The Knowledge-Leveraging Paradox: How to Achieve Scale without Making Knowledge Imitable

Author(s): Russell W. Coff, David C. Coff and Roger Eastvold

Source: The Academy of Management Review, Vol. 31, No. 2 (Apr., 2006), pp. 452-465
Published by: Academy of Management
Stable URL: http://www.jstor.org/stable/20159211 .
Accessed: 18/06/2014 15:13

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.

Academy of Management is collaborating with JSTOR to digitize, preserve and extend access to The Academy
of Management Review.

http://www.jstor.org

This content downloaded from 195.78.109.96 on Wed, 18 Jun 2014 15:13:38 PM

All use subject to JSTOR Terms and Conditions
 ?
Academy o?Management f?eview
2006,Vol. 31,No. 2, 452-465.

NOTE

THE KNOWLEDGE-LEVERAGING PARADOX:

HOW TO ACHIEVESCALEWITHOUTMAKING
KNOWLEDGEIMITABLE
RUSSELL W. COFF
Emory University

DAVID C. COFF
Xtelesis

ROGER E?STVOLD
e-Diagnostic Network

Tacit knowledge, one of the most promising resources, is among the toughest to

leverage. It is important because it is often inimitable. However, increasing its scale

may require codification, which may make it imitable. We explore and illustrate an

approach to applying information technologies to leverage scarce expertise without

codifying or transferring knowledge. Such technology may allow individuals to fur

ther specialize and generate more tacit knowledge, preserving the strategic properties
of knowledge while scaling up.

Tacit knowledge has strategic significance cult task by definition (Teece, 1977). Further
because of its potential to be valuable, rare, and more, should this be accomplished, the codified
hard for rivals to imitate or acquire (Barney, knowledge might be easier for rival firms to
1991). That is, the fact that the knowledge is hard acquire or imitate, thereby limiting the sustain
to transfer may help firms sustain an advan
ability of the advantage (Lado & Zhang, 1998).
tage. However, this creates management dilem Thus, the very property that grants strategic
mas within firms wishing to achieve an advan value also makes a very hard re
knowledge
tage using this resource (Coff, 1997; Zander & source to leverage.
Kogut, 1995). Once valuable tacit knowledge is We first highlight the paradox in the extant
identified, firmsmust transfer and replicate it to literature that tacit knowledge has great strate
increase the scale and meet the demand for the
gic value but that itmust be codified to exploit
scarce resource (Tsai, 2001). This, in turn, often that value. We then develop about how
theory
requires face-to-face communication and pro information technology can be to lever
applied
tracted learning curves
(Polanyi, 1966; Teece, tacit knowledge without transferring or cod
age
1977). Firms are therefore apt to face an internal new research
ifying the knowledge, suggesting
shortage of talent, hampering efforts to build a directions for the information technology litera
strategy that involves leveraging tacit knowl ture. We this using two vignettes in
explore
edge. which information technology was applied to
While information technology might be posed
leverage expertise: (1) silicon wafer manufactur
as a solution to this problem, there are many
ing and (2) a rural hospital in West Texas. In
hurdles to be overcome. The first is that infor
both cases the technology allowed experts to
mation technology often requires the knowledge work remotely and leverage their skills and
to be codified and/or transferred?a very diffi across locations. In this
knowledge multiple
way, a highly specialized and well-honed exper
tise was leveraged without codifying, transfer
We thank Maryam Alavi, Elliot Bendoly, Anandhi Bharad
comments and or even in face-to-face communi
waj, and Terri Griffith for helpful sugges ring, engaging
tions. cation. However, in the latter case, the

452

This content downloaded from 195.78.109.96 on Wed, 18 Jun 2014 15:13:38 PM

All use subject to JSTOR Terms and Conditions
 2006 Coff, Coff, and Eastvold 453

knowledge was not rare or unavailable to rivals petitive advantage and that knowledge-based
and did not appear to confer an advantage. In capabilities must typically be built rather than
deed, the hospital ultimately underwent a major purchased externally (Dierickx & Cool, 1989).
management shake-up owing to its poor perfor
mance.
The Dilemma of Scaling Up Knowledge to
In the next section we describe theoretical
Meet Demand
dilemmas in the use of information technology
to leverage tacit knowledge. We then develop However, obtaining the resource is only the
propositions about how such technology might be first of many knowledge management chal
applied to accomplish this and illustrate these lenges. If a firm is fortunate enough to obtain
concepts using the vignettes. Finally, we discuss such assets, those assets will still probably be
implications for using information technology in scarce, even within the firm. For example, ifan
support of resource-based advantages. individual has valuable tacit knowledge that
attracts customers, the demand for his or her
time will exceed
quickly supply (Teece, 1982).
THE KNOWLEDGE-SCALING PARADOX Thus, to leverage
tacit knowledge into a signif
icant advantage, the firmmust be able to "scale
The Problem of Acquiring Tacit Knowledge
up" tomeet demand for the resource.
In order to understand the dilemma of lever This is distinct from the corporate strategy
aging tacit knowledge with information technol literature, which focuses on leveraging knowl
ogy, we begin by reviewing broader problems edge across businesses. That context assumes
associated with tacit knowledge. It is widely that there is excess capacity in strategic re
accepted that knowledge creation and manage sources that cannot be fully utilized within a
ment can be a source of sustained advantage single business unit (Montgomery & Wernerfelt,
(Barney, 1991; Grant,1996). Tacit knowledge is 1988)?a very different problem from the scaling
especially promising because, although itmay dilemma at issue here, in which the knowledge
be closely tied to a valuable capability (e.g., a is in short supply.
cost or differentiation advantage), itmay also be To some extent, the broad problem of scaling
hard for rivals to imitate or acquire. may be generalizable to other firm-specific as
These isolating mechanisms that prevent ri sets. If a firm owns a
firm-specific technology,
vals from obtaining the resource also pose di the extent of any resulting advantage may be
lemmas for firms wishing to acquire resources. constrained by whether the resource can be
Such assets cannot generally be acquired exter scaled up tomeet demand. Itmay or may not be
nally for the same reasons that rivals cannot possible to invest further to increase the scale of
obtain the knowledge in labor markets. First, the core technology without making it available
such tacit knowledge is often firm specific and to rivals. If the technology generates rent, the
must be developed internally?there may be no firm should have financial resources to invest
market for the skills (Dierickx & Cool, 1989). Sec further. However, the firm-specific nature of the
ond, if the expertise is available, it may be resource will limit the role of outside suppliers
abundant and thus might not serve as a source or markets. This, along with the fact that rivals
of advantage. Third, the tacit nature of the cannot easily deploy the idiosyncratic technol
knowledge may make it hard for the firm to ogy, may help to sustain an advantage, even
identify job seekers with the desired knowledge. while the firm increases the scale on which it is
Such labor markets may function like Akerlof's applied.
(1970) markets for lemons, in that those seeking This leveraging problem is especially difficult
jobs will be disproportionately lower-quality for tacit knowledge, because the act of scaling
employees (Chiang & Chiang, 1990). Finally, if up may make the resource imitable and, thus,
the expertise is rare but easily identifiable, the may erode the advantage. Typically, the firm
individuals who own itmay be in a strong po must replicate the tacit knowledge by transfer
sition to extract most of the rent (Coff, 1999). ring it among employees so that the tacit knowl
Much of the literature on the resource-based edge is more widely shared and is therefore
view stops with the acknowledgment that available to meet demand (Tsai, 2001). Further
knowledge may be an important source of com more, knowledge creation requires that different

