1

**** DRAFT ****

EXPLICATING DYNAMIC CAPABILITIES:

ASSET SELECTION, COORDINATION, AND

ENTREPRENEURSHIP IN STRATEGIC MANAGEMENT THEORY

David J. Teece

Director, Institute of Management, Innovation and Organization

Professor, Haas School of Business
University of California, Berkeley

August 2003
 2

I. Introduction

Recent work on the theory of the firm stresses that firms consist of bundles or portfolios of fixed

assets and (production) competencies (“resources”). Competitive advantage flows from

whatever unique ability firms have to shape, reshape, configure and reconfigure those assets to

serve customer needs. The particular (non-immitability) capacity firms have to shape, reshape,

configure and reconfigure those assets so as to respond to changing technologies and markets is

what has come to be known as dynamic capabilities.

In this paper the implications of these concepts for strategic management and the theory of the

firm are elaborated further. Greater attention is given here than in Teece and Pisano (1994) and

in Teece, Pisano and Shuen (1997) to the orderly identification of the foundations of dynamic

capabilities. Furthermore, an effort is made to integrate elements of the organizational behavior,

marketing, economics and finance literature to show how they each contribute to the

understanding of dynamic capabilities and strategic management.

II. The (Fundamental) Economic Problem Addressed (Solved) By Dynamic Capabilities:

Coordination In The Face Of Thin Markets1

One of the most tauted virtues of a private enterprise economy is its ability to achieve

coordination of disparate actors -- both consumers and producers -- without central planners.

The price system is of course the mechanism to achieve this. Prices act as signals of scarcity or

1
While coordination/integration was identified as a major element in dynamic capabilities in Pisano and Teece,
1999, Teece et al 1997, it was not given the fundamental status it is given here. This is a deficiency in the earlier
treatment.
 3

abundance. Consumers adjust to price increases by reducing consumption; producers react to the

same signal by increasing production, and the market clears. This simple mechanism means that

a good deal of resource allocation can take place via market mechanisms -- quickly and

efficiently. Commodity markets certainly behave in this fashion.

However, a very large proportion of goods and most services aren’t simply exchanged in open

and organized markets. This is because they are forward transactions (the sale of good x at time

t in the future), contingent transactions (the sale of x if event e occurs), or they are simply

transactions which are so infrequent that a market doesn’t exist. When such transactions occur,

if they do, they are said to take place in thin markets. Such markets are exposed to contractual

hazards, or even if not exposed to hazards, experience significant liquidity discounts. Also, the

difference between bid and ask prices is likely to be large.

Frequently, transactions don’t occur at all because the services that an idiosyncratic asset

provides are difficult to describe and to define. If the asset is a competence, the valuation of the

competence is problematic because its value may depend on complementary assets owned by the

seller, the buyer, or third parties. All of this is to say that certain assets get built rather than

bought (because there may not be a market) and get redeployed inside the firm rather than sold

(because sale in a market is not a good way to extract value). This could be because the market

is simply too thin, or is simply non-existent.

Coordination as an economic problem only occurs because of change (Hayek, 1945). In a static

environment a short period of “set up” would be required to organize economic activity, but
 4

absent change in consumer tastes or technology, economic agents (both traders and managers)

would sort out the optimal flows of goods and services (together with methods of production).

Thereafter, there would be no need for their services.

Now introduce change. If there were a complete set of forward and contingent claims markets,

adjustments would occur automatically; absent a complete set of futures and contingent claims

markets, and there is the need for economic agents to engage in trading activities, and for

managers/entrepreneurs to “integrate, build, and reconfigure internal and external competences

to address rapidly changing environments” (Teece et al., 1997). Adapting effectively to

changing environments is part of a firm’s dynamic capabilities (Cyert and March, 1963).

Adaptation affects search and learning (March, 1994). If the goals of the business firm adapt

rapidly, the resulting indeterminacy of success and failure reduces the precision of learning and

thus reduces short-run returns. However, the same imprecision leads to more frequent subjective

failures, thus stimulates search and exploration that may increase longer-run returns (Levinthal

and March, 1993). Because of these (and other) imperfections, the firm must engage in active

search and imagination to create sustainable strategic opportunities (Winter, 2000).

In short, the need for firms to have dynamic capabilities stems from what can be thought of as

“market failures.” The “market failure” at issue is not one of high transaction costs and

contractual incompleteness. Rather, it is associated with the need to coordinate complementary

assets and make investment decisions -- something, which markets do not do as well as

managers, can do. G.B. Richardson (1960) has remarked upon the information problems
 5

associated with achieving coordination and investment decisions. It is this kind of coordination

which we are concerned with here.

Needless to say, the proficient achievement of the necessary coordination is by no means assured

inside the firm. Decision makers need information on changing consumer needs and technology.

Such information is not always available; or if it is available, is likely to be incomplete, or highly

subjective (Casson, 2000, p. 119; Simon, 1993). Managers are of course decision makers and

they must collect information, analyze it, synthesize it, and act upon it inside the firm. Situations

are dealt with in many ways, sometimes by creating rules, which specify how the organization

will respond to the observations made (March and Simon, 1958). If this path is chosen, then

rules may become codified and routinely applied (Casson, 2000, p 129) whenever certain

changes are detected2.

In some circumstances, new information and new situations may be best dealt with by forming a

new firm (Knight, 1921).3 Those who discover the new information, and can figure out the

appropriate response, need not be the same individual(s) who start a new enterprise; but given

the absence of a well functioning market for information about new market opportunities, the

discoverer and the enterprise founder may need to be one and the same.

2
Casson argues that rule making is entrepreneurial, but that rule implementation is routine, and is characterized by
managerial and administrative work.
3
Frank Knight was (probably) the first to argue a distinct entrepreneurial theory of the firm (Langlois & Cosgel ,
1993). In particular, Knight thought of entrepreneurs as possessing different judgments (and different capacities for
judgments) and acting upon (and profiting from) genuine uncertainty and unpredictability: “[I]t is true uncertainty
which by preventing the theoretically perfect outworking of the tendencies of competition gives the characteristic
form of ‘enterprise’ to economic organization as a whole and accounts for the peculiar income of the entrepreneur”
(Knight, 1921, p. 232).
 6

The need for coordination is not of course entirely exogenous to the firm. As Chandler has

reminded us, the activities of firms and their managers shape markets, as much as markets shape

firms (Chandler 1990; Teece, 1993; Simon, 1991). Put simply, firms and markets coevolve.

