Sie sind auf Seite 1von 32

Strategic Management Journal

Strat. Mgmt. J., 28: 13191350 (2007)

Published online 7 August 2007 in Wiley InterScience ( DOI: 10.1002/smj.640
Received 16 February 2004; Final revision received 20 June 2007


Institute of Management, Innovation and Organization, Haas School of Business,
University of California, Berkeley, California, U.S.A.

This paper draws on the social and behavioral sciences in an endeavor to specify the nature
and microfoundations of the capabilities necessary to sustain superior enterprise performance
in an open economy with rapid innovation and globally dispersed sources of invention, innova-
tion, and manufacturing capability. Dynamic capabilities enable business enterprises to create,
deploy, and protect the intangible assets that support superior long- run business performance.
The microfoundations of dynamic capabilitiesthe distinct skills, processes, procedures, orga-
nizational structures, decision rules, and disciplineswhich undergird enterprise-level sensing,
seizing, and reconfiguring capacities are difficult to develop and deploy. Enterprises with strong
dynamic capabilities are intensely entrepreneurial. They not only adapt to business ecosystems,
but also shape them through innovation and through collaboration with other enterprises, enti-
ties, and institutions. The framework advanced can help scholars understand the foundations of
long-run enterprise success while helping managers delineate relevant strategic considerations
and the priorities they must adopt to enhance enterprise performance and escape the zero profit
tendency associated with operating in markets open to global competition. Copyright 2007
John Wiley & Sons, Ltd.

INTRODUCTION fast-moving business environments open to global

competition, and characterized by dispersion in the
Recent scholarship stresses that business enter- geographical and organizational sources of inno-
prises consist of portfolios of idiosyncratic and vation and manufacturing, sustainable advantage
difficult-to-trade assets and competencies (re- requires more than the ownership of difficult-
sources).1 Within this framework, competitive to-replicate (knowledge) assets. It also requires
advantage can flow at a point in time from the unique and difficult-to-replicate dynamic capabili-
ownership of scarce but relevant and difficult-to- ties. These capabilities can be harnessed to con-
imitate assets, especially know-how. However, in tinuously create, extend, upgrade, protect, and
keep relevant the enterprises unique asset base.
Keywords: cospecialization; intangible assets; innovation; For analytical purposes, dynamic capabilities can
business ecosystems; entrepreneurship; managerial capi- be disaggregated into the capacity (1) to sense
talism; global competitiveness and shape opportunities and threats, (2) to seize
*Correspondence to: David J. Teece, F402 Haas School of
Business #1930, University of California, Berkeley, California opportunities, and (3) to maintain competitiveness
94720-1930, U.S.A. E-mail: through enhancing, combining, protecting, and,
The reference here is to the resource-based theory of the when necessary, reconfiguring the business enter-
enterprise advanced by Rumelt (1984), Wernerfelt (1984), Amit
and Schoemaker (1993), and others. Some of my earlier work prises intangible and tangible assets. Dynamic
(Teece, 1980, 1982) was also in this vein. capabilities include difficult-to-replicate enterprise

Copyright 2007 John Wiley & Sons, Ltd.

1320 D. J. Teece

capabilities required to adapt to changing cus- that multiple inventions must be combined to cre-
tomer and technological opportunities. They also ate products and/or services that address customer
embrace the enterprises capacity to shape the needs. The third is that there are well-developed
ecosystem it occupies, develop new products and global markets for the exchange of (component)
processes, and design and implement viable busi- goods and services; and the fourth is that the busi-
ness models. It is hypothesized that excellence ness environment is characterized by poorly devel-
in these orchestration2 capacities undergirds an oped markets in which to exchange technological
enterprises capacity to successfully innovate and and managerial know-how. These characteristics
capture sufficient value to deliver superior long- can be found in large sectors of the global econ-
term financial performance. The thesis advanced is omy and especially in high-technology sectors. In
that while the long-run performance of the enter- such sectors, the foundations of enterprise success
prise is determined in some measure by how the today depend very little on the enterprises abil-
(external) business environment rewards its her- ity to engage in (textbook) optimization against
itage, the development and exercise of (internal) known constraints, or capturing scale economies
dynamic capabilities lies at the core of enterprise in production. Rather, enterprise success depends
success (and failure). This paper first describes the upon the discovery and development of opportuni-
nature of dynamic capabilities, and then explicates ties; the effective combination of internally gener-
their microfoundations. ated and externally generated inventions; efficient
The ambition of the dynamic capabilities frame- and effective technology transfer inside the enter-
work is nothing less than to explain the sources of prise and between and amongst enterprises; the
enterprise-level competitive advantage over time, protection of intellectual property; the upgrading
and provide guidance to managers for avoiding the of best practice business processes; the inven-
zero profit condition that results when homoge- tion of new business models; making unbiased
neous firms compete in perfectly competitive mar- decisions; and achieving protection against imita-
kets. A framework, like a model, abstracts from tion and other forms of replication by rivals. It
reality. It endeavors to identify classes of relevant also involves shaping new rules of the game in
variables and their interrelationships. A framework the global marketplace. The traditional elements
is less rigorous than a model as it is sometimes of business successmaintaining incentive align-
agnostic about the particular form of the theoreti- ment, owning tangible assets, controlling costs,
cal relationships that may exist. Early statements of maintaining quality, optimizing inventoriesare
the dynamic capabilities framework can be found necessary but they are unlikely to be sufficient for
in Teece, Pisano, and Shuen (1990a, 1990b, 1997) sustained superior enterprise performance.
and Teece and Pisano (1994). An extensive lit- Executives seem to recognize new challenges
in todays globally competitive environments and
erature on dynamic capabilities now exists (e.g.,
understand how technological innovation is nec-
Helfat et al., 2007) that can be organized and inte-
essary but not sufficient for success. A. J. Lafley,
grated into the general framework offered here.
CEO of Proctor & Gamble, notes that the name
As indicated, the possession of dynamic capabil-
of the game is innovation. We work really hard to
ities is especially relevant to multinational enter-
try to turn innovation into a strategy and a process
prise performance in business environments that
. . . .3 Sam Pamisano, CEO of IBM, remarks that
display certain characteristics. The first is that the
innovation is about much more than new prod-
environment is open to international commerce and
ucts. It is about reinventing business processes and
fully exposed to the opportunities and threats asso-
building entirely new markets that meet untapped
ciated with rapid technological change. The sec- customer demand.4 Put differently, there is an
ond is that technical change itself is systemic in emerging recognition by managers themselves that
the foundations of enterprise success transcend
simply being productive at R&D, achieving new
The management functions identified are analogous to that of
an orchestra conductor, although in the business context the product introductions, adopting best practice, and
instruments (assets) are themselves constantly being created, delivering quality products and services. Not only
renovated, and/or replaced. Moreover, completely new instru-
ments appear with some frequency, and old ones need to be
abandoned. While flexibility is certainly an element of orches- Fortune, December 11, 2006: 4.
tration, the latter concept implies much more. Business Week, April 24, 2004: 64.

Copyright 2007 John Wiley & Sons, Ltd. Strat. Mgmt. J., 28: 13191350 (2007)
DOI: 10.1002/smj
Explicating Dynamic Capabilities: Nature and Microfoundations 1321

must the innovating enterprise spend heavily on new markets to emerge; and as discussed later, the
R&D and assiduously develop and protect its intel- ubiquity of platforms must now be recognized
lectual property; it must also generate and imple- (Evans, Hagiu, and Schmalensee, 2006).
ment the complementary organizational and man- While the development and astute management
agerial innovations needed to achieve and sustain of intangible assets/intellectual capital is increas-
competitiveness. ingly recognized as central to sustained enter-
As indicated, not all enterprise-level responses prise competitiveness, the understanding of why
to opportunities and threats are manifestations of and how intangibles are now so critical still
dynamic capabilities. As Sidney Winter (2003: remains opaque and is not addressed by orthodox
991) notes, ad hoc problem solving isnt neces- frameworks. What is needed is a new framework
sarily a capability. Nor is the adoption of a well- for business and economic analysis. As former
understood and replicable best practice likely to U.S. Federal Reserve Chairman Alan Greenspan
constitute a dynamic capability. Implementing best remarked, we must begin the important work of
practice may help an enterprise become or remain developing a framework capable of analyzing the
viable, but best practices that are already widely growth of an economy increasingly dominated by
adopted cannot by themselves in a competitive conceptual products.5 The dynamic capabilities
market situation enable an enterprise to earn more approach developed here endeavors to be respon-
than its cost of capital, or outperform its competi- sive to this challenge at the enterprise level.
tors. Likewise, invention and innovation by them- In an earlier treatment (Teece et al., 1997: 530)
selves are insufficient to generate success (Teece, it was noted that we have merely sketched an out-
1986). line for a dynamic capabilities approach. In what
Two yardsticks can be proposed for calibrating follows, the nature of various classes of dynamic
capabilities: technical fitness and evolutionary capabilities is identified, and an effort is made to
fitness (Helfat et al., 2007). Technical fitness is separate the microfoundations of dynamic capa-
defined by how effectively a capability performs bilities from the capability itself. Put differently,
its function, regardless of how well the capability important distinctions are made between the orga-
enables a firm to make a living. Evolutionary or nizational and managerial processes, procedures,
external fitness refers to how well the capability systems, and structures that undergird each class
enables a firm to make a living. Evolutionary fit- of capability, and the capability itself. One should
ness references the selection environment. Helfat note that the identification of the microfounda-
et al. (2007) further note that both technical and tions of dynamic capabilities must be necessarily
evolutionary fitness range from zero to some pos- incomplete, inchoate, and somewhat opaque and/or
itive value. These yardsticks are consistent with their implementation must be rather difficult. Oth-
the discussion here. Dynamic capabilities assist in erwise sustainable competitive advantage would
achieving evolutionary fitness, in part by helping erode with the effective communication and appli-
to shape the environment. The element of dynamic cation of dynamic capability concepts.
capabilities that involves shaping (and not just Of course, the existence of processes, proce-
adapting to) the environment is entrepreneurial in dures, systems, and structures already ubiquitously
nature. Arguably, entrepreneurial fitness ought to adopted by competitors does not imply that these
have equal standing with evolutionary fitness. have not in the past been the source of competitive
Dynamic capabilities have no doubt been rele- advantage, or might not still be a source of compet-
vant to achieving competitive advantage for some itive advantage in certain contexts. For example,
time. However, their importance is now ampli- studies of the diffusion of organizational innova-
fied because the global economy has become more tions (e.g., Armour and Teece, 1978; Teece, 1980)
open and the sources of invention, innovation, and
manufacturing are more diverse geographically 5
Chairman Alan Greenspan also noted recently, over the past
and organizationally (Teece, 2000), and multiple half century, the increase in the value of raw materials has
accounted for only a fraction of the overall growth of U.S. gross
inventions must be combined to achieve market- domestic product (GDP). The rest of that growth reflects the
place success (Somaya and Teece, 2007). Achiev- embodiment of ideas in products and services that consumers
ing evolutionary fitness is harder today than it was value. This shift of emphasis from physical materials to ideas as
the core of value creation appears to have accelerated in recent
before the millennium. Moreover, regulatory and decades. (Remarks of Alan Greenspan, Stanford Institute for
institutional structures must often be reshaped for Economic Policy Research, 2004.)

Copyright 2007 John Wiley & Sons, Ltd. Strat. Mgmt. J., 28: 13191350 (2007)
DOI: 10.1002/smj
1322 D. J. Teece

indicate that diffusion is by no means instanta- by Kirzner (1973), entrepreneurs can have differ-
neous, and that profits can persist for many years ential access to existing information. Second, new
before being competed away. Decade-long adop- information and new knowledge (exogenous or
tion cycles for new business structures and pro- endogenous) can create opportunities, as empha-
cedures (e.g., performance measurement systems) sized by Schumpeter (1934). Kirzner stressed how
are not uncommon. Uncertain imitability (Lippman the entrepreneurial function recognizes any dise-
and Rumelt, 1982) may also serve to slow the dif- quilibrium and takes advantage of it. The Kirzner-
fusion process and support persistent differential ian view is that entrepreneurship is the mechanism
performance. by which the economy moves back toward equi-
Fortunately, the existing literature on strategy, librium. Schumpeter, on the other hand, stressed
innovation, and organization and the new literature upsetting the equilibrium. As Baumol (2006: 4)
on dynamic capabilities have identified a panoply notes, the job of Schumpeters entrepreneur is
of processes and routines that can be recognized to destroy all equilibria, while Kirzners works
as providing certain microfoundations for dynamic to restore them. This is the mechanism under-
capabilities. For instance, Eisenhardt and Martin lying continuous industrial evolution and revolu-
(2000) identify cross-functional R&D teams, new tion. Equilibrium is rarely if ever achieved (Shane,
product development routines, quality control rou- 2003). Both forces are relevant in todays econ-
tines, and technology transfer and/or knowledge omy.
transfer routines, and certain performance mea- To identify and shape opportunities, enterprises
surement systems as important elements (micro- must constantly scan, search, and explore across
foundations) of dynamic capabilities. The effort technologies and markets, both local and dis-
here is not designed to be comprehensive, but to tant (March and Simon, 1958; Nelson and Winter,
integrate the strategy and innovation literature and 1982). This activity not only involves investment
provide an umbrella framework that highlights the in research activity and the probing and reprob-
most critical capabilities management needs to sus- ing of customer needs and technological possibili-
tain the evolutionary and entrepreneurial fitness of ties; it also involves understanding latent demand,
the business enterprise. the structural evolution of industries and mar-
kets, and likely supplier and competitor responses.
To the extent that business enterprises can open
SENSING (AND SHAPING) up technological opportunities (through engaging
OPPORTUNITIES AND THREATS in R&D and through tapping into the research
output of others) while simultaneously learning
Nature of the capability about customer needs, they have a broad menu
of commercialization opportunities. Overcoming a
In fast-paced, globally competitive environments, narrow search horizon is extremely difficult and
consumer needs, technological opportunities, and costly for management teams tied to established
competitor activity are constantly in a state of flux. problem-solving competences. Henderson (1994)
Opportunities open up for both newcomers and notes that General Motors (GM), IBM, and Dig-
incumbents, putting the profit streams of incum- ital Equipment Corporation (DEC) encountered
bent enterprises at risk. As discussed in Teece difficulties because they became prisoners of the
et al. (1997), some emerging marketplace trajecto- deeply ingrained assumptions, information filters,
ries are easily recognized. In microelectronics this and problem-solving strategies that made up their
might include miniaturization, greater chip density, world views, turning the solutions that once made
and compression and digitization in information them great into strategic straitjackets.
and communication technology. However, most When opportunities are first glimpsed, entrepre-
emerging trajectories are hard to discern. Sensing neurs and managers must figure out how to inter-
(and shaping) new opportunities is very much a pret new events and developments, which tech-
scanning, creation, learning, and interpretive activ- nologies to pursue, and which market segments
ity. Investment in research and related activities is to target. They must assess how technologies will
usually a necessary complement to this activity. evolve and how and when competitors, suppli-
Opportunities get detected by the enterprise ers, and customers will respond. Competitors may
because of two classes of factors. First, as stressed or may not see the opportunity, and even if they
Copyright 2007 John Wiley & Sons, Ltd. Strat. Mgmt. J., 28: 13191350 (2007)
DOI: 10.1002/smj
Explicating Dynamic Capabilities: Nature and Microfoundations 1323

