capabilities and industry: The evolution of mutual ture. Lexington Books, Toronto.
Pisano, G. (1994). ‘Knowledge integration and thefund processing’, Strategic Management Journal,
Winter Special Issue, 15, pp. 45–62. locus of learning: An empirical analysis of process
development’, Strategic Management Journal, Win-Levitt, B. and J. March (1988). ‘Organizational learn-
ing’, Annual Review of Sociology, 14, pp. 319–340. ter Special Issue, 15, pp. 85–100.
Porter, M. E. (1980). Competitive Strategy. Free Press,Link, A. N., D. J. Teece and W. F. Finan (October
1996). ‘Estimating the benefits from collaboration: New York.
Porter, M. E. (1990). The Competitive Advantage ofThe Case of SEMATECH’, Review of Industrial
Organization, 11, pp. 737–751. Nations. Free Press, New York.
Prahalad, C. K. and G. Hamel (1990). ‘The coreLippman, S. A. and R. P. Rumelt (1992) ‘Demand
uncertainty and investment in industry-specific capi- competence of the corporation’, Harvard Business
Review, 68(3), pp. 79–91.tal’, Industrial and Corporate Change, 1(1),
pp. 235–262. Richardson, G. B. H. (1960, 1990). Information and
Investment. Oxford University Press, New York.Mahoney, J. (1995). ‘The management of resources and
the resources of management’, Journal of Business Rosenberg, N. (1982). Inside the Black Box: Tech-
nology and Economics. Cambridge University Press,Research, 33(2), pp. 91–101.
Mahoney, J. T. and J. R. Pandian (1992). ‘The resource- Cambridge, MA
Rumelt, R. P. (1974). Strategy, Structure, andbased view within the conversation of strategic man-
agement’, Strategic Management Journal, 13(5), Economic Performance. Harvard University Press,
Cambridge. MA.pp. 363–380.
Mason, E. (1949). ‘The current state of the monopoly Rumelt, R. P. (1984). ‘Towards a strategic theory of
the firm’. In R. B. Lamb (ed.), Competitive Strategicproblem in the U.S.’, Harvard Law Review, 62,
pp. 1265–1285. Management. Prentice-Hall, Englewood Cliffs, NJ,
pp. 556–570.McGrath, R. G., M-H. Tsai, S. Venkataraman and
I. C. MacMillan (1996). ‘Innovation, competitive Rumelt, R. P. (1991). ‘How much does industry
matter?’, Strategic Management Journal, 12(3),advantage and rent: A model and test’, Management
Science, 42(3), pp. 389–403. pp. 167–185.
Rumelt, R. P., D. Schendel and D. Teece (1994).Milgrom, P. and J. Roberts (1982a). ‘Limit pricing and
entry under incomplete information: An equilibrium Fundamental Issues in Strategy. Harvard Business
School Press, Cambridge, MA.analysis’, Econometrica, 50, pp. 443–459.
Milgrom, P. and J. Roberts (1982b). ‘Predation, repu- Schmalensee, R. (1983). ‘Advertising and entry deter-
rence: An exploratory model’, Journal of Politicaltation and entry deterrence’, Journal of Economic
Theory, 27, pp. 280–312. Economy, 91(4), pp. 636–653.
Schumpeter, J. A. (1934). Theory of Economic Devel-Milgrom, P. and J. Roberts (1990). ‘The economics of
modern manufacturing: Technology, strategy, and opment. Harvard University Press, Cambridge, MA.
Schumpeter, J. A. (1942). Capitalism, Socialism, andorganization’, American Economic Review, 80(3),
pp. 511–528. Democracy. Harper, New York.