Hence, while the need for coordination and investment choices may be the fundamental problem

which the firm’s dynamic capabilities help address, the firm’s dynamic capabilities – particularly

its ability to introduce new products and services into the market – not only shapes markets; it

also requires firm-level responses by competitors, suppliers, and sometimes customers.

II. Resources/Competences (Distinguished from Dynamic Capabilities)

In advancing the view that dynamic capabilities are the basis of sustainable competitive

advantage, and help the economy make investment choices and achieve necessary coordination

of complementary assets, the distinction can be made between operating skills, coordination

skills, and investment skills. In making the distinctions and in showing interrelations, the

concept of dynamic capabilities is further developed to help display its utility in the

understanding of the distinct properties of the business enterprise (as compared to markets). In

addition, an effort is made to better illuminate dynamic capabilities in terms of the underlying

routines and skills that are implicated.

Some of the definitions of resources, competences, and dynamic capabilities, advanced earlier in

the literature, were quite fuzzy. Dosi, Nelson and Winter correctly noted the ‘terminology soup’

in the literature on organizational competences (Dosi, Nelson and Winter, 2000). However, the

emergent near-consensus seems to be that resources/competences map well into what historically

we have thought of as the firm’s operational capabilities. Dynamic capabilities, by contrast,

 7

relate to the firm’s ability to effectuate change. As Collis (1994) and Winter (2003) note, one

element of dynamic capabilities is that they govern the rate of change of ordinary capabilities. If

a firm possess resources/competences but lacks dynamic capabilities, it has a chance to make a

competitive return, but it cannot make a supra competitive return. It may earn Ricardian (quasi)

rents when demand increases for its output, but such quasi rents will be competed away. It

cannot earn Schumpeterian rents (because it’s not innovative). Nor is it likely to be able to earn

monopoly (Porterian) rents since these require exclusive behavior or strategic manipulation4.

The archetypical firm with competences/resources but lacking dynamic capabilities will in

equilibrium “earn a living by producing and selling the same product, on the same scale and to

the same customer population” (Winter, 2003). The operational competences involved include

such basic ones as order entry (to communicate what needs to be made/supplied), billings (to

collect from customers), purchasing (to decide what inputs to buy and to then pay suppliers),

financial controls (to restrict behavior and prevent theft), inventory controls (to minimize

inventory costs) financial reporting (to access capital), marketing (to identify customers), and

sales (to obtain orders). In order to properly perform operations, some level of assets will need

to be deployed. For a goods provider, this may or may not imply own manufacturing and

distribution. Management of these functions is commonly considered operations management.

Operations management is arguably at the foundation of basic management disciplines. While

knowledge of modern production systems took generations to develop, it is now widely diffused.

The division of labor, uniform standards, the moving assembly line, measurement techniques for

4
See Coleman and Teece (1998) for a discussion of Ricardian, Porterian, and Schumpeterian rents. Porterian
(monopoly) rents may or may not involve coordination. Signaling, for instance, can involve competitors. The focus
in this paper is not on rents due to contrived scarcity
 8

inspection and control all of course had to be invented and they now constitute what we now

think of as the (American) system of production5.

Competitive advantage can of course flow from superior operations. Indeed, the industrial

revolution saw significant differentials open up between craft systems and modern production

systems, and these innovations led to an almost complete reordering of the industrial landscape.

As Charles Babbage noted almost 200 years ago: “[W]e shall notice, in the art of making even

the most insignificant of [articles], processes calculated to excite our admiration by their

simplicity, or to rivet our attention by their unlooked-for results.” (Chapter I, p 3)

The post war period has led to great progress in the understanding of how production systems

work. Many useful techniques have been developed and improved. With developments in the

field of management science and operations research, precise answers to narrow problems exist.

Much is known about inventory management, scheduling, planning, quality control, and about

planning isolated subsystems, but according to one of the field’s pioneers, “we have not learned

very much about the relationships between these subsystems” (Buffa, 1982).

One implication is that non-imitable aspects relate to how to configure and reconfigure these

elements -- what has been called dynamic capabilities (Teece et al., 1994, 1997). Even with

respect to operations management it seems, the payoff today is in understanding how subsystems

are related and interact together. This is particularly important in the context of change.

5
Alfred Chandler (1977) writes “modern factory management had its genesis in the US in the Springfield armory.”
 9

Put differently, the understanding of the basic functions that constitute operations management is

today well known, at least in advanced economies. The diffusion of knowledge with respect to

such functions means that (1) they can be outsourced or (2) built inside the enterprise with

relative facility. In a static context, decisions can be made on such matters with reference to

standard contracting frameworks. Indeed, transaction cost economics (Williamson, 1975, 1985)

has evolved to a high level of sophistication and can aid in decision-making.

However, understanding how to change the relationship between the subsystems is a different

situation -- in part because the “services” such capabilities deliver cannot be readily outsourced

i.e. while one can contractually define requirements for requisite components or the subsystems,

the systems integration function is still inherently managerial and entrepreneurial. It involves

new combinations, and the requisite entrepreneurial services cannot be so readily outsourced.

Identifying and implementing such new combinations is one essence -- if not the essence -- of

dynamic capabilities. Strategic management skills are needed because there is no suitable

market in which one can rent entrepreneurial services for discrete projects.6 Of course, venture

capitalists might be thought of as renting entrepreneurs; in reality, however, entrepreneurs

coinvest with venture capitalists -- staking their careers on a deal structure where inventions,

ideas, passion, and vision are exchanged for equity in an enterprise funded by venture capitalists.

A useful distinction can be made between entrepreneurs, managers, and administrators.