do they may calibrate it differently. Their actions, assessing customer needs, expressed and latent.
along with those of customers, suppliers, standard- It involves learning, interpretation, and creative
setting bodies, and governments, can also change activity.
the nature of the opportunity and the manner in While certain individuals in the enterprise may
which competition will unfold. have the necessary cognitive and creative skills,
There are also constraints on the rules by which the more desirable approach is to embed scan-
competitive forces will play out. These constraints ning, interpretative, and creative processes inside
are imposed by regulators, standard-setting bod- the enterprise itself. The enterprise will be vulner-
ies, laws, social mores, and business ethics. The able if the sensing, creative, and learning functions
shape of the rules of the game is thus the are left to the cognitive traits of a few individuals.6
result of co-evolution and complex interaction Organizational processes can be put in place inside
between what might be thought of as (business) the enterprise to garner new technical information,
ecosystem participants. Because of uncertainty, tap developments in exogenous science, monitor
entrepreneurs/managers must make informed con- customer needs and competitor activity, and shape
jectures about the path ahead. These conjectures new products and processes opportunities. Infor-
become working hypotheses that can be updated as mation must be filtered, and must flow to those
evidence emerges. Once a new evolutionary path capable of making sense of it. Internal argument
becomes apparent, quick action is needed. and discussion about changing market and tech-
nological reality can be both inductive and deduc-
Microfoundations tive. Hypothesis development, hypothesis testing,
and synthesis about the meaning of information
The literature on entrepreneurship emphasizes that obtained via search are critical functions, and must
opportunity discovery and creation can originate be performed by the top management team. The
from the cognitive and creative (right brain) rigorous assembly of data, facts, and anecdotes can
capacities of individual(s). However, discovery can help test beliefs. Once a synthesis of the evidence
also be grounded in organizational processes, such is achieved, recurrent synthesis and updating can
as research and development activity. The ability be embedded in business processes designed by
to create and/or sense opportunities is clearly not middle management and/or the planning unit in
uniformly distributed amongst individuals or enter- the business organization (Casson, 1997). If enter-
prises. Opportunity creation and/or discovery by prises fail to engage in such activities, they wont
individuals require both access to information and be able to assess market and technological devel-
the ability to recognize, sense, and shape devel- opments and spot opportunities. As a consequence,
opments. The ability to recognize opportunities they will likely miss opportunities visible to others.
depends in part on the individuals capabilities and As noted in Teece et al. (1997), more decen-
extant knowledge (or the knowledge and learning tralized organizations with greater local autonomy
capacities of the organization to which the indi- are less likely to be blindsided by market and
vidual belongs) particularly about user needs in technological developments. Because of the prob-
relationship to existing as well as novel solutions. lem of information decay as information moves
This requires specific knowledge, creative activity, up (and down) a hierarchy, businesses must devise
and the ability to understand user/customer deci- mechanisms and procedures to keep management
sion making, and practical wisdom (Nonaka and
informed. Bill Hewlett and David Packard devel-
Toyama, 2007). It involves interpreting available
oped management by walking about (Packard,
information in whatever form it appearsa chart,
1995) as a mechanism to prevent top management
a picture, a conversation at a trade show, news of
at Hewlett-Packard from becoming isolated from
scientific and technological breakthroughs, or the
angst expressed by a frustrated customer. One must
accumulate and then filter information from profes- 6
In a limited sense, that is about decision making under uncer-
sional and social contacts to create a conjecture or tainty. As Knight observes, with uncertainty there is a necessity
a hypothesis about the likely evolution of technolo- to act upon opinion rather than knowledge (Knight, 1921: 268).
gies, customer needs, and marketplace responses. The problem is not just about knowledge asymmetries and incen-
tive problems as Alchian and Demsetz (1972) seem to suggest.
This task involves scanning and monitoring inter- Rather, it involves filtering and interpreting information about
nal and external technological developments and evolving technologies and marketplaces.

Copyright 2007 John Wiley & Sons, Ltd. Strat. Mgmt. J., 28: 13191350 (2007)
DOI: 10.1002/smj
1324 D. J. Teece

what was going on at lower levels in the enter- technology/components in a timely fashion will
prise, and outside the enterprise as well. In other lend to failure; conversely, success can some-
organizations (e.g., professional services) the man- times be achieved by continuous rapid design in.
agement ranks can be filled by leading profession- Indeed, continuous and rapid design around new
als who remain involved with professional work. technology/components developed elsewhere can
This protects them from the hazards of managerial itself be a source of durable competitive advan-
isolation. tage. Put differently, with rapid innovation by com-
The search activities that are relevant to sens- ponent suppliers, downstream competitive success
ing include information about whats going on in can flow from the ability of enterprises to continu-
the business ecosystem. With respect to technolo- ously tap into such (external) innovation ahead of
gies, R&D activity can itself be thought of as a the competition. External search and acquisition of
form of search for new products and processes. technology have been going on for decades, but as
However, R&D is too often usually a manifesta- Chesbrough (2003) explains, Open Innovation is
tion of local search. Local search is only one now a mandate for enterprise success.
component of relevant search. In fast-paced envi- The concept and practice of open innovation
ronments, with a large percentage of new prod- underscore the importance of broad-based external
uct introductions coming from external sources, search and subsequent integration involving cus-
search/exploration activity should not just be local. tomers, suppliers, and complementors. Establish-
Enterprises must search the core as well as to ing linkages between corporations and universities
the periphery of their business ecosystem. Search assists broad-based search, as university programs
must embrace potential collaboratorscustomers, are usually unshackled from the near at hand.
suppliers, complementorsthat are active in inno- Indeed, a recent study of patenting in the opti-
vative activity. cal disk industry (Rosenkopf and Nerkar, 2001)
Customers are sometimes amongst the first to seems to suggest that exploration that is more con-
perceive the potential for applying new technol- fined generates lower impacts, and that the impact
ogy. Visionary members of customer organizations of exploration is highest when exploration spans
are often able to anticipate the potential for new organizational (but not technological) boundaries.
technology and possibly even begin rudimentary However, it is not just a matter of searching for
development activities. Moreover, if the suppli- external inventions/innovations that represent new
ers of new technology do not succeed in properly possibilities. Frequently it is a matter of combining
understanding user/customer needs, it is unlikely complementary innovations so as to create a solu-
that new products they might develop will be suc- tion to a customer problem. The systemic nature
cessful. Indeed, one of the most consistent findings (Teece, 2000) of many innovations compounds the
from empirical research is that the probability that need for external search.
an innovation will be successful commercially is Sensing opportunities and threats can also be
highly correlated with the developers understand- facilitated if the enterprise and/or the entrepreneur
ing of user/customer needs (Freeman, 1974). Elec- explicitly or implicitly employ some kind of ana-
tronic computing and the Internet itself can rightly lytical framework, as this can help highlight what
be viewed as having a significant component of is important. The field of strategic management
user-led innovations. Business enterprises that are has been stranded for some time with a frame-
alert and sense the opportunity are often able to work that implicitly assumes that industry struc-
leverage customer-led efforts into new products ture (and product market share), mediated by
and services, as the users themselves are frequently enterprise behavior, determines enterprise perfor-
ill prepared to carry initial prototypes further for- mance. In Porters (1980) Five Forces frame-
ward. work, a good strategy involves somehow picking
Suppliers can also be drivers of innovation an attractive industry and positioning oneself to
important in the final product. Innovation in micro- be shielded from competition. Porters approach
processor and DRAMs is a classic case. This mandates industry analysis7 and the calibration
upstream or component innovation has impacted of five distinct industry-level forces: the role of
competition and competitive outcomes in personal
computers, cellular telephony, and consumer elec- 7
The Five Forces framework undergirds industry analysis in
tronics more generally. Failure to design in new business school curriculum and in practice. However, the very

Copyright 2007 John Wiley & Sons, Ltd. Strat. Mgmt. J., 28: 13191350 (2007)
DOI: 10.1002/smj
Explicating Dynamic Capabilities: Nature and Microfoundations 1325

potential entrants, suppliers, buyers, substitutes, game, (b) for factors inside the business enter-
and rivalry amongst competitors. Because of its prise that constrain choices, (c) for factors that
rather static nature and the fact that it ignores many impact imitation and appropriability issues, (d) for
aspects of the competitive environment includ- the role of supporting institutions, complementary
ing the role of complementarities, path dependen- assets, cospecialization, and network externalities,
cies, and supporting institutions, its application in or (e) for the blurred nature of industry boundaries.
the contexts outlined in the Introduction of this Also, as discussed later, in many platform indus-
paper will limit the ability of the entrepreneur tries or where there is significant outsourcing, scale
and/or the enterprise to properly sense opportu- is an industry asset.
nities and threats and properly calibrate strengths, The dynamic capabilities framework represents
weaknesses, and technological and market trajec- a strong break with Five Forces. Within the
tories. If network effects, path dependencies, and dynamic capabilities framework, the environmen-
the co-evolution of technologies and institutions tal context recognized for analytical purposes is
are significant, the Five Forces framework is of not that of the industry, but that of the busi-
limited utility. ness ecosystemthe community of organiza-
The Five Forces framework has inherent weak- tions, institutions, and individuals that impact the
nesses in dynamic environments. Fundamental is enterprise and the enterprises customers and sup-
that it implicitly views market structure as exoge- plies. The relevant community therefore includes
nous, when in fact market structure is the (endoge- complementors, suppliers, regulatory authorities,
nous) result of innovation and learning.8 Changes standard-setting bodies, the judiciary, and educa-
in science and technology create opportunities for tional and research institutions. It is a framework
innovation. Enterprises can search amongst new that recognizes that innovation and its support-
possibilities and engage in development activities. ing infrastructure have major impacts on com-
If successful, such development impacts the rela- petition. The dynamic capabilities framework is
tive fate of firms. This in turn determines market grounded in Kirznerian, Schumpeterian, and evolu-
structure. Outcomes for individual enterprises are tionary theories of economic change, whereas Five
shaped in part by the selection processes at work in Forces is grounded in the MasonBain paradigm
the business ecosystem. Relevant factors ignored of industrial economics.9 Also, whereas according
or underplayed by Five Forces include techno- to Porter the essence of strategy formulation is
logical opportunities, path dependencies, appropri- coping with competition (Porter, 1991: 11), in the
ability conditions, supporting institutions, installed dynamic capabilities tradition the essence of strat-
base effects, learning, certain switching costs, and egy involves selecting and developing technolo-
regulation. In short, in regimes of rapid technologi- gies and business models that build competitive
cal change with well-developed markets for goods advantage through assembling and orchestrating
and services (and poorly developed markets for difficult-to-replicate assets, thereby shaping com-
know-how), the Five Forces framework is com- petition itself.
promised because it has insufficient appreciation Even when utilizing the ecosystem as the orga-
(a) for the importance of and nature of innova- nizing paradigm for assessing developments in the
tion and other factors that change the rules of the business environment, the full import of particular
facts, statistics, and developments is rarely obvi-
ous. Accordingly, the evaluative and inferential
concept of an industry is itself of questionable value. If indus-
try boundaries exist, they are faint, at least in technologically skill possessed by an organization and its manage-
progressive environments. For instance, the telecommunications ment is important. Indeed, much of the informa-
industry may have had distinct boundaries over half a cen- tion gathered and communicated inside the enter-
tury ago around the telegraph and the telephone and associated
regulated services. However, by the 1960s, facsimile and data prise has minimal decision relevance. Even if rel-
services had begun to be overlaid on the public telephone net- evant, it often arrives too late. Management must
work. Today telephony is routinely carried by the Internet (using find methods and procedures to peer through the
Voice over IP) and cable TV networks.
Indeed, the (basic) market structureconductperformance
paradigm from industrial economics that undergirds the Five Developed in the 1930s, 1940s, and 1950s, it is still relevant
Forces approach has been in need of revision for quite some to some of the rust belt industries that experience low rates of
time. Phillips (1971) was perhaps the first to recognize that cau- technological innovation where complementors are not impor-
sation is the reverse of what is assumed, with market structure tant, and where the coevolution of technologies and institutions
being shaped by innovation. is not significant (Teece, 1990).

Copyright 2007 John Wiley & Sons, Ltd. Strat. Mgmt. J., 28: 13191350 (2007)
DOI: 10.1002/smj
1326 D. J. Teece

Processes to Tap
Processes to Direct
Developments in
Internal R&D and Select
Exogenous Science and
New Technologies.
Analytical Systems (and
Individual Capacities) to
Learn and to Sense, Filter,
Shape, and Calibrate
Processes to Tap Processes to Identify
Supplier and Target Market Segments,
Complementor Changing Customer
Innovation. Needs, and Customer

Figure 1. Elements of an ecosystem framework for sensing market and technological opportunities

fog of uncertainty and gain insight. This involves and Suarez, 1993; Malerba and Orsenigo, 1996).
gathering and filtering technological, market, and It implicitly recognizes inflexion points in tech-
competitive information from both inside and out- nological and market evolution. These inflexion
side the enterprise, making sense of it, and figur- points impact investment requirements and strate-
ing out implications for action. However, because gic choices. Implications for investment decisions
attention is a scarce resource inside the enterprise have been noted elsewhere (Teece, 1986) and
(Cyert and March, 1963), management must care- include staying flexible until the dominant design
fully allocate resources to search and discovery. emerges and then investing heavily once a design
The enterprises articulated strategy can become looks like it can become the winner. Any strat-
a filter so that attention is not diverted to every egy is, of course, likely to be fraught with hazards
opportunity and threat that successful search because of uncertainties. Moreover, the manner
reveals. Likewise, scenario planning can collapse and time at which an enterprise needs to place its
likely situations into a small number of scenarios bets depend on competition in the input markets
that can facilitate cognition, and then action, once and on the identity of the enterprise itself. Mitchell
uncertainty is resolved. Figure 1 summarizes indi- (1991) suggests that the timing of resource com-
vidual and enterprise traits that undergird sensing mitments can differ according to the enterprises
capabilities. existing positions with respect to the relevant com-
plementary assets. Enterprises that are well posi-
tioned can wait, while those that are not must
Addressing opportunities involves maintaining
Nature of the capability and improving technological competences and
Once a new (technological or market) opportunity complementary assets and then, when the opportu-
is sensed, it must be addressed through new prod- nity is ripe, investing heavily in the particular tech-
ucts, processes, or services. This almost always nologies and designs most likely to achieve mar-
requires investments in development and commer- ketplace acceptance. When network externalities
cialization activity. Multiple (competing) invest- are present, early entry and commitment are nec-
ment paths are possible, at least early on. The essary. The presence of increasing returns means
quintessential example is the automobile industry, that if one network gets ahead, it tends to stay
where in the early days different engine technolo- ahead. Getting ahead may require significant up-
giessteam, electric, and gasolineeach had front investments. Customers will not want an
their champions. Once a dominant design begins enterprises products if there are strong network
to emerge, strategic choices become much more effects and the installed base of users is rela-
limited. This paradigm, which was first offered tively small. Accordingly, one needs to strategize
by Abernathy and Utterback (1978) and then built around investment decisions, getting the timing
upon by Teece (1986, 2007), now has considerable right, building on increasing return advantages, and
evidence supporting it over a wide range of tech- leveraging products and services from one applica-
nologies (Klepper and Graddy, 1990; Utterback tion to another. The capacity to make high-quality,
Copyright 2007 John Wiley & Sons, Ltd. Strat. Mgmt. J., 28: 13191350 (2007)
DOI: 10.1002/smj
Explicating Dynamic Capabilities: Nature and Microfoundations 1327