This content downloaded from 195.78.109.96 on Wed, 18 Jun 2014 15:13:38 PM

All use subject to JSTOR Terms and Conditions
 454 Academy of Management Review April

bodies of knowledge be integrated across the rivals have in importing and imitating the
firm to create new knowledge (Grant, 1996; knowledge from scratch.
Nonaka, 1994). However, knowledge transfer However, this implicitly assumes that all tacit
generally requires some degree of codification knowledge has the same strategic relevance. If
(Zander & Kogut, 1995; Zoilo & Winter, 2002). In the knowledge is embedded in firm-specific ex
deed, the problem of knowledge stickiness (e.g., periences, itwill certainly be much more readily
why it is hard to transfer) is a central challenge available within the firm than to those outside
explored in the knowledge management litera the firm. In fact, this is an important basis for the
ture (Szulanski, 1996). knowledge-based theory of the firm (Kogut &
The knowledge-scaling paradox, then, is that Zander, 1992). Thus, tacitness alone is insuffi
in transferring and codifying knowledge to cient to predict the strategic impact of a knowl
achieve the requisite scale, itmay lose strategic edge base in terms of sustainability.
properties that keep it from rivals (Lippman & Figure 1 illustrates this interaction
between
Rumelt, 1982; Rivkin, 2001). That is, if the knowl tacitness and firm specificity. Quadrant I de
edge is codified, rivals may gain access to itmore picts the situation in which knowledge is highly
easily by hiring away employees or acquiring the tacit and firm specific. Accordingly, neither the
codified knowledge in some other form, such as originating firm nor its rivals can easily parlay
written documents or an expert system (Barney, the knowledge into an overwhelming competi
1991).We are leftwith a somewhat troubling ac tive advantage. This is the outcome that Rivkin
cepted view that knowledge is a promising source (2001) associates with high levels of tacitness.
of competitive advantage, but its promise must be In contrast, Quadrant II illustrates a context in
diminished in order to exploit the resource. Our which knowledge is tacit but broadly applicable.
contribution is to explore boundary conditions sur For example, most readers of this paper would
rounding this assertion. have difficulty absorbing and utilizing knowledge
Rivkin (2001) argues that firms must balance about the physics underlying carbon nanotubes?
the need to replicate tacit knowledge internally microscopic tubular structures made from pure
with the desire to keep the knowledge tacit so carbon (like diamonds) have important mechani
that rivals cannot imitate. At very high levels of cal andelectrical properties that make them
tacitness, firms fail to replicate the knowledge promising for producing composite materials.
internally and cannot realize a significant com However, nuclear physicists inmany different or
petitive advantage. At low levels of tacitness, ganizations have
the requisite background to
rivals are able to imitate the knowledge fairly comprehend this. While the first patent in nano
easily, and any advantage is short-lived. Ac tubes was issued in 1987, by January of 2004, 152
cordingly, he suggests that moderate levels of patents had been issued, 274 patents were pend
tacitness are most promising because they bal ing, and there were 6,026 papers published on the
ance the challenges apparent at each extreme. topic in scientific journals (deHeer, 2004).
Based on this, it is clear that there is a bur
geoning community of researchers who have the
requisite background to absorb new knowledge
Specificity of Tacit Knowledge and the Scaling
on this topic. Despite being ,tacit, for the most
Dilemma
part, this knowledge is not firm specific. In this
The assumption underlying Rivkin's (2001) case, firms imitating may not face the disadvan
case formoderate levels of tacitness is that the tage that Rivkin (2001) associates with highly
firm has better information than rivals about the tacit knowledge. The knowledge may flow more
capability. As such, moderate levels of tacitness easily to rivals who have the requisite absorp
pose more of a barrier for rivals wishing to im tive capacity than to pockets of the originating
itate than for firms wishing to replicate the ca organization that are not grounded in this
pability internally.Rivkin (2001) identifies the knowledge base. Thus, replicators would not
role of hands-on experience in conferring an necessarily enjoy an advantage over imitators
advantage to replicators over imitators. Put an at moderate levels of tacitness. Despite the tac
other way, the originating firm is assumed to itness, such "industry recipes" are unlikely to
have greater absorptive capacity to replicate confer a durable competitive advantage to the
moderately tacit knowledge in a new unit than originating firm (Spender, 1989).