Administrators are responsible for the day-to-day and the routine; they are not typically asked to

make decisions relating to the long run. Nor are they expected to engage in entrepreneurial acts,

6
Consultants can of course help firms perform entrepreneurial functions, but the client must provide focus and
coordination.
 10

i.e. they are not expected to sense new business opportunities. Nor are they typically expected to

discover the need for and to design new operating routines, although exceptional administrators

may do so.

The (strategic) management skill set is different from an administrative/operations skill set. Not

only does a (strategic) manager have to see the “big picture,”7 and make decisions that affect the

future course of the organization, but they must also oversee the selection of product strategy,

pricing strategy, branding strategy, procurement strategy, etc. The strategic management

function must not only appreciate how things are done in the firm; they must also appreciate why

they are done the way they are done.

The entrepreneurial function is different but related to the managerial function. The

entrepreneurial function is about getting things started, about finding new ways of doing things

that work. It is about coordinating the assembly of disparate elements, getting “approvals” for

non-routine activities.

In the (American) English language we have come to associate the entrepreneur with the

individual who starts a new business providing a new or improved product or service. Such

action is clearly entrepreneurial; but the entrepreneurial function should not be thought of as

confined to startup activities. As discussed later, dynamic capabilities reside with the firm’s top

management team and involve a significant element of entrepreneurial activity.

7
Including in the big picture are of course the industry, technological, and competitive environment.
 11

The distinctions made above with respect to different managerial functions are substantially

consistent with Porter (1996) when he claims that operational effectiveness is not strategy. He

recognizes that both operational effectiveness and strategy are essential to superior performance,

but notes:

“The quest for productivity, quality, and speed has spawned a remarkable number of
management tools and techniques, total quality management benchmarking, time based
competition, outsourcing, partnering, reengineering and change management. Although
the resulting operational improvements have been dramatic, many companies have been
frustrated by their inability to translate gains into sustainable profitability. And bit-by-
bit, almost imperceptibly, management tools have taken the place of strategy. As
managers push to improve on all fronts, they move farther away from viable competitive
positions.”

Dynamic capabilities can help bridge this divide which Porter seems to think exists. It is

important to note, however, that it is simply not true that “operations management” tools and

procedures cannot be the basis of competitive advantage. If there is a significant, tacit, non-

inimitable component of a firm superior operational competence, it has the potential to support

superior performance (it will in fact generate Richardian rents). Indeed, there is extensive

discussion in Teece et al. (1997) around replication, where we noted the importance of causal

ambiguity and tacit knowledge as factors underpinning immitability.

However, Porter is right to suggest that improved operational effectiveness alone is insufficient

for competitive advantage. If every firm adopts current best practice (as perhaps they should in a

customized way of course) profits will of course be competed away. But not all elements of

operational integration are instantaneously immitable.

 12

Porter correctly notes that “competition strategy is about being different.” However, that does

not only mean, as he suggests, “choosing a different set of activities to deliver a unique mix of

value.” One can produce the same products as ones competitors but establish advantage through

a very different business model (e.g. Walmart and Dell Computer); it is also possible to have

many different types of difficult to imitate operational advantages.

Dynamic capabilities can bridge operations management and strategic management. It is

certainly “strategic” to pursue a highly differentiated operating strategy -- witness e-commerce

versus bricks and mortar commerce used to deliver similar services. Dynamic capabilities is

about being different in non-immitable ways. Porter would, I hope, agree that if one is different

and better at operations, and those differences are non-inimitable, they can be a source of

competitive advantage.

III. Types of Dynamic Capabilities

Firms need to change (effectuate coordination of activities and investments) because of many

factors. Exogenous events (e.g. recession, foreign competition, regulation) will require

responses. So will technological innovations. However, not all firm level responses are

dynamic capabilities. As Sidney Winter (2003) notes “ad hoc problem solving” isn’t necessarily

a capability.

Dynamic capabilities have multiple origins, some rooted in routinized behavior, some rooted in

asset selection/investment choices, and some rooted in creative and differentiated entrepreneurial

acts which involve unusual skills that aren’t particularly imitable.

 13

IV.A. Dynamic Capabilities Through Routinezed Processes

In Teece and Pisano (1994) and Teece et al. (1997) routines (or “processes”) were an essential

element of dynamic capabilities. The treatment was specific in separating out production

routines to sustain current operations from learning routines designing to achieve improvement.

Eisenhardt and Martin (2000) recognize that some “dynamic” capabilities actually consist of

identifiable and specific routines that often have been the subject of extensive empirical research

in their own right. Examples include product development routines, quality control routines, and

technology transfer and knowledge transfer routines.

Eisenhardt and Martin correctly note that the classification of such processes as dynamic

capabilities also enables the empirical foundations of the dynamic capability literature to be

immediately expanded. They also point out that many dynamic capabilities have common

features and cannot therefore be used to undergrid differential performance. They reference

cross function R&D teams which are now widely recognized as essential for superior product

development performance. It is undoubtedly correct that some routine have common features;

but as Eisenhardt and Martin themselves point out, even though there may be common features,

there are also points of difference too. Indeed they note that successful innovation requires

knowledge of customer/user needs; firms that recognize this basic requirement have many

different ways (routines) for figuring out customer’s needs and preferences. Each approach has

different performance outcomes.

 14

However, the existence of common elements by no means suggests that even a particular routine

cannot be the sources of competitive advantage. Studies of the diffusion of organizational

innovations (e.g. Armour and Teece, 1978; Teece 1980) indicate that the diffusion of

organizational innovation is by no means immediate, and significant economic profits can be

earned before diffusion is significant enough to compete away superior returns. Decade long

adoption cycles are not common.

However, it is the firm’s panoply of dynamic capabilities that is a major source of competitive

advantage.8 To date much of the literature appears to define a firms dynamic capabilities almost

entirely in terms of its change routines or routines designing to renovate routines. In Teece et al.

(1997) learning routines were indeed featured. What is elaborated upon below was not featured

in Teece 1997 but was mentioned in Teece 1998. However, it deserves greater emphasis. The

activity to which I now refer is asset selection and orchestration skills achieved by

“intrapreneurship,” entrepreneurship and astute investing.