unbiased but interrelated investment decisions in future cash flow can be confidently projected. In
the context of network externalities, innovation, short, the new can lose out to the established unless
and change is as rare as decision-making errors management is sensitive to the presence of cer-
and biases are ubiquitous. tain biases in accepted investment decision pro-
However, the issue that the enterprise faces is cesses. An important class of dynamic capabilities
not just when, where, and how much to invest. emerges around a managers ability to override
The enterprise must also select or create a partic- certain dysfunctional features of established deci-
ular business model that defines its commercial- sion rules and resource allocation processes.
ization strategy and investment priorities. Indeed, It helps to begin by recognizing that decision-
there is considerable evidence that business suc- making processes in hierarchically organized enter-
cess depends as much on organizational innova- prises involve bureaucratic features that are use-
tion, e.g., design of business models, as it does on ful for many purposes, but they nevertheless may
the selection of physical technology. This is true muzzle innovation proclivities. In particular, a for-
at the enterprise level as well as at the economy- mal expenditure process involving submissions
wide level (Nelson, 2005). Indeed, the invention and approvals is characteristic of well-managed
and implementation of business models and asso- companies. Decision making is likely to have a
ciated enterprise boundary choices involve issues committee structure, with top management requir-
as fundamental to business success as the devel- ing reports and written justifications for signif-
opment and adoption of the physical technologies icant decisions. Moreover, approvals may need
themselves. Business models implicate processes to be sought from outside the organizational unit
and incentives; their alignment with the physical in which the expenditure is to take place. While
technology is a much overlooked component of this may ensure a matching up of expenditures
strategic management. The understanding of the to opportunities across a wider range of economic
institutional/organizational design issues is typi- activity, it unquestionably slows decision making
cally more limited than the understanding of the and tends to reinforce the status quo. Commit-
technologies themselves. This ignorance affords tee decision-making structures almost always tend
considerable scope for mistakes around the proper toward balancing and compromise. But innovation
design of business models and the institutional is often ill served by such structures, as the new
structures needed to support innovation in both the and the radical will almost always appear threat-
private and public sectors. ening to some constituents. Strong leaders can fre-
In theory, one could imagine transactions quently overcome such tendencies, but such lead-
between entities that scout out and/or develop ers are not always present. One consequence is a
opportunities, and those that endeavor to execute program persistence bias. Its corollary is various
upon them. In reality, the two functions cannot forms of anti-innovation bias, including the anti-
be cleanly separated, and the activities must be cannibalization basis discussed in a later section.
integrated inside a single enterprise, where new Program persistence refers to the funding of pro-
insights about marketsparticularly those that grams beyond what can be sustained on the merits,
challenge the conventional wisdomwill likely and follows from the presence or influence of pro-
encounter negative responses. The promoters/ gram advocates in the resource allocation process.
visionaries must somehow defeat the naysayers, This proclivity almost automatically has the coun-
transform internal views, and facilitate necessary tervailing effect of reducing funds available to new
investment. Some level of managerial consensus initiatives.
will be necessary to allow investment decisions to One should not be surprised, therefore, if an
be made. Investment will likely involve commit- enterprise senses a business opportunity but fails
ting financial resources behind an informed con- to invest. In particular, incumbent enterprises tend
jecture about the technological and marketplace to eschew radical competency-destroying innova-
future. However, managers of established product tion in favor of more incremental competency-
lines in large organizations can sometimes have enhancing improvements. The existence of layer
sufficient decision-making authority to starve the upon layer of standard procedures, established
new business of financial capital. This posture can capabilities, complementary assets, and/or admin-
be buttressed by capital budgeting techniques that istrative routines can exacerbate decision-making
more comfortably support investments for which biases against innovation. Incumbent enterprises,
Copyright 2007 John Wiley & Sons, Ltd. Strat. Mgmt. J., 28: 13191350 (2007)
DOI: 10.1002/smj
1328 D. J. Teece

relying on (path-dependent) routines, assets, and quality of the enterprises routines, decision rules,
strategies developed to cope with existing tech- strategies, and leadership around evaluating new
nologies, are handicapped in making and/or adopt- investment opportunities. Business historians (e.g.,
ing radical, competency-destroying, noncumula- Chandler, 1990a; Lazonick, 2005) and others have
tive innovation (Nelson and Winter, 1982; Tush- reminded us that over the long run the ability of
man and Anderson, 1986; Henderson and Clark, enterprises to commit financing and invest astutely
1990). This is true whether the competence is around new technologies is critical to enterprise
external to the firm or internal to the firm. performance.10
Evidence also shows that decision-makers dis- In regimes of rapid technological innovation, it
count outcomes that are merely probable in com- is clear that making investment choices requires
parison with outcomes that are certain. This has special skills not ubiquitously distributed amongst
been called the certainty effect (Kahneman and management teams. Nor are they ubiquitously dis-
Lovallo, 1993). It contributes to excessive risk tributed amongst investors.11 Resource/asset align-
aversion when choices involve possible losses. ment and coalignment issues are important in
Further, to simplify choices between alternatives, the context of innovation, but they are quite
individuals generally evaluate options in isolation. different from portfolio balance issues faced by
Viewing each alternative as unique leads decision- financial investors. The presence of increasing
makers to undervalue possibilities for risk pooling. returns means that one also needs to strategize
This approach to decision making may produce around investment decisions, getting the timing
inconsistent preferences and decision biases (timid right, building on increasing return advantages, and
choices) that lead to outcomes that block inno- leveraging products and services from one appli-
vation (Kahneman and Tversky, 1979; Kahneman cation to another. Value-enhancing investments
and Lovallo, 1993). An opposing bias to loss/risk inside the knowledge-based enterprise are often
aversion is excessive optimism. This leads to cospecialized12 to each other. Also, the nature of
investment in low or negative return projects. As a the portfolio balance needed inside the enterprise
result, entry decisions often fail. Audretsch (1995) is different from the portfolio balance sought by
found that over the period 197686 the average pure financial investors. The economics of cospe-
10-year failure rate in two-digit SIC manufacturing cialization are not the economics of covariance
sectors ranged from 75.8 percent to 54.8 percent. with which investors are familiar. In short, the
Similar failure rates have been reported in other task of making astute project- and enterprise-level
studies (Dunne, Roberts, and Samuelson, 1988; investment decisions is quite challenging because
Klepper and Miller, 1995). However, these failure of cospecialization, and irreversibilities.
rates disguise wide variation amongst particular The project finance and related literatures pro-
enterprises and between new entrants and incum- vide tools and clear decision rules for project
bents. selection once cash flows are specified, uncertainty
The existence of established assets and routines and/or risk are calibrated, and interdependencies
exacerbates problems of excessive risk aversion.
Specifically, both the isolation effect and the cer- 10
Consider the development of civilian jet transport aircraft in
tainty effect can be intensified by the existence of the United States in the 1950s. As Phillips (1971: 126) noted:
established assets, causing incumbent enterprises Any one of Boeing, Douglas, Lockheed, or Corvair might
have been first. . . . The technology was there to adapt tonot
to become comparably more risk averse than new risklessly or costlessly to be sure, but it was there. Perhaps the
entrants. In terms of innovative activity, this exces- biggest risk in 1953 was not technological in character. Instead,
sive risk aversion leads to biased decision making it was risk with respect to what sort of jet to build and when to
build it.
and limits the probability that incumbent enter- 11
The decision skills required of management have limited
prises will explore risky radical innovations. In commonality with those of an investor. One difference is the
short, success in one period leads to the establish- illiquidity and irreversibility of most managerial investment
decisions. Another is the need to achieve continuous alignment
ment of valid processes, procedures, and incen- amongst the assets at work in the enterprise. Both public and
tives to manage the existing business. This can private equity investors typically lack this kind of orchestration
have the unintended effect of handicapping the and integration capability or capacity. Moreover, their skills are
most applicable when investments are liquid.
new business. The proficiency with which such 12
Cospecialization is defined and discussed in Teece (1986) and
biases are overcome and a new opportunity is explored further in the section below entitled Managing threats
embraced is likely to depend importantly on the and reconfiguration.

Copyright 2007 John Wiley & Sons, Ltd. Strat. Mgmt. J., 28: 13191350 (2007)
DOI: 10.1002/smj
Explicating Dynamic Capabilities: Nature and Microfoundations 1329

between and amongst cash flows are ignored. apparent. No matter how much analytical work is
However, the essence of the investment decision done, tacit investment skills are of great impor-
for the (strategic) manager is that it involves esti- tance. Chandler further argues that success in the
mating interdependent future revenue streams and late-nineteenth and much of the twentieth century
cost trajectories, and understanding a panoply of came to those enterprises that pursued his three-
continuous and interrelated cospecialized invest- pronged strategy: (1) early and large-scale invest-
ment issues.13 The returns to particular cospecial- ments behind new technologies; (2) investment in
ized assets cannot generally be neatly apportioned product-specific marketing, distribution, and pur-
or partitioned. As a result, the utility of traditional chasing networks; and (3) recruiting and organiz-
investment criteria is impaired. Thus while project- ing the managers needed to supervise and coordi-
financing criteria (e.g., discounted cash flow, pay- nate functional activities. The first and second ele-
back periods and the like) and techniques for deci- ments require commitment to investments where
sion making under uncertainty are well known, irreversibilities and cospecialization are identified.
there is little recognition of how to value intangi- While the nature of required investments may
bles and take into account features such as cospe- have changed in recent decades (less decompos-
cialization, irreversibility, and opportunity costs.14 able/more interrelated), investment decision skills
Nor is the concept of a strategic investment rec- remain important.
ognized in the finance literature. Finance theory
provides almost no guidance with respect to how to Microfoundations
estimate future cash flows, although making such
estimates is as much, if not more, the essence of Selecting product architectures and business
good decision making as are the methodologies models
and procedures for analyzing cash flow. The design and performance specification of prod-
In short, managers need to make unbiased judg- ucts, and the business model employed, all help
ments under uncertainty around not just future define the manner by which the enterprise deliv-
demand and competitive responses associated with ers value to customers, entices customers to pay
multiple growth trajectories, but also around the for value, and converts those payments to profit.
pay-offs from making interrelated investments in They reflect managements hypothesis about what
intangible assets. In the world of tangible assets, customers want and how an enterprise can best
this can sometimes be precisely modeled; not meet those needs, and get paid for doing so. They
so for the world of cospecialized intangibles. embrace: (1) which technologies and features are
In essence, the organizational challenge appears to be embedded in the product and service; (2) how
to be that in environments experiencing rapid the revenue and cost structure of a business is to be
change, activities are not fully decomposable. designed and if necessary redesigned to meet
Cross-functional activities and associated invest- customer needs; (3) the way in which technolo-
ments must take place concurrently, rather than gies are to be assembled; (4) the identity of market
sequentially, if enterprises are to cut time-to- segments to be targeted; and (5) the mechanisms
market for new products and processes. Man- and manner by which value is to be captured.
agerial judgments (decision-making skills) take The function of a business model is to articu-
on great significance in such contexts. This was late the value proposition, select the appropriate
also true during prior centuries, as Alfred Chan- technologies and features, identify targeted market
dlers (1990a, 1990b) analysis of successful enter- segments, define the structure of the value chain,
prises from the 1870s through the 1960s makes and estimate the cost structure and profit potential
(Chesbrough and Rosenbloom, 2002: 533534). In
Monteverde and Teeces (1982) study of the automobile indus- short, a business model is a plan for the organi-
try showed that systems integration considerations impacted zational and financial architecture of a business.
makebuy decisions. This evidence hints at the value to be
created from figuring out heuristics and protocols likely to aid
This model makes assumptions about the behav-
decisions involving interrelated investments. Evans, Hagiu, and ior of revenues and costs, and likely customer and
Schmalensee (2006) recognize multisided market interdependen- competitor behavior. It outlines the contours of the
cies which likewise require a systems perspective.
solution required to earn a profit, if a profit is avail-
Ghemawat (1991) and many others have examined uncertainty
and irreversibilities. However, cospecialization has received very able to be earned. Once adopted it defines the way
little attention. the enterprise goes to market. Success requires
Copyright 2007 John Wiley & Sons, Ltd. Strat. Mgmt. J., 28: 13191350 (2007)
DOI: 10.1002/smj
1330 D. J. Teece

that business models be astutely crafted. Other- other dot-coms demonstrated just the opposite).
wise, technological innovation will not result in Both Dell Inc.s and Wal-Marts business models
commercial success for the innovating enterprise. were different, superior, and hard for competitors
Generally there is a plethora of business models to replicate. They have also constantly adjusted
that can be designed and employed, but some will and improved their processes over time.16
be better adapted to the ecosystem than others. One might be tempted to argue that designing,
Selecting, adjusting, and/or improving the business implementing, and validating business models is
model is a complex art. straightforward, but this simply is not so. Aspects
Nevertheless, the importance of business mod- of designing (and redesigning) a business model
els has been given short shrift in the academic are undoubtedly readily routinized and codified.
literature, at least until quite recently. Important Note the plethora of business books providing
(business model) choices include technological instruction on how to write a business plan. Such
choices, market segments to be targeted, financial manuals can provide some discipline to the busi-
terms (e.g., sales vs. leasing), choices with respect ness model questions that one should ask. How-
to bundled vs. unbundled sales strategies, joint ever, designing a new business requires creativity,
ventures vs. licensing vs. go-it-alone approaches, insight, and a good deal of customer, competitor,
etc. For example, in the early days of the copier and supplier information and intelligence. There is
industry, Xerox focused on leasing rather than a significant tacit component. Entrepreneurs and
selling copiers. This stemmed from a belief that executives are forced to make many informed
customer trial would lead to further use. Another guesses about customer and competitor behavior,
example from the United States is Southwest Air- as well as the behavior of costs. Indeed, validat-
lines, which believes that most customers want ing a business model and a business plan requires
low frills, reliability, and low cost. It eschews both effort and judgment. It takes detailed fact-
the hub-and-spoke model, does not belong to any specific inquiry including: a keen understanding of
alliances, and does not allow interlining of passen- customer needs and customer willingness to pay;
gers and baggage. Nor does it sell tickets through an understanding of procurement cycles and the
travel agenciesall sales are direct. All aircraft sales cycle; knowledge of supply and distribution
are Boeing 737s. Its business model is quite dis- costs; and an understanding of competitor position-
tinct from the major carriers, although many have ing and likely competitive responses. Put differ-
tried (without much success) to copy elements of ently, selecting the right architecture for a busi-
the Southwest model.15 ness requires not just understanding the choices
The capacity an enterprise has to create, adjust, available; it also requires assembling the evidence
hone, and, if necessary, replace business models needed to validate conjectures and hunches about
is foundational to dynamic capabilities. Choices costs, customers, competitors, complementors, dis-
around how to capture value all help determine tributors, and suppliers.
the architecture or design of a business. Having Designing good business models is in part
a differentiated (and hard-to-imitate) yet effective art. However, the chances of success are greater
and efficient strategic architecture to an enter- if enterprises (1) analyze multiple alternatives,
prises business model is important. Both Dell Inc. (2) have a deep understanding of user needs,
and Wal-Mart have demonstrated the value associ- (3) analyze the value chain thoroughly so as to
ated with their business models (Webvan and many understand just how to deliver what the customer
wants in a cost-effective and timely fashion, and
Let us take another example. A rock star might decide to use (4) adopt a neutrality or relative efficiency per-
concerts as the key revenue generator, or the concert may be spective to outsourcing decisions. Useful tools
used primarily to stimulate sales of recordings. The star could include market research and transaction cost eco-
decide to spend less time performing at concerts, and more time
in the recording studio. There is clearly a choice of various nomics. Chesbrough and Rosenbloom (2002) sug-
media to extract value: live productions, movies, sale of CDs gest that established enterprises often have blinders
through stores, online sale of music through virtual stores such
as the iTunes store offered by Apple, etc. The emergence of
the Internet, Napster, and Napster clones in turn requires artists Indeed, a critical element of Dells success is not just the
(and record companies) to rethink their business models. The way it has organized the value chain, but also the products that
ability to reconfigure business models for delivering and pricing it decides to sell through its distribution system. The initial
music profitably is undoubtedly a dynamic capability for both products were personal computers, but now include printers,
the record companies and the artists. digital projectors, and computer-related electronics.