This content downloaded from 195.78.109.96 on Wed, 18 Jun 2014 15:13:38 PM

All use subject to JSTOR Terms and Conditions
2006 Coff, Coff, and Eastvold 455

FIGURE 1
The Interaction of Tacitness and Firm Specificity

Firm specificity

High Low

I. Small durable advantage II. Industry recipe

Potentially durable source of The tacit knowledge requires absorptive

advantage but hard for the capacity to utilize but is readily diffused

originator to replicate in order to to firms that have the requisite

High increase the scale. knowledge base.

Example: Learning curve/experience or Example: Complex knowledge arising from

firm-sponsored training education/professional training

Tacitness

III. Substitution limits durability IV. Common knowledge

Originator can easily increase scale. The knowledge is quickly replicated
Specificity forces rivals to seek internally but also diffused throughout
substitutes (e.g., their own firm the industry. It becomes the "ante"
Low
specific routines). required to compete.

Example: Firm-level standard Example: Standardized, widely available

operating procedures and routines skills/general human capital

In Quadrant IIII the knowledge is firm specific to leverage an individual who has tacit knowl
but not tacit. This means that rivals can compre edge is through hierarchy or specialization of
hend the knowledge but cannot apply it directly labor. For example, researchers leverage their
in their context. However, the fact that they un knowledge by hiring research assistants for
derstand the knowledge spurs them to search mundane tasks. Similarly, doctors and lawyers
for comparable substitutes. In other words, they often tasks to paraprofessionals
delegate
will most likely seek their own firm-specific assistants, secretaries,
(nurses, paralegal etc.).
knowledge that accomplishes the same end. In this case, the knowledge may not be trans
Such knowledge may confer an advantage, but
ferred if the principal provides proper incentives
its duration will be bounded by rivals1 ability to and monitoring. This may leverage individuals
find comparable substitutes.
without codifying,
replicating, or transferring
Finally, Quadrant IV depicts knowledge that
the tacit knowledge. As such, this may be the
can easily be replicated and imitated because it
oldest solution to the knowledge-scaling para
is codifiable and general in nature. Accordingly,
dox. Indeed, this harkens back to Adam Smith's
it should diffuse rapidly throughout the industry
and form no basis for a competitive advan (1776) use of a pin factory to underscore his ob
servation that specialization of labor can en
tage?even one of a temporary nature.

From Figure 1we see that none of the quadrants hance efficiency.
It is also one of the foundations of the knowl
yields a highly promising source of a sustainable
there would edge-based theory of the firm, which seeks to
competitive advantage. Accordingly,
the scope and boundaries of the firm
be tremendous potential value to be realized if a explain
firm could figure out how to scale up highly tacit using the knowledge management process. The
and firm-specific knowledge (quadrant 1). This is existing literature suggests that hierarchies
truly a thorny problem for the strategy literature. emerge because of the need for a common lan
guage or routines
to facilitate knowledge trans
Knowledge Leveraging and the Theory of the fer (Kogut & Zander, 1992; Nelson & Winter, 1982)
Firm or because there is a need to supervise or coor
dinate those with distinct bases
This scaling challenge highlights a core prob knowledge (Ar
lem in organizational theory. Clearly, one way row, 1974; Conner & Prahalad, 1996).

This content downloaded from 195.78.109.96 on Wed, 18 Jun 2014 15:13:38 PM

All use subject to JSTOR Terms and Conditions
 456 Academy of Management Review April

Either way, it is assumed in existing research are specificto storage, whereas expert systems
that tacit knowledge accumulates at the lower are specific to knowledge application.
levels of the hierarchy (Castanias & Helfat, However, most of these applications manipu
1991). This differs from knowledge leveraging, in late codified knowledge so that it can be trans
which in authority positions
those tend to be ferred and used in the firm, or even by
by many
repositories of tacit knowledge and attempts to
multiple firms. Indeed, the manage
knowledge
transfer the knowledge are minimal. Neverthe ment literature focuses on database
system
less, we consider this to be otherwise consistent driven applications designed to store explicit
with the knowledge-based theory of the firm. Consider the knowledge
knowledge. manage
However, as a means to leverage ment
hierarchy, systems used by many large consulting
tacit knowledge, has a number of limitations. firms to codify, store, retrieve, and transfer
First, agency problems arise if it is hard to pro about how to conduct various
knowledge types
vide powerful incentives or monitoring. Anyone of projects. Here, a consultant might document
who has had a research assistant afflicted with the key elements of a complex project proposal
"senioritis" can attest to the potential formoti so that others who must develop similar propos
vational problems. If the tasks are hard tomea
als can retrieve them and borrow from them as
sure or observe directly, the principal may be needed. This type of knowledge management
restricted in his or her ability to delegate. A
system is designed specifically to manage the
second factor is that since there is a limit to how
codifiable portion of the knowledge to be lever
many subordinates an individual can manage
aged. In practice, managers often use such sys
(Barnard, 1938), this method may not achieve the
tems to identify others who can help them with a
desired scale before hitting the point of declin
particular problem; in short, such a system fa
ing marginal returns.
cilitates social network formation to a greater
extent than the storage of tacit knowledge.
Other knowledge management techniques
Limits of Using Information Technology to are even more aggressive in codifying tacit
Leverage Tacit Knowledge knowledge. Expert systems require that an ex
pert be interviewed to translate tacit heuristics
Arguably, the second oldest form of knowl
into a codified set of decision rules. These rules
edge leveraging involves
the application of
then form the outline of the application. Specif
technology. Simple tools may increase an indi
vidual's productivity considerably. In the case of ically, the end user enters the critical parame
ters indicated by the expert. The system then
knowledge management, parchment and a quill
applies the decision rules to generate an anal
offered a breakthrough innovation thousands of
and, a recommendation. Here,
years However, writing
ago. involves codifying ysis ultimately,
there is a conscious attempt to codify tacit
knowledge by definition?that is, putting it into
words. In this way, many solu knowledge in order tomake it available to non
technological
tions carry the same limitation as other knowl experts (Gill, 1996).
the potential These applications may be problematic from
edge management techniques:
value of the knowledge must be diminished a strategic standpoint, since the knowledge be
in order to achieve comes codified in the process of applying the
through codification scale.
This general observation is in the information technology. This means that the
apparent
information technology literature. Alavi and knowledge may become imitable and/or avail
Leidner (2001) describe the role of information able to rival firms (Lado & Zhang, 1998; Mata,
in facilitating creation, Fuerst, & Barney, 1995). That is, in applying the
technology knowledge
transfer, and technology, the knowledge will have lost some
knowledge storage, knowledge
of its value-generating properties, at least in
knowledge application. They identify group
ware, communication, and intranet technologies terms of the sustainability of the advantage. In
as tools that may be useful in facilitating all of fact, there is very little discussion in the litera
these processes by enhancing communication ture about applications of information technol
and coordination in organizations. Other tech ogy to facilitate knowledge management that do
nologies are specific to knowledge management not involve codifying and/or transferring the
processes. For example, knowledge repositories knowledge.