IV.B. Dynamic Capabilities Through Intrapreneurship and Entrepreneurship

In addition to the routinized foundations of dynamic capabilities stemming from R&D, quality

circles, knowledge transfer and the like, there is perhaps a far more important source of dynamic

capabilities -- the ability not just to sense changing market and technological opportunities, but

to seize them through effectuating “new combinations.” This is where entrepreneurial aspects of

management come into play and where the distinctive (transactional) competence of the firm

8
As noted in Teece et al (1997) and Teece and Coleman (1988) firms can generate Ricardian, Schumpterian, and
Porterian (monopoly) rents. Dynamic capabilities are usually the source of Schumpterian rents. Whereas Ricardian
rents are quasi rents and will be competed away, Schumpterian rents have the possibility of being sustained
indefinitely so long as the dynamic capability is maintained.
 15

meets knowledge/skill competences. The behavior in question is not necessarily a routinized

activity in the sense of Winter (2003); dynamic capabilities of this kind require judgment,

passion, conviction, superior insight, and leadership. 9 It is not ad hoc problem solving. The

activity is inherently entrepreneurial. It involves much more than minimizing transaction costs

and/or avoiding contractual hazards. It involves value creation as well as value protection.

Firms have resources which can be deployed into multiple product arenas. This

conceptualization was first advanced by Penrose (1959). However, Penrose didn’t explain just

how this happened or what astute performance requires for this activity.10 At minimum, it would

appear to require deep knowledge of the firm’s internal capabilities. This might appear to be

readily accomplished by management. However, in enterprises of significant size it is a non-

trivial challenge. While firms may have a codified inventory of tangible assets, intangible assets

are less readily categorized. Indeed, management is sometimes quite ignorant of the firm’s

knowledge assets.

Astute performance of “integration” functions is difficult when firms erect internal boundaries or

“silos.” For instance, the automobile companies in the United States in the 1980s began to once

again experiment with and relearn how to form cross-divisional and cross-functional teams. This

required the tearing down or at least bridging “silos.” These capabilities might sound modest;

however, achieving effective internal cooperation is not a trivial accomplishment. It requires

good incentive alignment and shared goals throughout the organization. It is not just a matter of

9
This is consistent with March’s theme of the relevance of Don Quixote for leadership, see for instance March
(1996).
10
In Teece (1982) attention is given to one element of astute performance, namely, transactional efficiency. This
led to a view that diversification was efficient (from a cost minimizing prospective) it led to efficient transfer of
know how to related businesses.
 16

finding and implementing good integration routines (as discussed above) although this is

undoubtedly part of it. Particularly in (large traditional) hierarchies, achieving functional

“integration” is a major organizational accomplishment. It requires persistent and astute

management (intrapreneurship) to succeed.

The concept of orchestration appears to be related to Porter’s concept of “fit.” As Porter (1996)

notes,

“[S]trategic fit among many activities is fundamental not only to competitive advantage
but also to sustainability of that advantage. It is harder for a rival to match an array of
interlocked activities than it is merely to imitate a particular sales force approach, match a
process technology, or replicate a set of product features.” (p 73) [And] “when activities
complement each other, rivals will get very little benefit from imitation unless they
successfully match the whole system -- frequent shifts in positioning are costly -- strategy
is creating a fit among a company’s activities. The success of strategy depends on doing
many things well – not just a few in an integrating among them. If there is not fit among
activities, there is not distinctive strategy and little sustainability. Management reverts to
the simple task of overseeing independent functions and operational effectiveness
determines an organizations relative performance.” (p 77)

The need for “fit” gives purpose to orchestration, and thus to dynamic capabilities. Orchestration

should be aimed at achieving what Porter calls “fit.”11 Indeed general managements core

function is, in Porter’s view, “defining and communicating the company’s unique position,

making tradeoffs, and forging fit amongst activities. The leader must provide the discipline to

decide which industry changes and customer needs the company will respond to.” On this, we

can agree.

In many ways, the above quotes are statements about key components of a firm’s dynamic

capabilities. However, dynamic capabilities introduce a broader lens than the “fit” analysis that

11
“Fit” appears to be rather poorly defined. It is not clear what objective criteria (if any) lie behind it.
 17

Porter advances. Put differently, Porter has defined strategy around only one of several sources

of dynamic capabilities. Each source of dynamic capabilities can be an independent basis of

sustainable competitive advantage. Each is in that sense “strategic.”

IV.C. Dynamic Capabilities through the External Orchestration of Assets (Alliance, Joint

Venture, and Vertical Integration Decisions)

As mentioned earlier, in stable environments firms can readily calculate whether and when to

outsource. Where the revenue implications of different outsourcing/firm boundary decisions are

clear, then minimizing (optimizing) transactions costs makes sense. Indeed, the implicit

assumptions in transaction cost economics (Williamson, 1985) are that revenues and production

costs are invariant to the choice of governance mode. This is an analytical convenience, but the

ceteris paribus assumption rarely holds in reality. A broader framework requires one to identify

at least three elements of an integration (boundary choices) decision: governance,

reconfiguration, and asset selection.

IV. C(i). Governance Costs

In the transaction cost economic framework, firm boundary issues get decided purely on certain

governance cost considerations. The emphasis is not on selecting the activities that need to be

done; rather it is on determining where (organizationally) they are to be done; and if done outside

the firm, the emphasis is on designing contracts that are efficient, in the sense that “hold up” or

recontracting costs are avoided. Managing contractual hazards is undoubtedly important; but as

a practical matter, there are other elements of governance that are relevant too. For instance,
 18

does the firm have sufficient surplus management/governance resources available to provide

governance and/or contractual oversight?

The very nature of the contract will itself determine how much governance is required. For

instance, a procurement contract can be a take-or-pay, or a requirements contract. There may be

many performance conditions which need to be monitored and enforced. Clearly, the amount of

contract administration and the exposure to contractual hazards is not a constant, but is itself a

function of contractual terms, the (trust) relationship between the parties, and jurisdictional

issues.