Copyright 2007 John Wiley & Sons, Ltd. Strat. Mgmt. J., 28: 13191350 (2007)
DOI: 10.1002/smj
Explicating Dynamic Capabilities: Nature and Microfoundations 1331

with respect to alternative business modelsand stream, downstream, as well as externally, is

that this prevails even if the technology is spun partly driven by the need to build capabilities,
off into a separate organization, where other (path- particularly when such capabilities are not widely
dependent constraints) are less likely to exist. distributed in the industry. Of course, vertical spe-
In short, designing the business correctly, and cialization is not itself independent of enterprise
figuring out what John Seeley Brown refers to strategy, and vice versa (Macher and Mowery,
as the architecture of the revenues17 (and costs), 2004). Studies of the early vertical evolution of
involve processes critical to the formation and suc- the petroleum industry stressed the need to align
cess of new and existing businesses. No amount of upstream and downstream capacities in an envi-
good governance and leadership is likely to lead ronment where qualified business partners were
to success if the wrong business model is in place. limited (Teece, 1976). Pisano, Shan, and Teece
Good business models achieve advantageous cost (1988: 202) developed a framework for under-
structures and generate value propositions accept- standing R&D outsourcing that recognized that the
able to customers. They will enable innovators to locus of world-class research/productive capabil-
capture a large enough portion of the (social) value ity might lie external to the enterprise, requiring
generated by innovation18 to permit the enterprise outsourcing as a way to compete.19 Jacobides and
at least to earn its cost of capital. Winter (2005: 398) have also clearly stated that it
is necessary to look at the distribution of produc-
Selecting enterprise boundaries tive capabilitiesto understand when enterprises
are integrated and when they are not. It becomes
In regimes of rapid technological progress, setting clear that vertical specialization must be in part a
the enterprise boundaries correctly is important, function of heterogeneity in productive capabili-
and can be viewed as an element of getting the ties along the value chain. They also note that the
business model right. In Teece (1986), Chesbrough capability development process itself changes as a
and Teece (1996), and Teece (1986, 2007) norma- consequence of changing scope. Recognition that
tive rules were advanced indicating how enterprise systemic innovation favors integration, for both
boundaries ought to be set to ensure that innovation transaction costs and capability reasons, is also
is more likely to benefit the sponsor of the inno- embedded in the saga of the development of the
vation rather than imitators and emulators. Key diesel electric locomotive (Teece, 1988). The abil-
elements of this framework were: (1) the appropri- ity of enterprises to procure technology externally
ability regime (i.e., the amount of natural and legal as well as develop it internally are critical skills,
protection afforded the innovation by the circum- as discussed above and in Teece (1986), Ches-
stances prevailing in the market); (2) the nature brough and Teece (1996), and Teece (2000). Firms
of the complementary assets (cospecialized or oth- must dispel prejudices against technology from
erwise) that an innovating enterprise possessed; the outside, and hone their absorptive capacity
(3) the relative positioning of innovator and poten- through learning activities and skill accumulation.
tial imitators with respect to complementary assets; Enterprises may require alliance arrangements to
and (4) the phase of industry development (pre or actively learn and upgrade relevant skills (Branzei
post the emergence of a dominate design). The and Vertinsky, 2006).
framework is prescriptive not only as to strategy The critical strategic element associated with
but also as to outcomes. capturing value from innovation is the ability of
Enterprise boundary decisions need to reflect the innovating enterprise to identify and control the
other criteria too. A companys integration up- bottleneck assets or choke points in the value
chain from invention through to market (Teece,
Quoted in Chesbrough and Rosenbloom (2002: 529). 1986, 2006). Outsourcing those assets/services that
A recent effort to establish a new business model is exem-
plified by the efforts of Rambus to rely exclusively on patent
licensing to capture value from its significant technological con- The model identified transaction costs, the locus of capabilities
tributions to the design of semiconductor memory devices. Such (inside or outside the enterprise), and appropriability regimes
an approach avoids building fabrication facilities (which are as three relevant classes of factors driving enterprise boundary
extremely expensive) but its viability depends entirely on Ram- decisions. In particular, it was noted that transaction cost factors
buss ability to enforce its patents in an environment in which must be weighed against any losses in productive efficiency
courts are sometimes reluctant to enjoin infringers and where that result from being less skilled than specialists in the relevant
enforcing broad patents may engender antitrust challenges. stages of production.

Copyright 2007 John Wiley & Sons, Ltd. Strat. Mgmt. J., 28: 13191350 (2007)
DOI: 10.1002/smj
1332 D. J. Teece

are in competitive supply is, of course, consis- products are often systems. These systems con-
tent with such a strategy. In short, the boundaries sist of interdependent components resting on plat-
of the enterprise need to be artfully contoured forms. There is strong functional interdependence
for each major innovation, using decision crite- amongst components of the system. End user
ria referenced above. Failure to do so is likely to demand is for the system, not the platform. There
be associated with the failure to stimulate market is often a multisided market phenomenon at work
development (especially of complementary tech- as well. For instance, electronic game consoles are
nologies) and incomplete capture of the profits not much use without games; computer operating
available from innovation. systems are not much use without a suite of appli-
cation programs; credit cards are not much use
to cardholders without merchants that will accept
Managing complements and platforms
them, and vice versa; and hydrogen cars are not
Investment choices in many high-technology much use without hydrogen filling stations, and
industries today are driven by imperatives quite vice versa. This important class of situations has
different from the (industrial) contexts that have highlighted the importance of cospecialization, and
animated strategy research over the past half cen- strategic decision making must now take this into
tury. Scale and scope economy mandates, which account.
to some strategists dictate the scale and scope of The phenomenon is not newthe automobile
the enterprise, have given way to a different set industry depended first on the general store and
of mandates around developing (or encouraging) then specialized retail outlets to make gasoline
complementary investments and capturing cospe- ubiquitously available to motorists. The role of
cialization benefits. The reason for this is that in complementary assets and cospecialization has
many industries outsourcing has made scale an already been recognized in the innovation process,
industry asset, in the sense that economies of scale and a decision framework outlined to chart the
can be captured by outsourcing to contract manu- innovator on a course more likely to lead to a
facturers who, in the face of competition, pass on higher share of the available profit (Teece, 1986,
the benefits of scale. Witness the contract semicon- 2007). What is new is that complements often sit
ductor fabricators. They enable fabless semicon- on top of what might be thought of as platforms,
ductor designers to capture most of the benefits which are managed by an incumbent enterprise
of scale without engaging in manufacturing. Like- (Evans et al., 2006). In these circumstances, entry
wise, in the clothing industry, small-scale design- decision and boundary conundrums exist. The
ers of footwear and outerwear can source at com- platform owner needs complementary products to
petitive rates from large suppliers, thereby cap- be provided by others, particularly when it has
turing the benefit of scale economics previously little or no relevant skills to develop them itself.
enjoyed only by large integrated manufacturers. Fostering innovation and entry by the providers of
With competition, scale advantages are not propri- complementary products may, in fact, require the
etary, and are unlikely to be a source of sustainable platform manager to commit (by word or deed) not
differentiation. to provide certain complements. When the inter-
When intermediate (product) markets are well face between the complementors and the platform
developed, neither economies of scale nor is itself evolving, decision rules become ever more
economies of scope need define the scale and scope complex. The platform owner and the complemen-
of the enterprise. Contractual access (on compet- tors might also need to consider whether the plat-
itive terms) to scale-based facilities vitiates the form needs to be open or proprietary, and whether
need for enterprise scale and scope. This was the tools and other incentives should be provided to
major theme in Teece (1980) but the importance stimulate investment by the complementors. Deci-
of the argument was often not appreciated. Today sion frameworks that recognize the importance
its importance is more evident. of network effects, dispersion in the sources of
While the importance of scale and scope innovation of complementary products, interoper-
economies to enterprise boundary decisions may ability issues, and installed base trajectories must
have been softened, the significance to enter- all be factored into decisions. Quality decisions
prise strategy of cospecialization has been ele- will require uncommon foresight and the ability to
vated. As viewed by customers, high-technology shape outcomes. In this regard, the existing asset
Copyright 2007 John Wiley & Sons, Ltd. Strat. Mgmt. J., 28: 13191350 (2007)
DOI: 10.1002/smj
Explicating Dynamic Capabilities: Nature and Microfoundations 1333

base of the platform manager, including its finan- is also essential. Management also needs to create
cial resources, is of considerable significance. The an environment where the individuals involved in
distribution of (development) capabilities between making the decision, at both the management and
the platform manager and the complementors will board level, feel free to offer their honest opinions,
also be important. Also, as discussed below, the and look at objective (historical) data in order to
boundaries of the enterprise (i.e., whether the plat- escape from closed thinking. Incentives must also
form manager is also providing complements) is be designed to create neutrality when assessing
likely to be of significance, possibly deterring (or investments in the old and the new.
encouraging) entry and innovation by complemen- Considerable progress in combating biases has
tors. been made. Advisors call upon managers to adopt
radical, nonformulaic strategies in order to over-
come the inertias that inhibit breakthrough inno-
Avoiding bias, delusion, deception, and hubris20
vation (Davidow and Malone, 1992; Handy, 1990).
As noted, proclivities toward decision errors are Specifically, corrective strategies encourage
not uncommon in managerial decision making, change through two basic mechanisms: (1) design-
particularly in large organizations. Investment deci- ing organizational structures, incentives and rou-
sion errors already identified include excessive tines, to catalyze and reward creative action; and
optimism, loss aversion, isolation errors, strate- (2) developing routines to enable the continual
gic deception, and program persistence. As Nelson shedding of established assets and routines that no
and Winter (2002: 29) note, organizational deci- longer yield value. Strategies that provide struc-
sion processes often display features that seem to tures, incentives, and processes to catalyze and
defy basic principles of rationality and sometimes reward creative action serve to attenuate problems
border on the bizarre. These errors can be espe- of excessive risk aversion. For example, strategies
cially damaging in fast-paced environments with that call on the enterprise to cut overhead and
path dependencies and network effects, as there is increase divisional authority can be interpreted as
less opportunity to recover from mistakes. When efforts to reduce the number of management lay-
investments are small and made frequently, there ers of the enterprise and to push decision making
are many opportunities to learn from mistakes. down to lower levels to minimize the inherent iso-
Since large investments are usually occasional, lation errors associated with multilevel, hierarchi-
major investment decisions are likely to be (poten- cal decision-making processes. These recommen-
tially) more vulnerable to error. dations can be viewed as organizational processes
Fortunately, biases can be recognized ahead of and strategic mechanisms to mitigate decision-
time. Enterprises can bring discipline to bear to making biases.
purge bias, delusion, deception, and hubris. How- Perhaps most importantly, executives must
ever, the development of disciplines to do so is still acknowledge the interaction effect between own-
in its infancy. The implementation of procedures ing established assets and decision-making biases.
to overcome decision-making biases in enterprise Many recommended strategies (such as cannibal-
settings is, accordingly, not yet a well-distributed izing profitable product lines and licensing your
skill, and may not be for decades to come. Accord- most advanced technology) call for the shed-
ingly, competitive advantage can be gained by ding of established capabilities, complementary
early adopters of techniques to overcome decision assets and/or administrative routines to reduce the
biases and errors. intensity of decision-making biases. By jettisoning
Overcoming biases almost always requires a dead or dying assets, the enterprise is no longer
cognitively sophisticated and disciplined approach shackled with an asset base that can be a crutch
to decision making. Being alert to the incentives and provide a false sense of security, and sustain
of the decision-makers and to possible informa- groups inside the enterprise that persist in torpedo-
tion asymmetries is a case in point. Obtaining an ing new initiatives. In abandoning dead or dying
outside view through the review of external data assets, the enterprise frees itself of certain routines,
can help eliminate bias. Testing for errors in logic constraints, and opportunities for undesirable pro-
tective action inside the enterprise.
I would like to thank Dan Lovallo for inspiration and help in Sources of the anti-cannibalization bias men-
this section. tioned earlier can also be attacked. Self-serving
Copyright 2007 John Wiley & Sons, Ltd. Strat. Mgmt. J., 28: 13191350 (2007)
DOI: 10.1002/smj
1334 D. J. Teece

behavior inside the enterprise to protect incum- and other constituencies. Organizational identifi-
bent constituencies undergirds this bias. Flawed cation (and commitment, which is the corollary)
investment frameworks may also contribute. Entry can dramatically augment enterprise performance,
into a market by an enterprise with a new and although it is doubtful it can override completely
superior technology will cause rapid deprecia- misaligned incentives. Nevertheless, group loy-
tion of the economic value of an incumbents alty is a powerful altruistic force that conditions
plant and equipment. However, the incumbent may employee goals and the cognitive models they
well make business decisions based on exam- form of their situation (Simon, 1993: 160). Top
ining accounting profits that reflect depreciation management through its action and its commu-
rates specified by accepted accounting standards. nication has a critical role to play in garnering
If decision-makers confuse depreciation calculated loyalty and commitment and achieving adherence
according to generally accepted accounting prin- to innovation and efficiency as important goals.
ciples (GAAP) with real economic depreciation, Since there is already an extensive literature on
and conclude that the existing business is still culture, commitment, and leadership, these issues
profitable when, in fact, it is not, then the busi- are not discussed further. However, it would be
ness enterprise may eschew profit-enhancing can- a significant oversight in a summary statement of
nibalization of its own products. To guard against the dynamic capabilities framework to ignore them
this bias, investment decision-makers and incum- completely. Their full integration into the frame-
bents must use accounting data cautiously. In par- work is left to others. However, it is recognized
ticular, they must also consider the opportunity that to the extent such properties are not ubiq-
cost associated with not cannibalizing their own uitously distributed amongst business enterprises,
products. Capital-budgeting procedures implicitly they can be a very important source of superior
biased against projects with long-term horizons performance. Figure 2 summarizes the microfoun-
must be jettisoned or used cautiously. That is not dations identified in this section of the paper.
to say that incumbents need to invest on the same
schedule as new entrants. As Teece (1986) and
Mitchell (1991) demonstrate, incumbents need not MANAGING THREATS AND
be the first movers. Superior positioning in com- RECONFIGURATION
plementary assets may enable incumbents to let
the new entrants do the prospecting, investing later
once market and technological risk has diminished. The successful identification and calibration of
There is an obvious role for leadership in making technological and market opportunities, the judi-
quality decisions, communicating goals, values, cious selection of technologies and product attri-
and expectations, while also motivating employees butes, the design of business models, and the

Delineating the Customer Solution and Selecting Enterprise Boundaries to

the Business Model Manage Complements and Control
Selecting the Technology and Product
Calibrating Asset Specificity;
Controlling Bottleneck Assets;
Designing Revenue Architectures; Assessing Appropriability;
Selecting Target Customers; Recognizing, Managing, and
Designing Mechanisms to Capture Capturing Cospecialization
Value. Economies.
Enterprise Structures,
Procedures, Designs and
Incentives for Seizing
Selecting Decision-Making Protocols Opportunities
Building Loyalty and Commitment

Recognizing Inflexion Points and Demonstrating Leadership;

Complementarities; Effectively Communicating;
Avoiding Decision Errors and Recognizing Non-Economic Factors,
Anticannibalization Proclivities. Values, and Culture.