This content downloaded from 195.78.109.96 on Wed, 18 Jun 2014 15:13:38 PM

All use subject to JSTOR Terms and Conditions
 2006 Coff, Coff, and Eastvold 457

However, some information technology appli for a more rapid or efficient transfer of tacit
cations rely relatively less on codification. For knowledge. These
technologies, however, may
example, virtual teams and groupware may also not speed this task so much as reduce the
be tools for facilitating knowledge creation. amount of face-to-face contact required to trans
Such electronic
coordination is valuable but fer knowledge. Indeed, substituting communica
may the nature of the knowledge
alter created tion technology for face-to-face communication
by emphasizing explicit elements of knowledge may cause the knowledge transfer to be more
that are more easily shared in electronic com lengthy and/or inefficient (Alavi & Leidner, 2001).
munication (Griffith, Sawyer, & Neale,
channels This may allow functions to be more geograph
2003). Tacit knowledge may be transferred in ically distributed than would otherwise be the
electronic media, but it often must be codified to case, but the task of knowledge transfer remains
a greater extent. Thus, the technology makes formidable.
explicit knowledge more available and may
shift emphasis away from relatively tacit com
Limiting Codification and Knowledge Transfer
ponents that require face-to-face communica
tion. This, in turn, lessens the focus on develop The discussion above assumes that technol
ing tacit knowledge that has greater strategic ogy may be applied primarily to assist in the
significance.1 Therefore, the following proposi codification and transfer of knowledge; indeed,
tion integrates perspectives from the knowl this has been a major thrust in the information
edge, information technology, and strategy liter technology literature (Alavi & Leidner, 2001;
ature. Lado & Zhang, 1998; Mata et al., 1995). However,

1: Other given that codification may reduce the strategic

Proposition things being value of the knowledge and technology may not
equal, the more technology requires enhance the speed or efficiency of knowledge
knowledge to be codified, the more
transfer, how useful is information technology
imitable the knowledge will become.
as a tool for leveraging knowledge as a strate
This would appear to cast doubt on the effi gic asset?2
cacy of applying information technology to le We believe that information technology can be
verage knowledge into a sustained competitive very valuable in this context but that it requires a
advantage. However, some technologies are not departure from the usual approaches to applying
specifically oriented toward codifying. For ex technology to leverage knowledge. We propose an
ample, telecommunications technology may re alternative that may avoid or limit codification
quire relatively little codification per se. Firms and transfer but still apply technology to leverage
rely heavily on voice and, increasingly, video knowledge. In so doing, thiswill preserve the stra
telecommunications applications that may help tegic properties of the knowledge and generate a
convey tacit knowledge (e.g., through tone or more sustainable advantage.
facial expression). Although these channels are Technology has long been used to enhance
relatively rich compared to text-based media, individual productivity. For example, personal
they
are
clearly not as rich as face-to-face com computers, such as the one used to write this
munication. paper, ease the process of writing. Similarly, we
Although these may not diminish the strategic suggest that it is possible to use technology so
value of the underlying knowledge in the way that individuals can leverage their tacit knowl
that text-based communication does, they have edge without transferring or codifying that
been a focus primarily in the context of knowl knowledge. In short, technology can be used to
edge transfer. Transfer is, by definition, a diffi make individuals more productive, limiting the
cult and time-consuming task for tacit knowl need to replicate, transfer, or codify the knowl
edge (Szulanski, 1996; Teece, 1977). Thus, one edge. Since this retains the strategic properties
might think that an effective tool would allow of the tacit knowledge, the advantage may be
even more sustainable than through traditional

1
Note that we hold constant the degree of protection af
2
forded by other isolation mechanisms, such as intellectual For the moment we set aside the value of establishing
property rights (Teece, 1986). geographically dispersed networks.

This content downloaded from 195.78.109.96 on Wed, 18 Jun 2014 15:13:38 PM

All use subject to JSTOR Terms and Conditions
 458 Academy of Management Review April

applications of information technology to Proposition 2: When technology is

knowledge management. used to leverage tacit knowledge
One might ask what type of technology might without codification, firm-specific re
leverage those with tacit knowledge while min sources (both technology and knowl
imizing the codification required. We begin with edge) are created that may enhance
an example of tacit knowledge in the form of sustainability of an advantage.
pattern recognition, where an expert can ob
serve a set of stimuli, discern patterns, and Although increased firm specificity may make
translate them into recommendations (Simon, an advantage more durable, itmay also tend to
1987). While this type of knowledge enhance the power of those with essential tacit
might be
amenable to expert systems, were it codified knowledge, since their knowledge would be
to a set of decision come even more valuable. These individuals
and reduced rules, as dis
cussed, that might reduce the strategic proper would become even more critical to rent produc
ties of the knowledge. tion once the technological intervention was in
However, technology
still be used to organize and track infor place. This is an important contrast to methods
might
mation. This would leave the key task of spot of leveraging requiring the codification or trans