Shapiro, C. (1989). ‘The theory of business strategy’, of rapid technological progress’, Journal of Economic
Behavior and Organization, 18(1), pp. 1–25.RAND Journal of Economics, 20(1), pp. 125–137.
Shuen, A. (1994). ‘Technology sourcing and learning Teece, D. J. (1993). ‘The dynamics of industrial capi-
talism: Perspectives on Alfred Chandler’s Scale andstrategies in the semiconductor industry’, unpub-
lished Ph.D. dissertation, University of California, Scope (1990)’, Journal of Economic Literature,
31(1), pp. 199–225.Berkeley.
Sutton, J. (1992). ‘Implementing game theoretical mod- Teece, D. J. (1996) ‘Firm organization, industrial struc-
ture, and technological innovation’, Journal of Econ-els in industrial economies’, In A. Del Monte (ed.),
Recent Developments in the Theory of Industrial omic Behavior and Organization, 31, pp. 193–224.
Teece, D. J. and G. Pisano (1994). ‘The dynamicOrganization. University of Michigan Press, Ann
Arbor, MI, pp. 19–33. capabilities of firms: An introduction’, Industrial
and Corporate Change, 3(3), pp. 537–556.Szulanski, G. (1995). ‘Unpacking stickiness: An empiri-
cal investigation of the barriers to transfer best Teece, D. J., R. Rumelt, G. Dosi and S. Winter (1994).
‘Understanding corporate coherence: Theory and evi-practice inside the firm’, Academy of Management
Journal, Best Papers Proceedings, pp. 437–441. dence’, Journal of Economic Behavior and Organi-
zation, 23, pp. 1–30.Teece, D. J. (1976). The Multinational Corporation
and the Resource Cost of International Technology Tushman, M. L., W. H. Newman and E. Romanelli
(1986). ‘Convergence and upheaval: Managing theTransfer. Ballinger, Cambridge, MA.
Teece, D. J. (1980). ‘Economics of scope and the unsteady pace of organizational evolution’, Califor-
nia Management Review, 29(1), pp. 29–44.scope of the enterprise’, Journal of Economic
Behavior and Organization, 1, pp. 223–247. Wernerfelt, B. (1984). ‘A resource-based view of the
firm’, Strategic Management Journal, 5(2),Teece, D. J. (1981). ‘The market for know-how and
the efficient international transfer of technology’, pp. 171–180.
Wernerfelt, B. and C. Montgomery (1988). ‘Tobin’s QAnnals of the Academy of Political and Social
Science, 458, pp. 81–96. and the importance of focus in firm performance’,
American Economic Review, 78(1), pp. 246–250.Teece, D. J. (1982). ‘Towards an economic theory of
the multiproduct firm’, Journal of Economic Williamson, O. E. (1975). Markets and Hierarchies.
Free Press, New York.Behavior and Organization, 3, pp. 39–63.
Teece, D. J. (1984). ‘Economic analysis and strategic Williamson, O. E. (1985). The Economic Institutions
of Capitalism. Free Press, New York.management’, California Management Review,
26(3), pp. 87–110. Williamson, O. E. (1991). ‘Strategizing, economizing,
and economic organization’, Strategic ManagementTeece, D. J. (1986a). ‘Transactions cost economics and
the multinational enterprise’, Journal of Economic Journal, Winter Special Issue, 12, pp. 75–94.
Williamson, O. E. (1996) The Mechanisms of Govern-Behavior and Organization, 7, pp. 21–45.
Teece, D. J. (1986b). ‘Profiting from technological ance. Oxford University Press, New York.
Womack, J., D. Jones and D. Roos (1991). Theinnovation’, Research Policy, 15(6), pp. 285–305.
Teece, D. J. 1988. ‘Technological change and the nature Machine that Changed the World. Harper-Perennial,
New York.of the firm’. In G. Dosi, C. Freeman, R. Nelson,
G. Silverberg and L. Soete (eds.), Technical Change Zander, U. and B. Kogut (1995). ‘Knowledge and the
speed of the transfer and imitation of organizationaland Economic Theory. Pinter Publishers, New York,
pp. 256–281. capabilities: An empirical test’, Organization
Science, 6(1), pp. 76–92.Teece, D. J. (1992). ‘Competition, cooperation, and
innovation: Organizational arrangements for regimes

10. teece pisano dynamic capabilities

More Related Content

What's hot (19)

Similar to 10. teece pisano dynamic capabilities (20)

More from Facultad Cea (14)

Recently uploaded (20)

10. teece pisano dynamic capabilities