Figure 2. Strategic decision skills/execution

Copyright 2007 John Wiley & Sons, Ltd. Strat. Mgmt. J., 28: 13191350 (2007)
DOI: 10.1002/smj
Explicating Dynamic Capabilities: Nature and Microfoundations 1335

commitment of (financial) resources to investment As the enterprise grows, it has more assets to
opportunities can lead to enterprise growth and manage and to protect against malfeasance and
profitability. Profitable growth will lead to the aug- mismanagement. Shirking, free riding, the strategic
mentation of enterprise-level resources and assets. manipulation of information, and internal compla-
Success will cause the enterprise to evolve in a cency are all issues that established enterprises
path-dependent way. A key to sustained profitable will confront continuously. As discussed earlier,
growth is the ability to recombine and to recon- over time successful enterprises will develop hier-
figure assets and organizational structures as the archies and rules and procedures (routines) that
enterprise grows, and as markets and technolo- begin to constrain certain interactions and behav-
gies change, as they surely will. Reconfiguration iors unnecessarily. Except in very stable envi-
is needed to maintain evolutionary fitness and, if ronments, such rules and procedures are likely
necessary, to try and escape from unfavorable path to require constant revamping if superior perfor-
dependencies. In short, success will breed some mance is to be sustained. It is not uncommon to
level of routine, as this is necessary for operational find that a once functional routine becomes dys-
efficiency. Routines help sustain continuity until functional, providing inertia and other rigidities
there is a shift in the environment. Changing rou- that stand in the way of improved performance
tines is costly, so change will not be (and should (Leonard-Barton, 1995; Rumelt, 1995). As a result,
not be) embraced instantaneously. Departure from less well-resourced enterprises (sometimes estab-
routines will lead to heightened anxiety within the lished enterprises that have divested certain assets,
organization, unless the culture is shaped to accept sometimes new entrants) end up winning in the
high levels of internal change. If innovation is marketplace.
incremental, routines and structures can probably Traditional management approaches endorse
be adapted gradually or in (semi-continuous) steps. strong hierarchies with at least three levels of
management: top, middle, and lower. Control is
When it is radical, possibly because it is science
exerted at the top and cascades down through mul-
based, then there will be a mandate to completely
tiple levels. Employees tend to end up beholden
revamp the organization and create an entirely new
to the management and CEO, and not the cus-
break out structure (Teece, 2000) within which an
tomer. The existence of independent profit centers
entirely different set of structures and procedures
can lead to internal boundaries that stand in the
is established.
way of providing integrated solutions that bene-
As discussed earlier, the anti-cannibalization fit customers. With centralized structures, strategic
bias is a particular manifestation of incentive decisions made at the top tend to become iso-
and structural problems that can thwart innova- lated from marketplace realities. Customer care is
tion in established enterprises. Incumbent enter- relegated to employees who are lower down in
prises possessing fixed assets may further tend to the organization. In short, the systems and rules
limit their new investments to innovations that needed to manage many layers of organization
are close-in to the existing asset base. They tend to create structural rigidities and perversities
tend to narrowly focus search activities to exploit that in turn handicap customer and technological
established technological and organizational assets. responsiveness. To sustain dynamic capabilities,
This effect makes it difficult for these enterprises decentralization must be favored because it brings
to see potential radical innovations. In addition, top management closer to new technologies, the
incumbent enterprises tend to frame new prob- customer, and the market.
lems in a manner consistent with the enterprises Top management leadership skills are required
current knowledge base, assets, and/or established to sustain dynamic capabilities. An important man-
problem-solving heuristics and established busi- agerial function is achieving semi-continuous asset
ness model. This second effect means that man- orchestration and corporate renewal, including the
agers may not successfully address opportunities or redesign of routines. This is because the sustained
potential innovations even when they do recognize achievement of superior profitability requires semi-
them. Managers face and must overcome at least continuous and/or continuous efforts to build,
two constraintscognitive limitations and fram- maintain, and adjust the complementarity of prod-
ing biasesarising from established assets (Teece, uct offerings, systems, routines, and structures.
2000). Inside the enterprise, the old and the new must
Copyright 2007 John Wiley & Sons, Ltd. Strat. Mgmt. J., 28: 13191350 (2007)
DOI: 10.1002/smj
1336 D. J. Teece

complement. If they do not, business units must to quasi-independent profit centers. The abandon-
be disposed of or placed in some type of sepa- ment of functional structures in favor of the mul-
rate structure. Otherwise, work will not proceed tidivisional form has been analyzed by Chan-
efficiently, and conflicts of one kind or another dler (1962), Williamson (1975), and many others.
will arise. Put differently, periodic if not continu- The basic rationale of this reconfiguration was
ous asset orchestrationinvolving achieving asset to achieve greater accountability of managerial
alignment, coalignment, realignment, and rede- decisions so that the recognition of opportunities
ploymentis necessary to minimize internal con- and threats could proceed more thoroughly and
flict and to maximize complementarities and pro- expeditiously. With functional internal structures,
ductive exchange inside the enterprise. day-to-day problems tend to distract management
Redeployment and reconfiguration (Capron, from long-run strategic issues. Studies showed
Dussauge, and Mitchell, 1998) may also involve that decentralization along product or market lines
business model redesign as well as asset-realign- with independent profit centers led to performance
ment activities, and the revamping of routines. improvements in many industries, at least during
Redeployment can involve transfer of nontrad- the period in which these organizational innova-
able assets to another organizational or geographic tions were diffusing (Armour and Teece, 1978;
location (Teece, 1977, 1980). It may or may not Teece, 1980, 1981). More recent scholarship has
involve mergers, acquisitions, and divestments.21 suggested that even further decentralization and
Helfat and Peteraf (2003: 1006) suggest that capa- decomposition in large organizations may be ben-
bility redeployment takes one of two forms: the eficial (Bartlett and Ghoshal, 1993). There is also
sharing of capability between the old and the new, some evidence that modern human resource man-
and the geographic transfer of capability from one agement techniquesinvolving delayering, decen-
market to another. Both are possible, but neither is tralization of decision rights, teamwork, flexi-
easy. ble task responsibilities and performance-based
rewardsalso improve performance (Jantunen,
Microfoundations Of course, achieving decentralization can com-
Achieving decentralization and near promise the organizations ability to achieve inte-
decomposability gration. There is little harm and much benefit from
decentralization when the customer does not ben-
Every system comprises subsystems (elements)
efit from an integrated product offering, or when
that are to some extent interdependent and inde-
sourcing and other inputs do not benefit from inte-
pendent. However, as discussed earlier, enterprises
gration and/or aggregation. If customer and sup-
are unlikely to be continuously responsive to cus-
ply considerations allow decomposability (because
tomers and new technologies absent a high degree
the required integration between units is less than
of decentralization. With decentralized decision
within units), then managements ability to iden-
making, different managers observe different infor-
tify and implement decomposable subunits should
mation and control different decisions, but there is
enhance performance. However, if firm-specific
not the need for communication to a single central
economies of scale and scope are available, they
decision-maker, and hence no comprehensive roll-
must be capturedotherwise the enterprise is tan-
up of information is required. Decentralization
tamount to a conglomerate. This tension can be
must be pursued as enterprises expand, otherwise
managed through a collaborative nonhierarchical
flexibility and responsiveness will erode.
management style assisted by establishing councils
One well-documented restructuring that is and other integration forums. Middle management
widely adopted as enterprises grow is the adoption can also play a critical role when such forums
of the multidivisional form. This involves decom- are established. They can also design and imple-
position and the devolution of decision rights ment tight financial controls and performance-
based reward systems. Since intangibles are key
As Capron et al. (1998) explain, failures in the market for drivers of performance, their enhancement and
resources sometimes cause enterprises to buy and sell business. protection must become a managerial priority.
What they refer to as market failure appears to relate to the
thin market problem discussed by the author in this paper and The open innovation model of Chesbrough
elsewhere (Helfat et al., 2007). (2003) also recognizes the benefits of relying on
Copyright 2007 John Wiley & Sons, Ltd. Strat. Mgmt. J., 28: 13191350 (2007)
DOI: 10.1002/smj
Explicating Dynamic Capabilities: Nature and Microfoundations 1337

a distributed model of innovation where the enter- to match an array of interlocked activities than it is
prise reaches out beyond its own boundaries to merely to imitate a particular sales force approach,
match a process technology, or replicate a set of
access and integrate technology developed by oth- product features. (Porter, 1996: 73)
ers. By way of example, Henderson and Cock-
burn (1994) found that an enterprises ability to Despite Porters clear recognition of the concept
integrate knowledge from external sourcestheir of fit, neither complementarities nor cospecializa-
architectural competencewas positively asso- tion are recognized in the Five Forces framework.
ciated with research productivity, as measured by However, complementarities and cospecialization
patent counts. Likewise, Iansiti and Clark (1994) are recognized in various ways in Teece (1986,
found that integration capability in the automo- 2000, 2006, 2007), Brandenburger and Nalebuff
bile industry and in the computer industry was (1996), and Santoro and McGill (2005). Economic
associated with positive enterprise performance, historians (Rosenberg, 1982; Hughes, 1983) have
again demonstrating the importance of knowledge also noted the phenomenon at a general level; but
integration skills. In the end, it appears that in in most analyses of competition and competitive
fast-paced environments organizational units must advantage, it is common to stress that various inno-
have considerable autonomy (to make decisions vations are substitutes, rather than complements
rapidly) but remain connected to activities that that may be cospecialized to each other. Indeed,
must be coordinated. Achieving this delicate bal- Schumpeter (1934) stressed that successful inno-
ance is what Simon (2002) called near decom- vations/enterprises are threatened by swarms of
posability and implementing it is an important imitators, all striving to produce me-too substi-
microfoundation of dynamic capabilities. tutes.22 He completely neglected complementari-
Managing cospecialization However, complementary innovation and com-
The field of strategic management and the dynamic plementary assets are of great significance, partic-
capabilities framework recognizes that strategic ularly in industries in which innovation might be
fit needs to be continuously achieved. However, characterized as cumulative, and/or where indus-
unless the concept is operationalized it has lim- try platforms exist or are needed. Examples of
ited utility. The key dimension of fit emphasized complementary innovation are ubiquitous. In the
in the dynamic capabilities framework is that of enterprise software industry, business applications
cospecialization. The concept of cospecialization, can be especially valuable to users if they can
introduced in Teece (1986) and discussed in the somehow be integrated into a single program, or
Managing complements and platforms section into a tightly integrated suite. The development
above, operationalizes at least one dimension of of gyroscopic stabilizers made imaging devices
the otherwise rather vague concept of organiza- such as video cameras and binoculars easier to
tional adaptation and fit. Cospecialization can be use by minimizing the impact of camera shake,
of one asset to another, or of strategy to struc- and enhanced the product, especially when the
ture, or of strategy to process. It is important to new feature was able to be introduced at low cost.
both seizing and reconfiguration. In environments Likewise, better high-energy rechargeable batteries
of rapid change, there is a need for continuous or enable laptop computers and cell phones to operate
at least semi-continuous realignment. for longer times. Situations of complementarities
In many ways, much of the traditional literature where there is also cospecialization between tech-
on organizational adaptation and fit (e.g., Miles nologies, and between technologies and other parts
and Snow, 1994) is consistent with dynamic capa-
bilities. In particular, both the strategy and organi- Schumpeter wrote (1934: 223) that innovations/new combi-
nations carried out by entrepreneurs are not, as one would
zational behavior literature emphasize fit between expect according to general principles of probability, evenly
and amongst strategy, structure, and processes. distributed through time . . . but appear, if at all, discontinu-
While it is not central to his framework, Michael ously in groups or swarms. This swarming of innovations
and innovative activity occurs exclusively because the appear-
Porter does note that: ance of one or a few entrepreneurs facilitates the appearance
of others, and these the appearance of more, in even increasing
[S]trategic fit among many activities is fundamental number (Schumpeter, 1934: 228). Recent studies that analyze
not only to competitive advantage but also to sus- patent races have also reinforced the view that innovations are
tainability of that advantage. It is harder for a rival substitutes, not complements.

Copyright 2007 John Wiley & Sons, Ltd. Strat. Mgmt. J., 28: 13191350 (2007)
DOI: 10.1002/smj
1338 D. J. Teece

of the value chain, are common, yet until recently at the GM labs. The earliest use was in sub-
poorly analyzed in economic analysis and in strat- marines. Alfred P. Sloan, GMs chairman, saw
egy formulation. the possibility of applying the technology to make
Cospecialized assets are a particular class of diesel-electric locomotivessteam power was, at
complementary assets where the value of an asset the time, completely dominant. To accomplish this,
is a function of its use in conjunction with other GM needed capabilities resident in the locomotive
particular assets.23 With cospecialization, joint use manufacturers and at Westinghouse Electric. Lan-
is value enhancing.24 Cospecialization results in glois (1992: 115) notes that the three sets of capa-
thin markets; i.e., the assets in question are bilities might have been combined by some kind
idiosyncratic and cannot be readily bought and of contract or joint venture, but the steam man-
sold in a market. Capturing cospecialization ben- ufacturersAlco, Baldwin, and Limafailed to
efits may require integrated operations (Teece, cooperate.26 In short, both innovation and reconfig-
1980). Cospecialization allows differentiated prod- uration may necessitate cospecialized assets being
uct offerings or unique cost savings. The inher- combined by management in order for (systemic)
ent thin market environment surrounding spe- innovation27 to proceed. Managers do not always
cific assets means that competitors are not able to succeed in doing so, sometimes because they do
rapidly assemble the same assets by acquisition, not sense the need or the opportunity, and some-
and hence cannot offer the same products/services times because they do but they are unable to effec-
at competing price points. tuate the integration. If the assets cannot be pro-
Managements ability to identify, develop, and cured externally, they will need to be built inter-
utilize in combination specialized and cospecial- nally.
ized assets built or bought is an important dynamic The ability of management to identify needs
capability, but it is not always present in enterprise and opportunities to invest in cospecialized
settings. Special value can be created (and poten- assets (through its own development or astute
tially appropriated by another party) through asset purchase) is fundamental to dynamic capabilities.
combinations, particularly when an asset owner Mere horse-trading skills (which market agents
is not cognizant of the value of its assets to possess) will not suffice to build sustainable
another party that owns assets whose value will competitive advantage, and decisions on when and
be enhanced through combination.25 This arises how to investwhether and when to build or
because the markets for cospecialized assets are buy cospecialized assetswill depend upon many
necessarily thin or nonexistent. Langlois (1992) factors, including transaction costs. In particular,
highlights the case of the diesel-electric locomo- it will depend on managements entrepreneurial
tive where, in the 1920s, Charles Kettering had capacities with respect to matching up and
developed advanced lightweight diesel technology integrating relevant cospecialized assets.
It is apparent that cospecialization involves
Lippman and Rumelts (2003a, 2003b) recent work on devel- lock-in and is a particular form of complemen-
oping the microfoundations for resource-based theory is very tarity that exists when technologies and other
complementary to my development of the microfoundations of assets need to be part of a tightly integrated sys-
dynamic capabilities. I acknowledge their efforts in modeling
cospecialized and complementary assets. In particular, they use tem to achieve the performance that customers
the concept of supermodularity to bring in the tools of cooper- want. Business success in such circumstances
ative game theory. The idea of supermodularity was introduced requires the coordination of R&D investment and
by Donald Topkins as a way to formalize complementarity, and
is also used by economists such as Milgrom and Roberts (see, in alliance activity. The manner and timing with
particular, Milgrom and Roberts, 1990) and evolutionary game which such coordination needs to be accomplished
theorists to model (strategic) complementarities (for instance, in is important to success (Teece, 1986; Mitchell,
models of R&D spillovers).
Complete cospecialization is a special case of economies of
scope where not only are complementary assets more valuable
in joint use than in separate use, but they may, in fact, have zero This was not because the companies feared holdup in the face
value in separate use and high value in joint use. Cospecialization of highly specific assets. Rather, it was because they actively
may stem from economies of scope, but they could also stem denied the desirability of the diesel and fought its introduction
from the revenue enhancement associated with producing a at every step. GM was forced to create its own capabilities in
bundled or integrated solution for the customer. locomotive manufacturing.
25 27
Even if they are cognizant, they do not have the bargaining For a discussion of systemic innovation, see Teece (1988,
power to take advantage of the situation. 2000).