IV.C.(ii) Configuration and Reconfiguration Costs and Benefits

Strategic flexibility requires that firms constantly rejigger both their internal and external

portfolios of assets and resources (see IV.D. below). How the firm achieves this is of course a

capability in and of itself. Figuring out what to own/control through integration involves an

understanding of both the likely demand for various assets, the firm’s ability to integrate desired

assets, and its ability to divest undesirable assets. Indeed, the framework in Teece (1986) is

designed to provide guidance on such decisions in the context of innovation. Effectuating asset

selection/investment and integration strategies is itself an acquired skill.

IV.D. Asset “Selection” Issues

IV. D.(i) General

Setting aside governance and reconfiguration costs and benefits, there is a real issue of whether

an activity/business is a “good business” for the firm to be in or not, or a good investment to

 19

make at this time. This assessment has many elements common to the assessment an investor

would bring to an asset. Strip the decision of governance (including integration issues) as well as

potential reconfiguration issues, and management is left with a pure asset selection (investment)

decision. In this regard, the decision skills required of management have something in common

with a private equity investor. The main difference is the ability of the strategic manager to

make complementary investment and the ability to simultaneously orchestrate portfolios of

assets. The private equity investor typically lacks this kind of integration capacity.

Overtime, management’s initial asset selection decisions and reinvestment decisions come to

define the asset positions of the firm. In Teece et al (1997), particular classes of assets were

recognized, including in technological assets, complementary assets, financial assets, operational

assets, and structural assets. If the firm owns and controls these assets, then at any point in time

management has some control over how these assets are going to be used, and the owners of the

firm have residual claims on the income generated by these assets, working in conjunction with

other assets.

The management literature is rather silent on the processes by which firms endeavor to make

important investment decisions. While there is of course discussion of project financing criterion

(e.g. discounted with flow, payback periods and the like) there is often little recognition that the

past and present investment decisions of managers are a significant determinant of firm level

performance. Of course, there is extensive discussion around selecting “attractive” industries.

 20

In this regard it is remarkable that the finance and strategic management literatures (and

approaches) rarely meet. There are reasons why investor decision theory sometimes doesn’t map

at all well onto strategic management decisions. Before discussing these decisions, one should

of course note that the project finance and related literatures do provide tools and clear decision

rules for project selection once cash flows are specified, with or without uncertainty bands.

However, the essence of strategic management is that it involves a panoply of interrelated

(complementary) investments. However, the returns to individual complementary assets cannot

generally be neatly petitioned -- and the utility of traditional criteria is thus significantly reduced.

Also, expected cash flows in new investment frequently depend very much on competitive

responses.

Accordingly, judgments need to be made around not just competitive responses but also the

payoffs from interrelated investments. Sometimes this can be precisely modeled; for instance, if

one builds a new chemical plant or petroleum refinery, investment in various processing units

can be modeled using linear programming tools. But in many circumstances, the linkages are

much weaker, and the required investments are not in specific “projects” per se. For instance,

Alfred Chandler’s (1990) assessment that the successful enterprises from the 1870s through the

1960s century were those that pursued his “three pronged” strategy: (1) early and large scale

investments behind new technologies; (2) investment in product-specific marketing, distribution,

and purchasing networks; and (3) recruiting and organizing the managers needed to supervise

and coordinate functional activities. Clearly, these interrelated investment decisions do not lend

themselves particularly well to the standard tools of project finance.

 21

There is also a remarkable dearth of analysis on how managerial investment decisions differ

from investor or “banker” perspectives. In the context of his analysis on the conglomerate form

of business organizations, Oliver Williamson (1975) introduced the idea of the capital budgeting

function of the firm creating in essence what he called a “miniature capital market.” The

distinguishing characteristics of this “market” were defined in informational terms: the internal

capital market has greater depth of information (because of managers deep situational

knowledge) whereas the external capital market, which is the domain of bankers (who have

greater breadth of information).

Other than through Williamson, the other two major interfaces between strategic management

issues and finance are Michael Jenson’s insightful but controversial work on corporate control

where he argues in essence that the discipline of debt is necessary to take away managers

financial and investment discretion. However, it is the prudent and visionary exercise of this

discretion, which is the essence of value creation at the enterprise level, as Chandler so

strenuously, argues. The tension between two of Harvard’s eminent scholars can perhaps be

resolved as follows: where the discipline of debt is value creating is when corporate governance

is weak or otherwise ineffective.

The other treatment is Helfat’s (1988) study of work on investment choices in industry. Helfat

used portfolio, and not capital market theory, to explain non-financial investment choices by

firms. She showed that firms attempt to minimize total firm investment risk. Managers simply

did not believe like capital market (portfolio) investors.

 22

None of these studies come firmly to grips with the nature of (strategic) investment decisions

made by top management, and the (equity) investment decisions made by individual or

institutional investors. This is quite unfortunate. The essence of the difference would appear to

be the following:

(1) Managers frequently “build” rather than buy assets.

(2) Investment decisions frequently have to be made with respect to complementary assets
e.g. at a trivial level, the gas station owner deciding whether to add a car wash; the
manufacturer or cars deciding whether to make trucks or SUVs.

(3) Relatedly, the assets they invest in are frequently intangible e.g. Chandler’s managerial
hierarchies, or marketing competences. Such assets create value only in conjunction with
others.

(4) Relatedly, many investment decisions are made incremental and in an evolutionary
fashion. This is because investment choices are often constrained by annual capital
budgets.

(5) An important component of the value for an investment decision is the real options
component. Many investments, including investments in excess capacity, investments in
early stage R&D, and investments in strategic alliances can be thought of in this fashion.
(see below).

The skill set required for the strategic management function clearly has overlap with general

investment skills; but as a general matter, managerial investing skills require a more detailed

understanding of interrelationships. The skill of a shrewd investor is clearly valuable in a

managerial setting, but much more is needed. One-way to state it, given the introductory

comments made earlier, is that what is needed is a competence for investing in “thin” markets.

The required competence is in part a contractual savvy (in which guarding against opportunism

is just one component) and in part a competence around understanding market and technological

developments, in part a competence in accessing information and making judgments about

 23

competitors behavior, and in part a traditional competence in respecting (where appropriate) and

understanding traditional (project) finance decision criterion.