trends in the information uncodified. How fer of knowledge and would therefore reduce the
ting
ever, the information that a decision maker power and influence of those who possessed the
use could be collected tacit knowledge.
would and displayed in
an interface that facilitated In Indeed, Osterloh and Frey (2000) note the im
interpretation.
portance of cooperation and motivation in the
this way, a decision maker could become more
knowledge transfer process. Furthermore, resis
productive without codifying or transferring the
tance to change is often the most significant
most critical knowledge?how to interpret pat
obstacle when implementing new technologies
terns in the data.
(Heracleous & Barrett, 2001). Thus, choosing
Such a system likely to increase
is firm
technologies that minimize codification might
specific investments a
in number of ways. First,
ease implementation considerably.
the technology itself may be idiosyncratic, since
This increase in power may encourage indi
it is customized to the knowledge management vidual experts to cooperate relative to alterna
task at hand. Second, over time, the system will
tive knowledge management methods. How
become embedded in the firm's routines, since it
ever, it may also grant central individuals
collects and organizes data that are specific to
considerable bargaining power (Coff, 1999). That
decisions requiring tacit knowledge (Birkin is, employees might seek compensation for the
shaw, Nobel, & Ridderstrale, 2002; Nelson & Win risk they would be assuming in investing in
ter, 1982). For example, as described earlier, narrow firm-specific knowledge that has limited
some consulting firms have designed and im value to other firms. Such knowledge normally
plemented idiosyncratic knowledge manage forms the basis for a bilateral monopoly (Wil
ment systems. These systems are then inte
liamson, 1975). In other words, the individual
grated into such routines as the proposal has power through his or her ability to limit or
development process, which may involve scan stop rent production. At the same time, he or she
ning the system for similar projects and consult to offer a credible
may not be able threat to
ing project managers across the firm for advice cease if the knowledge is firm spe
production
(e.g., access to tacit knowledge). Such systems cific and, therefore, is not valuable to other firms
also require individuals to routinely contribute (Becker, 1964). Using information technology to
information about projects to make it available an individual without codification
leverage
to others in the firm. Accordingly, individuals would tend to retain this basic structure but
invest in firm-specific knowledge to use the sys would increase the stakes. Thus, one would ex
tem effectively. This, in turn,may enhance stra pect the firm and individual to split the gains
tegic properties of the knowledge, since firm achieved through the application of information
specificity may act as a barrier to imitation (Pe technology.
teraf, 1993). Thus, the knowledge may actually But what split might be obtained? To the ex
have stronger strategic properties after the ap tent that the technological change results in
plication of technology than before. more firm-specific knowledge, itmay reduce the

This content downloaded from 195.78.109.96 on Wed, 18 Jun 2014 15:13:38 PM

All use subject to JSTOR Terms and Conditions
 2006 Coff, Coff, and Eastvold 459

external for the experts while

demand increas management. refers to generating
This the max
ing their value internally. This may hinder their imum output from each silicon wafer (e.g., min
ability to offer a credible threat of exit, thereby imizing waste). Among the most significant driv
reducing the portion of the rent that the experts ers behind the demand for yield management
can appropriate. While these experts might seek solutions are increasing device complexity, re
compensation the
for this risk before making duced product life cycles, and increased compe
investment, they would lose bargaining power tition. To meet market needs, the semiconductor
once the initial investment was made. If so,
industry is constantly increasing the functions
while initial pay may be high, pay increases incorporated into a single device, while simul
might not keep pace. Accordingly, the longer the device size.
taneously reducing
experts' tenure, the less rent they would be able These challenges are further complicated by
to appropriate.
competition within the semiconductor industry,
which drives an expectation for declining
Proposition 3: As technology-lever
individuals increase the firm prices, despite increasing product complexity.
aged
Thus, chip manufacturers simultaneously face
specific component of their knowl
continuing pressures for profitability, shorter
edge over time, they lose power to
rent. product cycles, and increasing manufacturing
appropriate
complexities. As chip prices drop, cost effective
ness becomes the key to superior profitability.
VIGNETTES ON USING INFORMATION Chip manufacturers increasingly turn to a few
TECHNOLOGY TO LEVERAGE TACIT firms that provide yield management equip
KNOWLEDGE ment and expertise to enhance efficiency. One
of these, KLA-Tencor, offers products, software,
The propositions above may at first seem sim
analysis, services, and expertise to help manu
ple and yet unrealistic. The information technol
facturers manage yield throughout the wafer
ogy literature has focused so heavily on the cod
ification and transfer of knowledge that the fabrication process?from R&D to production.
notion of avoiding these elements seems quite Scarce tacit manufacturing knowledge.
is the primary asset lever
foreign. In this section, however, we offer two Knowledge being
brief vignettes that illustrate how this might aged, for a variety of reasons. First, given the
work in practice. We do not use these examples rapid pace of innovation in manufacturing
to test propositions so much as to suggest that equipment and processes, yield management
must be well versed in the
they are not unrealistic and should be tested. support professionals
The first vignette is drawn from over twenty most current chip manufacturing technologies.
unstructured interviews with KLA-Tencor Corpo Second, the most important knowledge tends to
ration managers charged with implementing a be tacit. By the time defects are readily apparent
system that allowed experts in manufacturing in a plant's output, the process will have already

yield management to leverage their scarce entered a "deathspiral," requiring the facility to
knowledge across many locations. The inter be shut down for repairs. Accordingly, the cen
views took place throughout the planning and tral task is to identify problems from very subtle
implementation phases of the project. The sec signs that may be amplified over time absent
ond example is drawn from the literature on managerial intervention. require con
Experts
physicians' use of information technology to di siderable hands-on
experience to develop this
agnose and treat patients remotely. In both tacit knowledge; the early signs of a chip man
cases information technology was used to lever ufacturing process gone awry cannot be fully
age individuals in very knowledge-intensive codified and, therefore, are hard to teach others.
tasks. This is further complicated by the fact that
customer factories are miles or even continents
apart and are equipped with a multitude of
Leveraging Yield Management Expertise in
highly complex tools. Historically, firms in this
Silicon Wafer Manufacturing
industry have stationed engineers at customers'
Several factors inherent in the semiconductor plants so they can manage performance using
industry drive the need for comprehensive yield their proprietary monitoring "tools." These tools