Copyright 2007 John Wiley & Sons, Ltd. Strat. Mgmt. J., 28: 13191350 (2007)
DOI: 10.1002/smj
Explicating Dynamic Capabilities: Nature and Microfoundations 1339

1991). Common ownership of the parts facili- much know-how does leak out. Innovating busi-
tates system-wide innovation and economic per- ness enterprises with limited experience have been
formance (Teece, 2000) and protects against oppor- known to inadvertently compromise or lose their
tunism (Williamson, 1975). intellectual property rights. Failure to proactively
To summarize, entrepreneurs and managers can monitor and protect know-how and intellectual
create special value by combining cospecialized property is common.
assets inside the enterprise (Teece, 2007). This may The outsourcing of production and the prolif-
require investments to create the necessary cospe- eration of joint development activities likewise
cialized technologiesas illustrated by Thomas create requirements that enterprises develop gov-
Edison and the creation of electric power as ernance procedures to monitor the transfer of tech-
a system. It is not uncommon in technology- nology and intellectual property. Technology trans-
based industries to find that certain technologies fer activities, which hitherto took place inside the
are worth more to some market participants than enterprise, increasingly take place across enterprise
to others, based on the technology they already boundaries. The development of governance mech-
have, and their technology and product strat- anisms to assist the flow of technology while pro-
egy. tecting intellectual property rights from misappro-
priation and misuse are foundational to dynamic
Learning, knowledge management, and corporate capabilities in many sectors today. Figure 3 sum-
governance marizes the microfoundations of this third class of
dynamic capability.
With intangible assets being critical to enterprise There are also several other governance issues
success, the governance and incentive structures relevant to dynamic capabilities. At one level there
designed to enable learning and the generation of are governance and business model issues associ-
new knowledge become salient. There are many ated with an enterprises ability to achieve asset
types of learningincluding experiential, vicari- combinations and reconfiguration. As noted ear-
ous, individual, and organizationaland a large lier, there is a continuous need to modify product
literature that explores each type. Also sensing offerings, business models, enterprise boundaries,
requires learning about the environment and about and organizational structures. Decentralized struc-
new technological capabilities. R&D was seen as tures that facilitate near decomposability are likely
one way that the enterprise could promote such to assist in achieving reconfiguration.
learning. However, in the context of the dynamic One class of governance issues relate to incen-
capability discussed in this section, the ability to tive alignment. The microfoundations of incentive
integrate and combine assets including knowledge issues are embedded in an understanding of agency
is a core skill (Kogut and Zander, 1992; Grant, and incentive design issues, also discussed earlier.
1996). The combination of know-how within the Agency theory has long emphasized that the sep-
enterprise, and between the enterprise and organi- aration of ownership from control creates interest
zations external to it (e.g., other enterprises, uni- alignment problems, particularly around manage-
versities), is important. ment compensation and the allocation of corpo-
Integrating know-how from outside as well rate perquisites. The abuse of discretion and the
as within the enterprise is especially important use of corporate assets for private purposes can
to success when systems and networks are occur absent appropriate accountability/oversight.
present. Good incentive design and the creation These issues become more severe as an enter-
of learning, knowledge-sharing, and knowledge- prise grows and the separation between ownership
integrating procedures are likely to be critical to and management widens. Recent corporate gover-
business performance, and a key (micro)foundation nance scandals in the United States, Europe, and
of dynamic capabilities (Nonaka and Takeuchi, Japan indicate the need for continued vigilance.
1995; Chesbrough, 2003). Of equal importance However, increasing the mix of independent and
are monitoring and managing the leakage, mis- inside directors will not necessarily ameliorate
appropriation, and misuse of know-how, trade problems associated with strategic malfeasance.
secrets, and other intellectual property. Of course, There are likely to be benefits associated with
tacit know-how is difficult to imitate and has a participation at the board level by individuals who
certain amount of natural protection. However, can calibrate whether the top management team
Copyright 2007 John Wiley & Sons, Ltd. Strat. Mgmt. J., 28: 13191350 (2007)
DOI: 10.1002/smj
1340 D. J. Teece

Decentralization and Near

Decomposability Cospecialization

Adopting Loosely Coupled Structures; Managing Strategic Fit So That

Embracing Open Innovation; Asset Combinations Are Value
Developing Integration and Enhancing.
Coordination Skills.

Continuous Alignment and

Realignment of Specific Tangible
and Intangible Assets.

Governance Knowledge Management

Achieving Incentive Alignment;
Knowledge Transfer;
Minimizing Agency Issues;
Know-how Integration;
Checking Strategic Malfeasance;
Achieving Know-how and Intellectual
Blocking Rent Dissipation.
Property Protection.

Figure 3. Combination, reconfiguration, and asset protection skills

is sufficiently dynamic. The replacement of the rewrite uncompetitive supply contracts that are the
CEO and other members of the top management product of unrealistic collective bargaining actions
team, if they demonstrate weak sensing, seizing, in an earlier period. The ability of some enterprises
and reconfiguration capabilities (strategic malfea- to craft work specifications, attract and retain more
sance), is important to effectuate. That is not to committed talent, design reward systems, develop
say that guarding against financial malfeasance is corporate cultures, and blunt the formation of
unimportant. It will always remain as an important coalitions that extract quasi-rents through threat-
corporate governance function; but its significance ening to withhold participation, is an important
is likely to pale next to strategic malfeasance, managerial capacity.
which is harder to detect and evaluate. The cur- The design and creation of mechanisms inside
rent wave of governance reforms in the United the enterprise to prevent the dissipation of rents by
Stateswith its strong emphasis on accounting interest groups (both management and employees)
controls and systems integritymay inadvertently would also appear to be very relevant to dynamic
lead to much bigger strategic performance fail- capabilities, but has not been high on the agenda of
ures by management. Boards stacked with inex- strategy researchers. One exception is Gottschalg
perienced independent board members may not and Zollo (2007), who point out that the capac-
have the requisite talents to properly diagnose ity to continuously achieve incentive alignment
strategic malfeasance and respond accordingly. is an important performance-enhancing (and rent-
A related literature in economics has stressed protecting) dynamic capability.
how poorly designed incentives can produce ten- Many of the issues discussed here have, in the
sions between the actions of employees and the past, fallen under the rubric of human resource
actions needed to achieve profitable performance. management; a closer connection of these issues
Dysfunctional behavior, such as activity that gen- to strategic management issues would appear to
erates influence costs, has received considerable be warranted. The reason is that strategic man-
attention (Teece, 2003). Also, through use of col- agement is focused not only on how to gen-
lective bargaining, employees in industries insu- erate rent streams, but also on how to prevent
lated from global competition have been able them from being dissipated or captured by various
to appropriate economic surplus. Above-market entities or groups inside and outside the enter-
wageswhich characterized, and to some extent prise. For instance, the concepts of the appropri-
still characterize, certain enterprises in the auto, ability regime and isolating mechanisms were
steel, and airline industries in the United States developed by strategic management scholars to
are a case in point. These conditions can extend to help explain how rents from innovation and other
managerial ranks as well. Restructuring may then sources of superior performance can be protected
require the judicious use of bankruptcy laws to and guarded from dissipation by competitors and
Copyright 2007 John Wiley & Sons, Ltd. Strat. Mgmt. J., 28: 13191350 (2007)
DOI: 10.1002/smj
Explicating Dynamic Capabilities: Nature and Microfoundations 1341

others. However, the earlier focus on markets to increase value from the use of the assets the
or external competition did not address internal enterprise owns involves knowing the fine-grained
appropriation by interest groups. structure of the firms asset base, and filling in
the gaps necessary to provide superior customer
solutions. Gap filling may involve building new
DYNAMIC CAPABILITIES, assets, or acquisitions and strategic partnerships
ORCHESTRATION SKILLS, AND (Ettlie and Pavlou, 2006).
COMPETITIVE ADVANTAGE The dynamic capabilities framework recognizes
that the business enterprise is shaped but not
The general framework advanced here sees necessarily trapped by its past. Management can
dynamic capabilities as the foundation of enter- make big differences through investment choice
prise-level competitive advantage in regimes of and other decisions. Enterprises can even shape
rapid (technological) change. The framework indi- their ecosystem. In this sense, the framework
cates that the extent to which an enterprise devel- is quite Chandlerian (Chandler, 1990a, 1990b).
ops and employs superior (nonimitable) dynamic Managers really do have the potential to set
capabilities will determine the nature and amount technological and market trajectories, particularly
of intangible assets it will create and/or assemble early on in the development of a market (David,
and the level of economic profits it can earn (see 1992). Indeed, the enterprise and its environ-
Figure 4). Furthermore, the framework emphasizes ment frequently coevolve. However, because of the
that the past will impact current and future perfor- assumed contextregimes of rapid technological
mance. However, there is much that management change exposed to the full force of international
can do to simultaneously design processes and competitionthere is little room for big mistakes.
structures to support innovation while unshackling Hence, the dynamic capabilities framework is
the enterprise from dysfunctional processes and partially but not entirely in the spirit of evolution-
structures designed for an earlier period. ary theorizing. The dynamic capabilities frame-
In Teece and Pisano (1994) and Teece et al. work endeavors to capture the key variables and
(1997), we proposed three organizational and man- relationships that need to be manipulated to cre-
agerial processescoordination/integrating, learn- ate, protect, and leverage intangible assets so as to
ing, and reconfiguringas core elements of achieve superior enterprise performance and avoid
dynamic capabilities. These processes are a sub- the zero-profit trap. However, building and assem-
set of the processes that support sensing, seiz- bling tangible and intangible assets and effectuat-
ing, and managing threats. Together they might ing change is seen as difficult. Long-run success
be thought of as asset orchestration processes. is likely to require achieving necessary internal
A key strategic function of management is to creative destruction, possibly involving spin-outs
find new value-enhancing combinations inside the and spin-offs to help sustain superior performance.
enterprise, and between and amongst enterprises, Decision biases must also be neutralized. In short,
and with supporting institutions external to the enterprises may be more like biological organ-
enterprise. Because many of the most valuable isms than some economists, managers, and strategy
assets inside the firm are knowledge related and scholars are willing to admit; but they are also
hence nontradable, the coordination and integra- more malleable than some organizational ecolo-
tion of such assets create value that cannot be repli- gists are willing to recognize.
cated in a market. This establishes a distinctive role The enterprise will need sensing, seizing, and
for managers in economic theory and in the eco- transformational/reconfiguring capabilities to be
nomic system. Managers seek new combinations simultaneously developed and applied for it to
by aligning cospecialized assets (Teece, 2007). The build and maintain competitive advantage. Simul-
need to sense and seize opportunities, as well as taneity may not be necessary at the product level
reconfigure when change occurs, requires the allo- indeed, Helfat and Peteraf (2003) distinguish
cation, reallocation, combination, and recombina- between capability development and subsequent
tion of resources and assets. These are the key honing, grafting, and branding. Endeavoring to
strategic function of executives. Indeed, skills used simultaneously achieve sensing, seizing, and recon-
to identify and exploit complementarities and man- figuring at the individual product level could lead
age cospecialization are scarce. Figuring out how to chaos and lack of effectiveness, as routines and
Copyright 2007 John Wiley & Sons, Ltd. Strat. Mgmt. J., 28: 13191350 (2007)
DOI: 10.1002/smj
D. J. Teece

Copyright 2007 John Wiley & Sons, Ltd.


Processes to Processes to Tap Enterprise
Direct Internal Developments in Delineating the
Boundaries to Decentralization
R&D and Select Exogenous Customer
Manage and Near Cospecialization.
New Science and Solution and the
Enterprise Complements Decomposability
Technologies. Analytical Systems Technology. Business Model Continuous
SELECTED Structures, and Control
(and Individual Alignment and
MICRO- Procedures, Platforms.
Capacities) to Learn Realignment of
FOUNDATIONS and to Sense, Filter, Designs and
Processes to Specific
Shape, and Calibrate Incentives
Identify Target Tangible and
Opportunities. for Seizing
Market Selecting Intangible
Processes to Tap Opportunities Building Loyalty Knowledge
Segments, Decision-Making Governance Assets
Supplier and Changing and Commitment Management.
Complementor Protocols
Customer Needs
Innovation. and Customer

Figure 4. Foundations of dynamic capabilities and business performance

DOI: 10.1002/smj
Strat. Mgmt. J., 28: 13191350 (2007)
Explicating Dynamic Capabilities: Nature and Microfoundations 1343

rules in the organization would likely be in a con- To summarize, an enterprises ability to man-
tinuous state of flux. age competitor threats and to reconfigure itself
The first two capabilities recognized as funda- is dependent on its investment activity, which
mentalsensing and seizingare related to but is in turn dependent on its ability to sense an
different from Marchs (1991, 1996, 2006) con- opportunity. This aspect of dynamic capabilities
cepts of exploration and exploitation. March seems indicates that the likelihood of achieving finan-
clear that both are necessary for adaptation, but cial success depends on events and responses to
he has recognized the tensions, if not incompat- them. Formally, let the probability of a high eco-
ibilities, between the two. His argument in part nomic profits ranking for an enterprise, condi-
is that incompatibilities flow from the fact that tional on some extraordinary event E (e.g., an
exploration and exploitation compete for resources exogenous technological change that opens up the
and that the mindsets and organizational routines possibility of a new business opportunity) occur-
needed for one are different from the other, mak- ring,28 be Pr (|E). Then: Pr (|E = Pr (sense|E)
ing simultaneous pursuit difficult, if not impossible Pr (seize|E, sense) Pr (manage threats/transform
(March, 1996: 280). While there is merit to each |E, sense, seize) P r((|E), sense, seize,
assumption, both need to be put into perspective. manage threats/transform).
With respect to competition for resources, sensing As indicated throughout this paper and through-
does not necessarily involve large commitments out earlier treatments by this author, it is also
of resources, at least not relative to seizing. Cer- necessary to assess the issue of the sustainabil-
tain aspects, such as monitoring the environment, ity or nonimitability of both assets and capa-
can be a low-cost activity. Early-stage exploratory bilities. This in turn depends upon a number of
research is also relatively inexpensive. Mansfield factors summarized adequately by the twin con-
et al.s (1971: Table 6.2) studies of new prod- cepts of isolating mechanism and appropriabil-
uct development showed that the cost of early- ity regimes.29 When the appropriability regime is
stage research activities was a small percentage tight and the business enterprises own isolating
of the total new product development costs. For mechanisms are strong, differential performance
instance, the costs of pharmaceutical development can be sustained, at least for a time. Dynamic
typically far exceed those of pharmaceutical dis- capabililties of course require the creation, integra-
covery. Also, with respect to the different mindsets tion, and commercialization of a continuous stream
and routines, while there are undoubtedly tensions, of innovation consistent with customer needs and
these can be relieved by having different organi- technological opportunities.
zation units (or different parts of an organizational Note that in the dynamic capabilities framework,
unit) specializing to some degree on sensing as enterprises must employ sensing, seizing, and
compared to seizing. As Gupta, Smith, and Shal- reconfiguring mechanism to direct their financial
ley (2006: 697) note: exploration or exploitation resources consistent with marketplace needs and
in one domain may coexist with high levels of imperatives. However, as a matter of pure theory,
exploration or exploitation in the other domain. enterprises need not continuously reinvent them-
Of course, the outsourcing of manufacturing and selves. The need to reinvent depends on events,
other aspects of seizing reconciles the issues even anticipated or otherwise. If the ecosystem in which
more starkly, as the routines needed for proficient the enterprise is embedded remains stable, the need
manufacturing then lie external to the firm. to change can be modulated accordingly.30 Indeed,
The need for both exploration and exploitation
is well accepted for adaptative systems, and is 28
Alternatively, one could assess the unconditional probability
embedded in the literature on ambidexterity (e.g., Pr () of earning such profits. Pr () = Pr (|E) + Pr (|
OReilly and Tushman, 2007). This literature rec- E) with Pr (| E) defined analogously to the definition of
ognizes that both exploration and exploitation can Pr (|E) in the text. In competitive markets without dynamic
capabilities, Pr (| E) is likely to be zero.
be assisted by differentiated but partially or weakly 29
Intellectual property protection, the tacit nature of know-how,
integrated subunits (divisions, departments). Sens- and the inherent difficulty of the technology, all affect the ease of
ing activities need to be decentralized with the imitation. Another factor developed in this article is the unique
coalignment of specific assets. Achieving such combinations
information rolling up to top management. Tight may be difficult for imitators to effectuate.
planning will be a part of seizing, but less so of 30
This assumes that the ecosystem remains attractive. If it does
sensing. not, the enterprise will have to consider migrating to a different