IV.D.(ii) Asset Selection and “Positioning”

Porter has corrected noted that “positioning” remains critical to competitive advantage.

“Positions” were in fact quite central to Teece et al (1997), although the treatment was different.

The context of the discussion on “positions” was focused on non-tradable specific assets, and

perhaps neglected the breadth of investment decisions which managers make. For instance, the

decision of an integrated oil refinery to build another refinery will clearly have important

implications for its profitability. Refineries have a good deal of physical operating complexity

(and hence flexibility as to the crude oils they will run and the petroleum products they will

make). There are well-developed markets for both crude oil and refined products. Accordingly,

the decision to build a refinery (where there is little in the way of transaction specific assets) is

very much a pure investment decision or “asset play.” In making such investment decisions,

managers will need to consider whether they will want greater risk/return exposure to petroleum

refining, because performance of the asset will depend importantly on margins in the refining

sector.

While there is no strong consensus in the strategic management literature on the importance of

industry effects with Schmalansee (1985) believing them to be major, and Rumelt’s (1991) study

suggesting it is minor, there is undoubtedly a component of firm level profitability that can be

explained by choice of industry (Rumelt estimates this at 8.3%).

 24

The asset selection capabilities discussed here are not macro or cross industry in nature. They

are “situational.” For instance, management’s decision to invest in complementary assets is

typically not an industry level selection decision; the firm is usually already in an industry when

making investment choices regarding complementary assets (Teece, 1986). Investment decisions

with respect to complementary assets typically involve decisions as to whether to invest to a

greater or lesser degree i.e. they don’t implicate the choice of industry. However, they do

implicate “position” in the sense of Teece (1997).

It appears quite important to note, that when Porter discusses the origins of strategic positions, he

has something quite specific in mind. Porter (1996) states that positioning can be (1) based on

producing a subset of an industry’s products and services, (2) based on serving most or all the

needs of a particular group of customers, and (3) segmenting customers who are accessible in

different ways. Porter calls this access based positioning.

Porter’s view of positioning is much more customer and market focused. In the dynamic

capabilities framework, by contrast, positioning relates more to the assets/competences in which

the firm chooses to invest. This is because in the Penrosian view of the world which I adopted

long ago (Teece, 1980, 1982) resources are to some extent fungible; they represent a bundle of

potential services and can, for the most part, be defined independent of their use.

Nonetheless, Porter’s discussion of position suggests the need to separately identify marketing

decisions as a key dynamic capability. Indeed, a key element of dynamic capabilities involves

the processes by which firms decide what customer segments to serve and how to serve them
 25

(e.g. the specification of the product as to price–performance, etc, (Bacon, Beckman, Mowery

and Wilson, 1994)). Getting this right is obviously a key to performance. The marketing

literature is replete with methodologies for doing so.

IV.D.(iii) Real Options

The asset selection component of managerial decision making is where real options

considerations come into play. The value of an asset is not just its current cash flow, but also its

option value. This could be high if (for instance) both (1) demand for the services of the asset

fluctuates, and (2) one cannot access these services (contractually) with ease. Thin markets mean

that the option component of an assets value is likely to be higher than in highly liquid markets,

because if the asset is suddenly in demand, it cannot be readily accessed by owners, unless there

is a prearranged contractual right to do so.

In an Arrow-Debreu economy with a complete set of contingent claims markets (and without

governance and related concerns), there would of course be no need to own such assets in order

to capture the upside associated with asset appreciation. However, we don’t live in an Arrow-

Debreu economy, and moreover, governance cost and configuration/reconfiguration costs are

non trivial, and so integrated and diversified firms have an important role to play. What might

simply be financial decisions in a frictionless world become strategic decisions because of the

illiquidity of the assets, (due to the absence of markets).

This can be illustrated with further exploration of the concept of complementary assets. Teece

(1986) saw the asset selection component of complementary assets as being critical to profiting
 26

from innovation. Indeed, the Teece1986 framework collapsed asset selection, financial capacity,

and contracting costs together where arguably they should be separated in order to facilitate

decision-making.

Asset selection issues are not of course confined to external sourcing (own versus rent). Even

when the decision to build an internal capability has already been made, and the asset is already

inside the firm, the question still remains as to whether the asset ought be improved (through

reinvestment) or shed (divestiture). But once the asset is inside, such decisions are less rigorous

in the sense that the asset is relatively illiquid, and so spinning out marginal assets isn’t

necessarily feasible or desirable, even if the asset isn’t performing as expected.

Asset selection issues that management must face are of course highly path dependent. If a firm

is contemplating whether an activity is a “good business to be in,” it is inevitable that attention

will be focused on basic supply-demand elements: is the demand for the asset’s services strong,

does the firm have free cash flows, are competitors better positioned? In essence, asset selection

issues are highly path dependent. For instance, the relative positioning of a competitor, the

availability of cash, and the timeliness of a desired investment all depend on the firm’s internal

resources and the posture/position of competitors. Hence, the exercise of dynamic capabilities

must take this into account.

There are several functions to an astute asset selection (asset “picking”) capability: (1) real

options decision frames; (2) flexible organizations that are able to acquire and divest with

reasonable alacrity; (3) managers with a bias for action.

 27

V. Real Options Decision Frames

“In a narrow sense, the real options approach is the extension of financial options theory to

options on real (nonfinancial) assets.” (Amram and Kulatilaka, 1999) While the detailed

contractual terms (e.g. strike price, length) are precisely specified in most financial options, the

real options embedded in strategic investments need be identified and specified because they are

usually only implicit.

Stewart Meyers noted that “strategic planning needs finance.” The response here is in the

affirmative but with several caveats. As discussed earlier, finance theory alone isn’t sufficient to

create a robust framework for strategic management decisions. In particular, strategic planning

(and strategic management theory) needs dynamic capabilities as a framework to integrate

finance, governance, and organizational behavior issue. Absent all three, one does not have a

framework robust enough to make good management decisions.