This content downloaded from 195.78.109.96 on Wed, 18 Jun 2014 15:13:38 PM

All use subject to JSTOR Terms and Conditions
 460 Academy of Management Review April

are a
combination of hardware and software and no longer need to travel to remote locations,
used tomeasure the manufacturing process. the experts have specialized more and devel
As the number of tools has increased (corre oped additional tacit knowledge. Accordingly,
sponding tomanufacturing complexity), the task the net effect of applying the technology has
ofmaintaining them has become harder. On-site actually been to increase the firm's reliance on
engineers can manage the routine situations firm-specific tacit knowledge.
that arise. However, on-site support profession Implications for competitive advantage. KLA
als typically lack the deep knowledge to resolve Tencor is the world leader in the yield manage
complex or rare problems. In these cases, ment market, with a 41 percent market share
deeper, more specialized expertise is needed? (Pitzer, Kumar, & Chin, 2004). Its next closest
particularly to identify problems before they are competitor has only about a 12 percent market
noticeable in the firm's output. Yet this type of share. Moreover, in the years 2000-2003, KLA
expertise is expensive and is not necessarily Tencor was the only one among its most direct
needed for the routine problems at each plant peers that did not report losses in at least one
location. Furthermore, it is difficult to transfer or year. Accordingly, the firm exhibits the type of
replicate the tacit knowledge across such geo performance one might attribute to a resource
graphically dispersed sites. Accordingly, when based competitive advantage.
these more complex problems have emerged, The system described here creates some very
the firm has had to fly in experts from its head critical resources and capabilities embodied in
quarters in Silicon Valley. the specialized engineers that support the com
While this is a more cost-effective way to pany's advantage. First, the technology allows
maintain the depth of tacit knowledge and ex experts to be colocated and minimizes required
pertise, it complicates support in other ways. travel. This, in turn, facilitates knowledge cre
First, customers have to wait while experts are ation (Hedlund, 1994; Kogut & Zander, 1992;
in transit. Second, other customers, who may Nonaka, 1994), which produces the tangible re
have a similarly urgent need for expertise, must sult of enhanced product development on an
wait until
the expert is available to visit their ongoing basis. Thus, the use of information tech
site. Third, unexpected trips take specialists nology to leverage tacit knowledge has been
away from product development, causing a rip quite successful here.
ple effect on the firm's ability to introduce new This has strategic significance because the
products. Finally, travel makes the specialists' capabilities created meet the criteria set forth in
job harder and less attractive, exacerbating the the resource-based view: valuable, rare, and un
already difficult task of attracting qualified peo available to rivals (Barney, 1991).We begin with
ple. the question of value. As suggested, technology
Application of information technology. KLA enabled experts can solve customers' produc
Tencor has addressed this problem by having tion problems more effectively and cheaply. By
highly specialized engineers manage the tools developing deeper, more tacit knowledge, ex
from a central location. They monitor all of the perts are able to identify and correct problems
tools remotely to identify patterns associated before they lead to costly shutdowns or repairs.
with reduced process integrity. These experts Arguably, this can be the biggest source of
are often able to diagnose and resolve problems waste, so this knowledge development results in
without even notifying on-site engineers. In this significant savings. KLA-Tencor is now using
way, the experts can respond immediately and this customer support solution as a way to dif
can prevent inappropriate repair actions, lost ferentiate itself from rivals. This system is the
time from travel, mistaken diagnoses, and, ulti first in the semiconductor equipment industry to
mately, the down time required for major re combine all of these technological components
pairs. for the purpose of leveraging tacit knowledge.
This system involves minimal codification of In addition, the system actually generates
knowledge. Rather, the instrument readouts are rare resources via hyperspecialization. Through
made available for remote viewing and control. this application of information technology, KLA
As such, experts interface much as they would Tencor has been able to develop hyperspecial
were they using the tools tomonitor the process ized expertise on how its own idiosyncratic tools
on site. However, because they are colocated operate. That is, individual experts have deep

This content downloaded from 195.78.109.96 on Wed, 18 Jun 2014 15:13:38 PM

All use subject to JSTOR Terms and Conditions
 2006 Coff, Coff, and Eastvold 461

knowledge of a small subset of the firm's port one source of KLA-Tencor's stable financial per
folio of tools. Therefore, the company is assured formance relative to its direct rivals.
that the key resource?engineers with deep and
narrow scarce,
expertise?remains particularly
Leveraging Physician Expertise with
outside the firm.
Information Technology
Finally, it is especially
noteworthy that these
resources are inimitable and unavailable to ri Information technology is increasingly ap
vals. Not only has the intervention increased the plied to the field of medicine in a variety of
tacitness of the knowledge, it has also made it ways, ranging from robot-assisted surgery (Ger
more firm specific. Thus, as suggested in Prop hardus, 2003) to remotely delivered psychiatry
osition 2, it is possible to increase firm specific (Vought, Grigsby, Adams, & Shevitz, 2000). All
ity if the technology applied minimizes codifica are attempts to leverage scarce medical knowl
tion. In this way, both the high levels of edge to increase the quality of medical care.
tacitness and the firm specificity act as isolating Accordingly, the second vignette is the applica
mechanisms?illustrating Quadrant I of Figure tion of telecommunications technology to lever
1. KLA-Tencor's rivals (e.g., yield management age expertise from a Lubbock hospital to an
consulting and semiconductor manufacturing other located in rural West Texas. This vignette
equipment firms) are at a disadvantage in that draws on a detailed published case study by
KLA's employees have knowledge of the firm's Sykes and Mclntosh (1999).
idiosyncratic tools that will not easily transfer. Scarce physician knowledge in West Texas.
Furthermore, each expert has knowledge of a The Texas Tech Health Sciences Center (TTHSC)
small subset in the firm's portfolio
of the tools is charged with providing medical support and
and cannot convey enough knowledge for rivals services for rural West Texas. A key challenge is
to fully imitate its systems and tools. Accord the scarcity of professionals who wish to live in
ingly, rivals have relatively little to gain by lur and serve these rural areas. To answer this chal
ing away KLA employees because their knowl lenge, the TTHSC was awarded a grant in 1990
edge is so specific to KLA's proprietary tools and to study the potential of telecommunications for
technologies. rural health care. The project, known as MED
Appropriability is a final issue in assessing NET, used satellite and telephone lines to estab
the extent of advantage (Coff, 1999). In this case, lish links with several rural West Texas hospi
the key resources are the information technol tals. Through this system, doctors in these areas
ogy and those with deep expertise. As stated gained access to continuing education and in
earlier, the experts have a bilateral monopoly teractive medical consultations with specialists.
owing to their firm-specific knowledge (Becker, The hope was that providing this link would
1964; Williamson, 1975). That is, their skills are increase the viability of rural hospitals in sev
extremely valuable, and they conceivably have eral ways: (1) communication with experts at the
the ability to hold up others to appropriate rent. TTHSC would reduce the isolation felt by rural
However, at the same time, their skills are firm physicians and would encourage more to re
specific and may not be as valuable to rivals. main in the area; (2) the continuing education
Interestingly, this application has increased the units offered through TTHSC would help profes
firm-specific component of employees' knowl sionals retain their certification without travel
edge, potentially reducing demand for them, ing, important both for the convenience of the
which would tend to reduce employee bargain physicians and because, like the KLA-Tencor
ing power. setting, the professionals were unavailable
Interestingly, in this case, engineers have while in transit; and (3) the access to specialists
viewed the ability to limit their travel positively would enhance health care and improve patient
and have not sought additional compensation to retention.