Copyright 2007 John Wiley & Sons, Ltd. Strat. Mgmt. J., 28: 13191350 (2007)
DOI: 10.1002/smj
1344 D. J. Teece

if an enterprise controls standards, or can some- govern the rate of change of ordinary capabili-
how help stabilize its own environment, then it ties.31 However, the notion advanced here is that,
may not need to engage in the continuous and at least analytically, dynamic capabilities can be
costly exploration of radical alternatives (March, disaggregated into sensing, seizing, and transfor-
1991). Selecting suitable business models, making mational activities. Enterprises with good dynamic
the right strategic investment decisions, and pursu- capabilities will have entrepreneurial management
ing incremental innovation can keep an enterprise that is strategic in nature and achieves the value-
highly competitive for a decade or so (e.g., Boe- enhancing orchestration of assets inside, between,
ings decision to build the 747, which 30 years and amongst enterprises and other institutions
later is much improved and still competitive in within the business ecosystem. Dynamic capability
some configurations on some routes) if the envi- is a meta-competence that transcends operational
ronment is stable. Excessive internal change for competence. It enables firms not just to invent but
the sake of it can lead to internal chaos and per- also to innovate profitably (Teece, 1986, 2006).
formance failure. The dynamic capabilities framework is integra-
tive. Dosi, Nelson, and Winter (2000: 4) noted at
one point the terminological flotilla in the lit-
RESOURCES/COMPETENCES erature on organizational competences. However,
DISTINGUISHED FROM DYNAMIC perhaps there is now an emerging consensus that
CAPABILITIES resources/competences map well into what his-
torically we have thought of as the enterprises
The dynamic capabilities framework advances a operational capabilities, which help sustain techni-
neo-Schumpeterian theory of the firm and orga- cal fitness. Dynamic capabilities, by contrast, relate
nizational decision making that is recognizable to to high-level activities that link to managements
those familiar with the behavioral theory of the ability to sense and then seize opportunities, nav-
firm, with evolutionary theorizing in economics, igate threats, and combine and reconfigure spe-
and with a Schumpeterian characterization of the cialized and cospecialized assets to meet changing
innovation process. It also builds on what has come customer needs, and to sustain and amplify evolu-
to be known as the resource-based approach. While tionary fitness, thereby building long-run value for
the resource-based approach is inherently static, it investors.
is nevertheless relevant to dynamic capabilities. As If an enterprise possesses resources/competences
noted by Teece et al.: but lacks dynamic capabilities, it has a chance
to make a competitive return (and possibly even
the resource-based perspective also invites consid- a supra-competitive return) for a short period;
eration of strategies for developing new capabil- but it cannot sustain supra-competitive returns for
ities. Indeed, if control over scarce resources is the long term except due to chance. It may earn
the source of economic profits, then it follows
Ricardian (quasi-)rents when demand increases for
that such issues as skill acquisition and learning
become fundamental strategic issues . . . (Teece its output, but such quasi-rents will be competed
et al., 1990a: 9) away. It does not earn those Schumpeterian rents
associated with new combinations and subse-
Zott similarly recognizes that quent recombination, or Kirznerian rents associ-
ated with bringing markets back into equilibrium.
It might earn short-term Porterian rents associated
dynamic capabilities are more than a simple addi-
tion to resource based view since they manipulate with building defenses against competitive forces
the resources and capabilities that directly engender (Porter, 1991: 22), but this is far too reactive for
rents. (Zott, 2003: 120) long-term success. Dynamically competitive enter-
prises dont just build defenses to competition;
Collis (1994) and Winter (2003) also note that they help shape competition and marketplace out-
one element of dynamic capabilities is that they comes through entrepreneurship, innovation, and

ecosystem, or reshaping the ecosystem itself. Both are very As discussed here, dynamic capabilities certainly include this
challenging tasks. element, as well as several others.

Copyright 2007 John Wiley & Sons, Ltd. Strat. Mgmt. J., 28: 13191350 (2007)
DOI: 10.1002/smj
Explicating Dynamic Capabilities: Nature and Microfoundations 1345

semi-continuous asset orchestration and business about managing isolated subsystems. The pursuit
reconfiguration. of benchmarking and the adoption of best prac-
The archetypical enterprise with competences/ tices has helped with the diffusion of discrete
resources but lacking dynamic capabilities will skills, protocols, and procedures. However, accord-
in equilibrium earn a living by producing and ing to one of the fields pioneers, we have not
selling the same product, on the same scale and learned very much about the relationships between
to the same customer population (Winter, 2003: these subsystems (Buffa, 1982: 2). This is one
992). Such an enterprise might even be good place where dynamic capabilities come into play.
at invention, but it will likely fail to capitalize One implication is that special know-how
on its technological accomplishments. The oper- know-how that is difficult to obtain and applyis
ational/technical competences possessed might in- needed to sense opportunities, execute plans, and
clude basic ones such as order entry (to commu- configure and reconfigure assets and systems as
nicate what needs to be made/supplied), billings necessary. Skill in putting things together to cap-
(to collect from customers), purchasing (to decide ture cospecialization benefits is important. Even
what inputs to buy and to then pay suppliers), with respect to operations management, it seems
financial controls (to restrict behavior and prevent the pay-off today is in understanding how sub-
theft), inventory controls (to minimize inventory systems are related and interact together. Put dif-
costs) financial reporting (to access capital), mar- ferently, the understanding of the basic business
keting (to identify customers), and sales (to obtain functions that constitute business administration
orders). Management of these functions is com- and operations management is widely diffused and
monly considered operations management. hence well known, at least in advanced economies.
Operations management is arguably at the foun- The wide diffusion of knowledge with respect to
dation of basic management functions; but while such functions means that much can be outsourced
knowledge of modern production systems took or implemented inside any enterprise with rela-
generations to develop, it is now widely diffused. tive facility. However, by running hard at this, an
The division of labor, uniform standards, the mov- enterprise may manage only to stand stillwhat
ing assembly line, measurement techniques for some refer to as the Red Queen effect. Absent
inspection, and control all of course had to be a broader overarching set of dynamic capabilities,
invented and they now constitute what we now a firm that is merely competent in operations will
think of as the (American) system of production. fail. However, understanding how to enhance per-
Competitive advantage can in theory flow from formance of the enterprise through sensing future
superior operations, or what was referred to ear- needs, making quality, timely, and unbiased invest-
lier as technical fitness. Indeed, the Industrial ment decisions inside a well-designed business
Revolution saw significant differentials open up model, executing well on those decisions, effec-
between craft systems and modern production sys- tuating productive combinations, promoting learn-
tems, and these innovations led to an almost com- ing, reengineering systems that no longer work
plete reordering of the industrial landscape. As well, and implementing good governance remains
Charles Babbage noted almost 200 years ago: enigmatic. The requisite managerial services that
undergird dynamic capabilities cannot be out-
[W]e shall notice, in the art of making even the sourced. Understanding and implementing the pro-
most insignificant of [articles], processes calculated
to excite our admiration by their simplicity, or to
cesses and structures that undergird dynamic capa-
rivet our attention by their unlooked-for results. bilities is enterprise specific, and requires intimate
(Babbage, 1835: 3) knowledge of both the enterprise and the ecosys-
tem in which the enterprise cooperates and com-
However, the postwar period has led to great petes.
progress in the understanding of how produc- In this regard, a useful distinction can be made
tion systems work. Many useful techniques have between entrepreneurs, managers, and administra-
been developed and improved. With developments tors. Administrators are responsible for the day-to-
in the field of management science and opera- day operations and the routine; they help ensure
tions research, precise answers to narrow problems that the enterprise is technically fit, in the sense
exist. Much is known about inventory manage- defined earlier. They are not expected to engage in
ment, scheduling, planning, quality control, and entrepreneurial activities; e.g., they are not relied
Copyright 2007 John Wiley & Sons, Ltd. Strat. Mgmt. J., 28: 13191350 (2007)
DOI: 10.1002/smj
1346 D. J. Teece

on to sense new business opportunities. Nor are free trade and investment, global dispersion in the
they typically expected to discover the need for sources of new knowledge, and the multi-invention
and to design new enterprise-wide operating rou- or systemic character of such innovation have
tines, as this constitutes evolutionary fitness. The upped the ante for modern management. Improv-
distinctions made earlier are implicitly recognized ing quality, controlling costs, lowering inventories,
by Porter (1996) when he claims that operational and adopting best practices (technical fitness)
effectiveness is not strategy. He recognizes that will no longer suffice for long-run competitive
both operational effectiveness and strategy are success. Nor do traditional scale economies in pro-
essential to superior performance, but notes: duction always have the differentiating power they
may once have had. More than scale and scope
The quest for productivity, quality, and speed has advantage are needed. Success requires the cre-
spawned a remarkable number of management ation of new products and processes and the imple-
tools and techniques, total quality management
benchmarking, time-based competition, outsourc- mentation of new organizational forms and busi-
ing, partnering, reengineering and change manage- ness models, driven by an intensely entrepreneurial
ment. Although the resulting operational improve- genre of management constantly honing the evo-
ments have been dramatic, many companies have lutionary and entrepreneurial fitness of the enter-
been frustrated by their inability to translate gains prise. Entrepreneurial managers can sense and
into sustainable profitability. And bit-by-bit, almost
imperceptibly, management tools have taken the even help shape the future, unshackle the enter-
place of strategy. As managers push to improve prise from the past, and stay ahead by augment-
on all fronts, they move farther away from viable ing knowledge assets, protecting them with intel-
competitive positions (Porter, 1996: 61). lectual property rights, establishing new value-
enhancing asset combinations, and transforming
Yet it is perhaps an overstatement to say that oper- organizational and, if necessary, regulatory and
ations management tools and procedures cannot institutional structures. Dynamic capabilities reside
be the basis of competitive advantage, or work in large measure with the enterprises top manage-
against it. If there is a significant, tacit, non- ment team, but are impacted by the organizational
inimitable component of an enterprises superior processes, systems, and structures that the enter-
operational competence, it has the potential for prise has created to manage its business in the
a time to support superior performance (it will, past.
in fact, generate Ricardian rents).32 Nevertheless, Maintaining dynamic capabilities thus requires
superior operational efficiency, while valuable, is entrepreneurial management. The entrepreneurial
not a dynamic capability. management in question is different but related
to other managerial activity. Entrepreneurship is
about sensing and understanding opportunities,
CONCLUSION getting things started, and finding new and better
ways of putting things together. It is about cre-
For open economies exposed to rapid technolog- atively coordinating the assembly of disparate and
ical change, the dynamic capabilities framework usually cospecialized elements, getting approvals
highlights organizational and (strategic) manage- for nonroutine activities, and sensing business
rial competences that can enable an enterprise to opportunities. Entrepreneurial management has lit-
achieve competitive advantage, and then semi- tle to do with analyzing and optimizing. It is more
continuously morph so as to maintain it. The about sensing and seizingfiguring out the next
framework integrates and synthesizes concepts and big opportunity and how to address it.
research findings from the field of strategic man- We have come to associate the entrepreneur
agement, from business history, industrial eco- with the individual who starts a new business
nomics, law and economics, the organizational sci- providing a new or improved product or ser-
ences, innovation studies, and elsewhere. vice. Such action is clearly entrepreneurial, but
Implicit in the dynamic capabilities framework the entrepreneurial management function embed-
is a recognition that relatively open regimes of ded in dynamic capabilities is not confined to
startup activities and to individual actors. It is
Wal-Mart and Dell Inc. have both used differentiated business a new hybrid: entrepreneurial managerial capital-
models to anchor their competitive advantages.) ism. It involves recognizing problems and trends,
Copyright 2007 John Wiley & Sons, Ltd. Strat. Mgmt. J., 28: 13191350 (2007)
DOI: 10.1002/smj
Explicating Dynamic Capabilities: Nature and Microfoundations 1347

directing (and redirecting) resources, and reshap- economy with innovation, outsourcing, and off-
ing organizational structures and systems so that shoring.
they create and address technological opportu-
nities while staying in alignment with customer
needs. The implicit thesis advanced here is that ACKNOWLEDGEMENTS
in both large and small enterprises entrepreneurial
managerial capitalism must reign supreme for An earlier version of this paper formed the basis
enterprises to sustain financial success. Nor is of a lecture delivered in Finland on September
entrepreneurial management merely intrapreneur- 3, 2003, on the occasion of the award of the
ship, as there is a large role for the entrepreneurial Viipuri Prize in Strategic Management. I wish to
manager in external activities, including shaping thank Mie Augier, John Blair, Hank Chesbrough,
the ecosystem. Antonia Cusumano Binette, Connie Helfat, Dan
As discussed, there are obvious tensions and Lovallo, Cyndi Morrow, Richard Nelson, Ikijiro
interrelationships between and amongst the three Nonaka, Christos Pitelis, Dan Schendel, Valeska
classes of capabilities identified. The managerial Scheren-Guivel, Edward F. Sherry, and Sydney
skills needed to sense are quite different from those Winter for many helpful comments and sugges-
needed to seize and those needed to reconfigure. tions on early drafts. I also thank Richard Rumelt
All functions have a significant entrepreneurial for insights provided over the years, which have
and right brain component. Successful enter- undoubtedly shaped my thinking on strategy and
prises must build and utilize all three classes of on dynamic capabilities. Two anonymous refer-
capabilities and employ them, often simultane- ees provided trenchant comments on several dif-
ously. Since all three classes are unlikely to be ferent drafts. These were immensely helpful, and
found in individual managers, they must be some- I am much indebted to them. Patricia Murphy,
where represented in top management, and the Patricia Lonergan, Frances Darnley, and Elizabeth
principal executive officer must succeed in getting Olson provided tenacious assistance in preparing
top management to operate as a team. Of course, the manuscript for publication.
if the principal executive officer has depth in all
three classes of capabilities, the organization has a
better chance of success. REFERENCES
The dynamic capabilities framework goes
beyond traditional approaches to understanding Abernathy WJ, Utterback JM. 1978. Patterns of industrial
competitive advantage in that it not only empha- innovation. Technology Review 80(7): 4047.
sizes the traits and processes needed to achieve Alchian A, Demsetz H. 1972. Production, information
costs, and economic organization. American Economic
good positioning in a favorable ecosystem, but it
Review 62: 777795.
also endeavors to explicate new strategic consider- Amit R, Schoemaker PJH. 1993. Strategic assets and
ations and the decision-making disciplines needed organization rent. Strategic Management Journal
to ensure that opportunities, once sensed, can be 14(1): 3346.
seized; and how the business can be reconfigured Armour H, Teece DJ. 1978. Organizational structure
when the market and/or the technology inevitably and economic performance: a test of the multidivi-
sional hypothesis. Bell Journal of Economics 9(2):
is transformed once again. In this sense, dynamic 106122.
capabilities aspire to be a relatively parsimonious Audretsch DB. 1995. Innovation, growth, and survival.
framework for explaining an extremely seminal International Journal of Industrial Organization
and complicated issue: how a business enterprise 13(4): 441457.
and its management can first spot the opportu- Babbage C. 1835. On the Economy of Machinery and
Manufactures (4th edn). Charles Knight: London.
nity to earn economic profits, make the decisions Bartlett CA, Ghoshal S. 1993. Beyond the M-form:
and institute the disciplines to execute on that toward a managerial theory of the enterprise. Strategic
opportunity, and then stay agile so as to contin- Management Journal , Winter Special Issue 14:
uously refresh the foundations of its early success, 2346.
thereby generating economic surpluses over time. Baumol W. 2006. Entrepreneurship and invention: toward
restoration into microeconomic value theory. Working
If the framework has succeeded in some small paper, Ringberg Castle Presentation, Germany.
measure, then we have the beginnings of a gen- Brandenburger AM, Nalebuff BJ. 1996. Co-opetition.
eral theory of strategic management in an open Harvard Business School Press: Boston, MA.
Copyright 2007 John Wiley & Sons, Ltd. Strat. Mgmt. J., 28: 13191350 (2007)
DOI: 10.1002/smj
1348 D. J. Teece