As Kylaheiko’s (1998) points out, scholars adopt the “methodology of isolation in order to make

sense of the pervasive complexity of the real world” (p.320). In economics, partial equilibrium

analysis is an illustration. Transaction cost economics also adopts methodological isolation

when it focuses on costs and ignores benefits. Analyzing issues in “isolation” is of course

methodologically sound; particularly in static environments with low uncertainty. A case in

point is regulated industries or environments otherwise protected from Schumpeterian gates of

creative destruction, or just simply changes in consumer preferences. But where uncertainty
 28

(technological or market) and complexity are pervasive, methodological isolationism is fraught

with hazards and as commonly employed appears to obscure rather than expose central issues.

The focus in this section, however, is on how a competence in real options can support dynamic

capabilities. First, it is important to recognize that real options is first and foremost a way of

thinking. According to Amram and Kulatilaka, there are three components.

(i) Options are contingent decisions:

Options provide the opportunity to make a decision after you see how events unfold.

By contrast, fixed (noncontingent) decisions have linear payoffs because no matter

what happens, you make the same decisions. This observation in and of itself has

implications for what firms are about. To the extent that firms can be thought of as

bundles of assets/resources, the nature of the (implicit) optionality that firms have

over such assets is quite different from if those assets are owned by another firm

which has implicit optionality over the asset. The costs associated with

disassembling these implicit contracts and transferring assets between and amongst

firms are accordingly by no means trivial. That’s not to say that the services of assets

cannot be accessed if they belong to another enterprise; but except in unusual

circumstances, the costs of doing so are greater than the internal assets. In this regard

previous explanations provided for the differential [e.g. the flexibility of managerial

fiat versus the inflexibility of Contracts (Williamson, 1975)] are inadequate.

(ii) Option valuations are aligned with financial market valuations:

The real options approach endeavors to reference market prices, where possible, in

determining asset values and in making asset selection (and divestment) decisions.
 29

This discipline is favored by economists and it does enable more objective decision-

making. Although quantitative analysis can never be relied upon completely, it is

possible to calibrate option values (at least roughly) in the context of business

decision-making.

(iii) Strategic investments can be managed proactively:

The optionality in a strategic investment (whether inside the firm, or a potential

acquisition) must first be recognized before it can be measured. Once categorized

and measured it can be monitored and then proactively managed. Growth options,

flexibility options, exit options, and learning options (Amran and Kulatikala, 1999,

p.10-11) can all be recognized. Managers can then use the options they have (through

the firm’s portfolio of assets/competencies) to flexibly respond to unfolding events.

Real options thinking can be used when there are contingent investment decisions,

when uncertainty is high (and it’s sensible to wait for uncertainty to be at least

partially resolved), when the value is resident in future growth options rather than

current cash flow, and when there are opportunities for mid-course corrections to

investment strategies. Options to delay, to temporarily shut down, to abandon, and to

rewrite investment plans in midstream can significantly increase value. When a firm

“exercises” investment and disinvestments options it clearly changes its own

boundaries.

Also, in many fast-paced industries investment in one cycle of product development creates the

opportunity to invest in the next. In this regard, Andy Grove, former CEO of Intel, claimed that

all Intel was sure of when it succeeded with one generation of microprocessors is that it would
 30

have a seat at the table when it came to the next round of innovation. This does speak to the

fragility of Intel’s position; it also underscores the value of the “seat,” and the option value

associated with succeeding in each round of innovation.

VI. Operations Management, Strategic Management and Entrepreneurship

Strategic management can now be partially separated from operations management as the former

deals with designing and implementing difficult to imitate “change” routines, achieving efficient

asset selection (with due regard to complementarities and to real option values), and perform

integration/orchestration (“fit”) functioning. Operation management becomes strategic only

when new difficult to imitate methods of coordinating operations are being developed and

implemented.

The deep economic question to ask is what is the role of management in the dynamic capabilities

framework? If, as Winter and others suggest, dynamic capabilities are defined mainly around

high-level routines, is the role of (strategic) management reduced and relegated to selecting new

routines? Certainly if innovation becomes truly a routine in large firms, then the

manager/intrapreneur is of secondary importance.

In strict evolutionary views of the world, there is no specific agent and no hierarchy responsible

for regulating the evolutionary process (Cohendet, Llerena and Marengo, 2000). However, even

in a strongly evolutionary view of the world, economic agents of some kind (possibly managers)

must direct the reinvestment of cash flow, and must allocate resources as between exploration

and exploitation (March, 1991, 1994).

 31

In a less evolutionary view of the world, the entrepreneurial function takes on a far greater

significance. The entrepreneur need not be an individual; in the modern corporation it is a

function. As Schumpeter (1949) noted: “The entrepreneurial function may be and often is filled

cooperatively -- in many cases, therefore, it is difficult or even impossible to name an individual

that acts as “the entrepreneur.” (pp 71-72)

The entrepreneur/manager must articulate goals, set culture, build trust, and play a critical role in

the key strategic decisions. Clearly the role of the entrepreneur and the manager overlap to a

considerable extent. As Simon (1991) recognized:

“Especially in the case of new or expanding firms, the entrepreneur does not face an
abstract capital market. He or she exerts much effort to induce potential investors to share
the company’s views (often optimistic) about its prospects. This executive is much closer
to Schumpeter’s entrepreneur than to the entrepreneur of current neoclassical theory.
Whether the firm expands or contracts is determined not just by how its customers
respond to it, but by how insightful, sanguine and energetic its owners and managers are
about its opportunities” (p. 31).

The entrepreneur/manager plays a key role in achieving asset selection and the “coordination” of

economic activity, particularly when complementary assets must be assembled. The

entrepreneur/manager can buy or sell investments/assets, orchestrate internal assets

(intrapreneurship) and transact with the owners of external assets (entrepreneurship). The astute

performance of this function will help achieve what Porter calls “fit”, not just with internally

controlled assets, but with the assets of alliance partners. The entrepreneur/manager can also

shape learning processes with the firm. These are not functions which can be achieved by

markets divorced from entrepreneurs/managers.