encourage the investment in firm-specific hu Although the nearest hospital was often hun
man capital. This illustrates how nonfinancial dreds of miles away, patients were willing to
benefits can substitute for pay (Coff, 1997). drive if they lacked confidence in their local
Therefore, it appears that employees have not facility and emergency care was not required.
appropriated a large portion of the rent (consis MEDNET would allow more complex or serious
tentwith Proposition 3). Indeed, this may well be patients to be treated locally, where health care

This content downloaded from 195.78.109.96 on Wed, 18 Jun 2014 15:13:38 PM

All use subject to JSTOR Terms and Conditions
 462 Academy of Management Review April

costs were lower. It was estimated that this management team. Furthermore, the buyer did
would save more than $1,000 per consult over the not implement the MEDNET system in any of its
costs to transport the patient and the higher fees other facilities as a best practice to be emulated.
in the city. There were a number of important differences
Sykes and Mclntosh (1999) focus on the Big that may have contributed to this outcome. First,
Bend Regional Medical Center, one rural hospi the knowledge, while tacit, was not firm spe
tal in the network. This hospital was under fi cific. There were many specialists in different
nancial pressure, and it was hoped that MED firms/hospitals who had comparable expertise.
NET would strengthen its reputation and Accordingly, this example illustrates Quadrant
differentiate it.Certainly, the satellite link gave II of Figure 1.While the new technology did not
the facility access to more specialized medical reduce tacit knowledge, it did not increase firm
than other within the 200
expertise any facility specificity. Patients retained the ability to drive
mile radius. Since this was a pilot study, the the 200 miles and see a comparable at
specialist
technology was not available to rival facilities. a different facility. This distance may have been
However, there was a possibility that itwould an inconvenience, but rival clinics remained a
eventually be rolled out more broadly should it factor in all nonemergency cases where patients
be successful. could afford the time to choose. As a result, the
The system received publicity in 1990, when a was valuable and somewhat scarce
knowledge
doctor was able to save a baby in respiratory but remained available to rivals. Importantly,
distress by videoconferencing with a neonatal this was the perception of patients and the com
specialist 300 miles away, in Lubbock, Texas as a whole. Thus, while the technology
munity
(Stanley, 1991). The two-way video link allowed worked as planned, it could not serve as a basis
the specialist to "examine" the patient remotely for differentiating the no
facility and offered
and to view X-rays, along with other lab tests.
competitive advantage.
The specialist concluded that the baby had in From the standpoint of TTHSC, while the tech
haled waste in the amniotic fluid and pre worked its economic value
nology very well,
scribed a course of treatment that restored the was also questionable. It failed to divert very
baby's health. News of the event was amplified to the technology-enabled rural
many patients
when technology providers, like Southwestern and resulted in relatively little sav
hospitals
Bell Telephone, used the case in advertisements was to save
ings. The system used lives in the
to show how telecommunications was trans context of medical emergencies and trauma,
forming medicine. where travel to a larger facility was not possi
Implications for competitive advantage. Like ble. when still
However, feasible, patients pre
the KLA example, the objective of MEDNET was
ferred to drive to see a specialist
the extra miles
not generally to transfer tacit knowledge but to were not
directly. In the end, TTHSC specialists
allow experts to work remotely. The technology
better utilized after the technology was imple
involved little, ifany, codification of knowledge.
mented.
Finally, as the "miracle baby" example illus
trates, the information technology was quite ef
fective overall. Indeed, in addition to saving
money, relative to the alternatives, the system DISCUSSION AND CONCLUSION
allowed doctors to save a number of lives.
Paradox Lost
However, the overall result for the organiza
tions involved was very different. The financial These brief vignettes were not offered as tests
difficulties continued at the Big Bend Regional of the propositions but, rather, to illustrate the
Medical Center, suggesting that, if any advan efficacy of further exploration along these lines.
tage was conferred, it did not have overwhelm Much of the work on the resource-based view
ing economic value. The hospital continued to extols the potential benefits of leveraging
experience a shortage of professional staff, so knowledge-based assets since they are hard for
recruitment and retention were not strongly in rivals to imitate or acquire (Hall, 1993; McEvily &
fluenced. Ultimately, Big Bend Regional Medical Chakravarthy, 2002). Here, the tacitness of the
Center failed, and the assets were taken over by knowledge contributes greatly to the sustain
a large hospital chain, which fired the existing ability of a competitive advantage. However,