Branzei O, Vertinsky I. 2006. Pathways to product Ghemawat P. 1991. Commitment: The Dynamics of
innovation capabilities in SMEs. Journal of Business Strategy. Free Press: New York.
Venturing 21(1): 75105. Gottschalg O, Zollo M. 2007. Interest alignment and
Buffa E. 1982. Research in operations management. competitive advantage. Academy of Management
Journal of Operations Management 1(1): 17. Review 32(2): 418437.
Capron L, Dussauge P, Mitchell W. 1998. Resource rede- Grant RM. 1996. Prospering in dynamically competitive
ployment following horizontal mergers and acqui- environments. Organizational Science 7(4): 375387.
sitions in Europe and North America, 19881992. Gupta AK, Smith KG, Shalley CE. 2006. The interplay
Strategic Management Journal 19(7): 631661. between exploration and exploitation. Academy of
Casson M. 1997. Information and Organization: A New Management Journal 49(4): 693706.
Perspective on the Theory of the Enterprise. Oxford Handy C. 1990. The Age of Unreason. Harvard Business
University Press: New York. School Press: Boston, MA.
Chandler A. 1962. Strategy and Structure: Chapters in the Helfat C, Peteraf M. 2003. The dynamic resource-based
History of Industrial Enterprise. Harvard University view: capability lifecycles. Strategic Management
Press: Cambridge, MA. Journal, October Special Issue 24: 9971010.
Chandler A. 1990a. Scale and Scope: The Dynamics Helfat C, Finkelstein S, Mitchell W, Peteraf MA, Singh
of Industrial Capitalism. Harvard University Press: H, Teece DJ, Winter SG. 2007. Dynamic Capabilities:
Cambridge, MA. Understanding Strategic Change in Organizations.
Chandler A. 1990b. The enduring logic of industrial Blackwell: Oxford, U.K.
success. Harvard Business Review 68(2): 130140. Henderson RM. 1994. Managing innovation in the
Chesbrough H. 2003. Open Innovation: The New Impera- information age. Harvard Business Review 72(1):
tive for Creating and Profiting from Technology. Har- 100106.
vard Business School Press: Boston, MA. Henderson RM, Clark K. 1990. Architectural innovation:
Chesbrough H, Rosenbloom RS. 2002. The role of the the reconfiguration of existing product technologies
business model in capturing value from innovation: and the failure of established firms. Administrative
evidence from Xerox Corporations technology. Science Quarterly 35: 930.
Industrial and Corporate Change 11(3): 529555. Henderson RM, Cockburn I. 1994. Measuring com-
Chesbrough H, Teece DJ. 1996. Organizing for inno- petence? Exploring firm effects in pharmaceutical
research. Strategic Management Journal , Winter Spe-
vation: when is virtual virtuous? Harvard Business
cial Issue 15: 6384.
Review 74(1): 6573.
Hughes TP. 1983. Networks of Power Electrification
Collis DJ. 1994. Research note: how valuable are
in Western Society 18801930 . Johns Hopkins
organizational capabilities? Strategic Management
University Press: Baltimore, MD.
Journal , Winter Special Issue 15: 143152.
Iansiti M, Clark KB. 1994. Integration and dynamic
Cyert RM, March JG. 1963. A Behavioral Theory of the capability: evidence from product development in
Enterprise. Prentice-Hall: Englewood Cliffs, NJ. automobiles and mainframe computers. Industrial and
David P. 1992. Heroes, herds and hysteresis in Corporate Change 3(3): 557605.
technological history: Thomas Edison and the battle Jacobides MG, Winter S. 2005. The coevolution of capa-
of the system reconsidered. Industrial and Corporate bilities and transaction costs: explaining the institu-
Change 1(1): 129180. tional structure of production. Strategic Management
Davidow W, Malone M. 1992. The Virtual Corporation. Journal 26(5): 395413.
Harper Business: New York. Jantunen A. 2005. New HRM practices and knowledge
Dosi G, Nelson RR, Winter SG. 2000. Introduction. utilization. In Proceedings of the 5th International
In The Nature and Dynamics of Organizational Workshop on Human Resource Management, Seville,
Capabilities, Dosi G, Nelson, RR, Winter SG (eds). Spain.
Oxford University Press: New York; 124. Kahneman D, Lovallo D. 1993. Timid choices and bold
Dunne T, Roberts MJ, Samuelson L. 1988. Patterns of forecasts: a cognitive perspective on risk taking.
enterprise entry and exit in U.S. manufacturing Management Science 39(1): 1731.
industries. Rand Journal of Economics 19(4): Kahneman D, Tversky A. 1979. Prospect theory: an
495515. analysis of decisions under risk. Econometrica 47(2):
Eisenhardt K, Martin J. 2000. Dynamic capabilities: 263291.
what are they? Strategic Management Journal , Kirzner I. 1973. Competition and Entrepreneurship.
OctoberNovember Special Issue 21: 11051121. University of Chicago Press: Chicago, IL.
Ettlie JE, Pavlou PA. 2006. Technology-based new Klepper S, Graddy E. 1990. The evolution of new
product development partnerships. Decision Sciences industries and the determination of market structure.
37(2): 117147. Rand Journal of Economics 21(1): 2744.
Evans DS, Hagiu A, Schmalensee R. 2006. Invisible Klepper S, Miller J. 1995. Entry, exit, and shakeouts
Engines: How Software Platforms Drive Innovation in the United States in new manufactured products.
and Transform Industries. MIT Press: Cambridge, Internal Journal of Industrial Organization 13(4):
MA. 567591.
Freeman C. 1974. The Economics of Industrial Innova- Knight FH. 1921. Risk, Uncertainty, and Profit. Houghton
tion. Penguin: Harmondsworth, U.K. Mifflin: New York.
Copyright 2007 John Wiley & Sons, Ltd. Strat. Mgmt. J., 28: 13191350 (2007)
DOI: 10.1002/smj
Explicating Dynamic Capabilities: Nature and Microfoundations 1349
Kogut B, Zander U. 1992. Knowledge of the enterprise, Nelson RR, Winter SG. 2002. Evolutionary theorizing in
combinative capabilities and the replication of economics. Journal of Economic Perspectives 16(2):
technology. Organizational Science 3(3): 383397. 2346.
Langlois R. 1992. Transactions-cost economics in real Nonaka I, Takeuchi H. 1995. The Knowledge Creating
time. Industrial and Corporate Change 1(1): 99127. Company. Oxford University Press: New York.
Lazonick W. 2005. The innovative firm. In The Oxford Nonaka I, Toyama R. 2007. Strategic management as
Handbook of Innovation, Fagerberg J, Mowery D, distributed practical wisdom (phronesis). Industrial
Nelson RR (eds). Oxford University Press: New York; and Corporate Change 16(3): 371394.
2955. OReilly CA, Tushman M. 2007. Ambidexterity
Leonard-Barton D. 1995. Wellsprings of Knowledge- as a dynamic capability: resolving the inno-
Building and Sustaining the Sources of Innovation. vators dilemma. (March 2007).
Harvard Business School Press: Boston, MA. abstract=978493 [24 June 2007].
Lippman SA, Rumelt RP. 1982. Uncertain imitability: an Packard D. 1995. The HP Way: How Bill Hewlett and I
analysis of interfirm differences under competition. Built Our Company. HarperCollins: New York.
Phillips A. 1971. Technology and Market Structure: A
Bell Journal of Economics 13(2): 418438.
Study of the Aircraft Industry. Heath Lexington Books:
Lippman SA, Rumelt RP. 2003a. A bargaining perspec-
Lexington, MA.
tive on resource advantage. Strategic Management
Pisano G, Shan W, Teece DJ. 1988. Joint ventures and
Journal 24(11): 10691086. collaboration in the biotechnology industry. In Inter-
Lippman SA, Rumelt RP. 2003b. The payments perspec- national Collaborative Ventures in U.S. Manufactur-
tive: micro-foundations of resource analysis. Strate- ing, Mowery D (ed.). Ballinger: Cambridge, MA;
gic Management Journal , October Special Issue 24: 183222.
903927. Porter M. 1980. Competitive Strategy: Techniques for
Macher JT, Mowery DC. 2004. Vertical specialization Analyzing Industries and Competitors. Free Press:
and industry structure in high technology industries. New York.
Advances in Strategic Management 21: 317356. Porter M. 1991. How competitive forces shape strategy.
Malerba F, Orsenigo L. 1996. The dynamics and In Strategy: Seeking and Securing Competitive
evolution of industries. Industrial and Corporate Advantage, Montgomery C, Porter M (eds). Harvard
Change 5(1): 5187. Business School Press: Boston, MA; 1126.
Mansfield E, Rapoport J, Schnee J, Wagner S, Ham- Porter M. 1996. What is strategy? Harvard Business
burger M. 1971. Research and Innovation in the Mod- Review 74(6): 6180.
ern Corporation. WW Norton: New York. Rosenberg N. 1982. Inside the Black Box. Cambridge
March JG. 1991. Exploration and exploitation in University Press: New York.
organizational learning. Organizational Science 2(1): Rosenkopf L, Nerkar A. 2001. Beyond local search:
7187. boundary-spanning, exploration and impact in the
March JG. 1996. Continuity and change in theories optical disc industry. Strategic Management Journal
of organizational action. Administrative Science 22(4): 287306.
Quarterly 41: 278287. Rumelt R. 1984. Towards a strategic theory of the
March JG. 2006. Rationality, foolishness, and adaptive enterprise. In Competitive Strategic Management,
intelligence. Strategic Management Journal 27(3): Lamb RB (ed). Prentice-Hall: Englewood Cliffs, NJ;
201214. 556570.
March JG, Simon HA. 1958. Organizations. Wiley: New Rumelt R. 1995. Inertia and transformation. In Resource
York. Based and Evolutionary Theories of the Enterprise,
Miles R, Snow C. 1994. Fit, Failure, and the Hall of Montgomery C (ed.). Kluwer Academic: Boston, MA;
Fame: How Companies Succeed or Fail . Free Press: 101132.
Santoro D, McGill JP. 2005. The effect of uncer-
New York.
tainty and asset cospecialization on governance in
Milgrom P, Roberts J. 1990. The economics of modern
biotechnology alliances. Strategic Management Jour-
manufacturing: technology, strategy, and organization. nal 26(13): 12611269.
American Economic Review 80(3): 511528. Schumpeter J. 1934. The Theory of Economic Develop-
Mitchell W. 1991. Dual clocks: entry order influences ment. Harvard University Press: Cambridge, MA.
on industry incumbent and newcomer market share Shane S. 2003. A General Theory of Entrepreneurship.
and survival when specialized assets retain their value. Edward Elgar: Northampton, MA.
Strategic Management Journal 12(2): 85100. Simon HA. 1993. Altruism and economics. American
Monteverde K, Teece DJ. 1982. Supplier switching costs Economic Review 83(2): 156161.
and vertical integration in the U.S. automobile Simon HA. 2002. Near decomposability and the speed
industry. Bell Journal of Economics 13(1): 206213. of evolution. Industrial and Corporate Change 11(3):
Nelson RR. 2005. Technology, Institutions, and Economic 587599.
Growth. Harvard University Press: Cambridge, MA. Somaya D, Teece DJ. 2007. Patents, licensing and
Nelson RR, Winter SG. 1982. An Evolutionary Theory entrepreneurship: effectuating innovation in multi-
of Economic Change. Harvard University Press: invention contexts. In Entrepreneurship, Innovation,
Cambridge, MA. and the Growth Mechanism of the Free-Market
Copyright 2007 John Wiley & Sons, Ltd. Strat. Mgmt. J., 28: 13191350 (2007)
DOI: 10.1002/smj
1350 D. J. Teece

Enterprise, Sheshinski E, Strom RJ, Baumol WJ Teece DJ. 2007. Managers, markets, and dynamic capa-
(eds). Princeton University Press: Princeton, NJ; bilities. In Dynamic Capabilities: Understanding
185212. Strategic Change in Organizations, Helfat C, Finkel-
Teece DJ. 1976. Vertical Integration and Vertical stein S, Mitchell W, Peteraf MA, Singh H, Teece DJ,
Divestiture in the U.S. Oil Industry. Stanford Winter SG (eds). Blackwell: Oxford, U.K.; 1929.
University Institute for Energy Studies: Stanford, CA. Teece DJ, Pisano G. 1994. The dynamic capabilities of
Teece DJ. 1977. Technology transfer by multinational enterprises: an introduction. Industrial and Corporate
enterprises: the resource cost of transferring tech- Change 3(3): 537556.
nological know-how. Economic Journal 87: (June): Teece DJ, Pisano G, Shuen A. 1990a. Enterprise capabil-
242261. ities, resources and the concept of strategy. Consor-
Teece, DJ, 1980. Economies of scope and the scope tium on Competitiveness and Cooperation, Working
of the enterprise. Journal of Economic Behavior and paper CCC 90-8, Institute of Management, Innovation
Organization 1(3): 223247. and Organization, University of California, Berkeley,
Teece DJ. 1981. Internal organization and economic per- CA.
formance: an empirical analysis of the profitability of Teece DJ, Pisano G, Shuen A. 1990b. Firm capabilities,
principal enterprises. Journal of Industrial Economics resources and the concept of strategy. Economic
30(2): 173199. Analysis and Policy Working Paper EAP-38,
Teece DJ. 1982. Towards an economic theory of the Institute of Management, Innovation and Organization,
multiproduct firm. Journal of Economic Behavior and University of California, Berkeley, CA.
Organization 3(1): 3963. Teece DJ, Pisano, G, Shuen, A. 1997. Dynamic capabili-
Teece DJ. 1986. Profiting from technological innovation. ties and strategic management. Strategic Management
Research Policy 15(6): 285305. Journal 18(7): 509533.
Teece DJ. 1988. Technological change and the nature Tushman M, Anderson P. 1986. Technological disconti-
of the enterprise. In Technical Change and Eco- nuities and organizational environments. Administra-
nomic Theory, Dosi G, Freeman C, Nelson RR, Sil- tion Science Quarterly 31: 439465.
verberg G, Soete, L (eds). Pinter: London; 256281. Utterback J, Suarez F. 1993. Innovation, competition, and
Teece DJ. 1990. Contributions and impediments of eco- market structure. Research Policy 22(1): 121.
nomic analysis to the study of strategic management. Wernerfelt B. 1984. A resource-based view of the firm.
In Perspectives on Strategic Management , Fredrick- Strategic Management Journal 5(2): 171180.
son JW (ed.). HarperCollins: New York; 3980. Williamson OE. 1975. Markets and Hierarchies. Free
Teece DJ. 2000. Managing Intellectual Capital: Orga- Press: New York.
nizational, Strategic, and Policy Dimensions. Oxford Winter SG. 2003. Understanding dynamic capabilities.
University Press: Oxford, U.K. Strategic Management Journal , October Special Issue
Teece DJ. 2003. Expert talent and the design of 24: 991996.
(professional services) enterprises. Industrial and Zott C. 2003. Dynamic capabilities and the emergence
Corporate Change 12(4): 895916. of intraindustry differential firm performance: insights
Teece DJ. 2006. Reflections on profiting from innovation. from a simulation study. Strategic Management
Research Policy 35(8): 11311146. Journal 24(2): 97125.

Copyright 2007 John Wiley & Sons, Ltd. Strat. Mgmt. J., 28: 13191350 (2007)
DOI: 10.1002/smj