 32

Thus the entrepreneur/manager function in the dynamic capabilities framework is in part

Schumpeterian (he introduces novelty and seeks new combinations) and in part evolutionary (he

endeavors to provide and shape learning). Both intrapreneur and entrepreneur, the function

senses new opportunities and engages the organization to seizing them. The entrepreneur/

manager must therefore lead. These are roles not recognized by economic theory; but these roles

are the essence of dynamic capabilities and are absolutely critical to the theory of strategic

management.
 33

REFERENCES

Martha Amram and Nalin Kulatilaka (1999). Real Options: Managing Strategic Investments in
an Uncertain World. Oxford: Oxford University Press.

Armour, Henry and David J. Teece (1978). “Organizational Structure and Economic Performance:
A Test of the Multidivisional Hypothesis,” The Bell Journal of Economics, 9:2 (Spring): 106-122.

Chandler, Alfred (1977). The Visible Hand: The Management Revolution in American
Business. Belknap Press.

_______________(1990). Scale and Scope. Bellknap Press.

Babbage, Charles (1835). On the Economy of Machinery and Manufactures (4th edition).

Bacon, Glenn, Sara Beckman, David C. Mowery, and Edith Wilson (1994). "Managing Product
Definition in High-Technology Industries: A Pilot Study," California Management Review, 36:3
(Spring): 32-56.

Buffa, Elwood (1982). “Research in Operations Management,” Journal of Operations

Management, 1(1): 1-7.

Casson, Mark (2000). “An Entrepreneurial Theory of the Firm,” in Nicolas Foss and
Volker Mahnke (eds.), Competence, Governance, and Entrepreneurship, Ch. 6. Oxford
University Press.

Cohendet, Patrick, Patrick Llerna, and Luigi Marengo (2000). “Is There A Pilot in the
Evolutionary Firm?” in N. Foss and V. Mahnke, op. cit.

Collis, David J. (1994). “Research Note: How Valuable are Organisational Capabilities?”
Strategic Management Journal, 15: 143-152.

Cyert, Richard M. and James G. March (1963). A Behavioral Theory of the Firm. Oxford:
Blackwell Press.
Dosi, Giovanni, Richard Nelson and Sidney G. Winter (2000). “Introduction,” in Dosi, Nelson
and Winter (eds), The Nature and Dynamics of Organizational Capabilities. Oxford University
Press.

Eisenhardt, Kathleen and Jeffrey Martin (2000). “Dynamic Capabilities: What are they?”
Strategic Management Journal, 21: 1105-1121.

Hayek, F.A. (1945). “The Use of Knowledge in Society,” in F.A. Hayek, Individualism and
Economic Order. Chicago: University of Chicago Press.
 34

Helfat, Constance (1988). Investment Choices in Industry. Cambridge, MA: MIT Press.

Jensen, Michael (1989). “Eclipse of the Public Corporation,” Harvard Business Review
(September – October), 89(5): 61-74.

Knight, Frank (1921). Risk, Uncertainty and Profit. Boston, MA: Houghton Mifflin.

Langlois, Richard and M. Cosgel (1993). “Frank Knight on Risk, Uncertainty, and the Firm,”
Economic Inquiry, 31: 456-465.

Levinthal, Daniel and James March (1993). “The Myopia of Learning,” Strategic Management
Journal, 14: 95-112.

March, James G. (1991). “Exploration and Exploitation in Organizational Learning,”

Organization Science, 2: 71-87.

_______________ (1994). “Three Essays on Efficiency and Adaptation in Organizations,”

Swedish School of Economics and Business Administration.

_______________ (1996). “A Scholars Quest,” Stanford Graduate School of Business

Magazine, 1996.
_______________ and Herbert A. Simon (1958). Organizations. Oxford: Blackwell.

Penrose, Edith (1959). The Theory of the Growth of the Firm. Oxford University Press.

Porter, Michael (1994). “Toward a Dynamic Theory of Strategy,” in Richard P. Rumelt, Dan E.
Schendel and David J. Teece (eds.), Fundamental Issues in Strategy. Boston: Harvard
Business School Press.

_______________ (1996). “What is Strategy,” Harvard Business Review (November/December).

Richardson, George B. (1960). Information and Investment. Oxford University Press.

Schmalansee, Richard (1985). “Do Markets Differ Much?” American Economic Review, 75 (3):
167-185.

Schumpter, Joseph (1949). "Economic Theory and Entrepreneurial History," in Change and the
Entrepreneur. Prepared by the Research Center in Entrepreneurial History, Harvard University.
Cambridge, MA: Harvard University Press, pp. 63-84.

Simon, Herbert A. (1991). “Organizations and Markets,” Journal of Economic Perspectives 5: 25-
44.

_______________ (1993). “Strategy and Organizational Evolution,” Strategic Management Journal,

14: 131-142.
 35

Teece, David J.(1993). “The Dynamics of Industrial Capitalism: Perspectives on Alfred

Chandler’s Scale and Scope,” Journal of Economic Literature, 31: 199-225.

_______________ (1986). "Profiting from Technological Innovation," Research Policy, 15(6): 285-
305.

_______________ (1982): "Towards an Economic Theory of the Multiproduct Firm," Journal of

Economic Behavior and Organization, 3, 39-63.

_______________ (1980). “The Diffusion of an Administrative Innovation,” Management Science,

26(5) (May): 464-470.

_______________ and Mary Coleman (1998). “The Meaning of Monopoly: Antitrust Analysis in
High-Technology Industries,” The Antitrust Bulletin (Fall-Winter): 801-857.

_______________ and Gary Pisano (1994). "The Dynamic Capabilities of Firms: An Introduction,"
Industrial and Corporate Change, 3(3).

Teece, David J., Gary Pisano and Amy Shuen (1997). “Dynamic Capabilities and Strategic
Management,” Strategic Management Journal, 18(7): 509-533.

Williamson, Oliver (1985). Economic Institutions of Capitalism. New York: The Free Press.

_______________ (1975). Markets and Hierarchies. New York: The Free Press.

Winter, Sidney (2000). “The Satisfying Principle in Capability Learning,” Strategic

Management Journal, 21: 981-996.

_______________ (2003). “Dynamic Capabilities,” unpublished manuscript.