This content downloaded from 195.78.109.96 on Wed, 18 Jun 2014 15:13:38 PM

All use subject to JSTOR Terms and Conditions
 2006 Coff, Coff, and Eastvold 463

relatively little of the research has addressed Implications for Further Research
how firms can achieve the desired scale with
Future research should explore the contexts in
this type of resource.
which such applications of information technol
Furthermore, while the knowledge literature
ogy are most likely to add value. What are the
has focused on how firms might leverage such
boundary conditions under which this contrib
resources, the bulk of this literature presumes utes to a competitive Our
advantage? vignettes
that knowledge management requires codifica have described unique situations in which in
tion. It would appear that this contradicts re to
formation technology has been used leverage
source-based assumptions about why tacit tacit knowledge. Unlike many applications, the
knowledge is such a valuable resource, but this in both cases involved rela
technology applied
paradox has rarely been explored directly tively little codification. As such, the core knowl
(Rivkin, 2001, is an important exception). edge remained tacit and embodied in the ex
Much of the information technology literature perts. In one context, it even allowed for greater
serves to exacerbate this paradox. Here, the specialization, which should have hindered im
need for codification is perceived to be even itation even more than before the information
stronger, since information technology can be technology was applied.
an especially powerful tool for manipulating However, our vignettes suggest that the locus
and managing codified knowledge (Alavi & of the advantage continues to reside with the
Leidner, 2001). Although this is consistent with knowledge and not with the information tech
the knowledge management literature, it could nology. In one case the technology appeared to
lead to the erosion of the value of knowledge create an advantage, whereas in the other it did
based resources from a strategic standpoint. not. The information technology can serve as an
enabler or amplifier of strategic assets but is
Interestingly, counter to trends in the litera
ture, information technology can be deployed in not, in itself, a source of competitive advantage.
a way that creates even more tacit knowledge, if That is, if the knowledge is valuable, rare, and
the application does not require codification or unavailable to rivals, technology might be used
to achieve more scale without adversely affect
knowledge transfer and increases the degree of
In this way, information technol ing its strategic properties. Accordingly, there is
specialization.
much we can learn from these settings about
ogy can allow firms to leverage knowledge
how to develop technological solutions to
while enhancing its strategic properties?the
knowledge management challenges that en
key knowledge management challenge.
hance strategic position.
This runs counter to Rivkin's (2001) sugges
That said, we have not explored the limita
tion that moderate levels of knowledge tacit
tions on how much scale can be achieved
ness and complexity are most promising for a
through this type of technology. While there are
knowledge-based advantage. He suggests
generally declining returns to specialization, it
that, in this range, knowledge management would appear that
the technology allows for
dilemmas (transfer and codification) are bal
greater specialization. However, it is not clear
anced against the need to minimize knowl
that this will be sufficient to achieve the desired
edge spillovers that would allow rivals to im
scale in all settings. Rather, it adjusts the point
itate more easily. at which declining returns to specialization are
In contrast, we suggest that it is possible to reached. Further research is needed to identify
leverage relatively high levels of tacit knowl the bounds on when this type of solution is most
edge into a competitive advantage. First, a high
likely to add value.
of tacitness can be coupled with firm
degree Thus far, our discussion has ignored the
specificity so that the risk of spillovers is quite tacit knowledge and other intangible assets
limited. Furthermore, the dilemmas of replica that may be associated directly or indirectly
tion and transfer can be avoided if the technol with information
implementing technology.
ogy allows individuals to utilize their tacit Indeed, of information technol
applications
knowledge more effectively. In short, individu ogy can be a source of advantage when linked
als become more efficient, reducing the need to to knowledge and expertise (Bharadwaj, 2000).
steep others in the knowledge base. Put another way, firms often fail to implement

This content downloaded from 195.78.109.96 on Wed, 18 Jun 2014 15:13:38 PM

All use subject to JSTOR Terms and Conditions
 464 Academy of Management Review April

technology effectively, and the ability to do so knowledge management systems: Conceptual founda
tions and research issues. MIS Quarterly, 25: 107-136.
may confer an advantage if it is sufficiently
valuable and rare. If the technology requires Arrow, K. 1974. The limits of organization. New York: Norton.

and creates additional tacit knowledge, there Barnard, C. 1938. The functions of the executive. Cambridge,
may be some component of an advantage MA: Harvard University Press.

more closely tied to the technology itself. Fur Barney, J.B. 1991. Firm resources and sustained competitive
ther research should explore whether this advantage. Journal of Management, 17: 99-120.

might extend competitive advantage even Becker, G. S. 1964. Human capital. New York: Columbia Uni
more. versity Press.

Bharadwaj, A. S. 2000. A resource-based perspective on in

formation technology capability and firm performance:
An empirical investigation. MIS Quarterly, 24: 169-196.
Conclusion
Birkinshaw, J.,Nobel, R., & Ridderstrale, J. 2002. Knowledge
We have by integrating the strat
contributed as a contingency variable: Do the characteristics of

egy and knowledge management literature to knowledge predict organization structure? Organization
understand how firms can effectively leverage
Science, 13: 274-289.

knowledge into a resource-based advantage. Castanias, R. P., & Helfat, C. E. 1991. Managerial resources

Additional studies should focus more carefully and rents. Journal of Management, 17: 155-171.

on how firms can leverage knowledge effec Chiang, S.-H., & Chiang, S.-C. 1990. General human capital
itsmost valuable strate as a shared investment
tively while preserving
under asymmetric information.
Canadian Journal of Economics, 23: 175-189.
gic properties. Of course, we call for further re
search that directly tests our propositions. These Coff, R. W. 1997. Human assets and management dilemmas:
with hazards on the road to resource-based the
strike at the very core of research on competitive Coping
ory. Academy of Management Review, 22: 374-402.
advantage?but have nevertheless been under
researched. Coff, R. W. 1999. When competitive advantage doesn't lead to

performance: The resource-based view and stakeholder

Indeed, our propositions represent a very dif
bargaining power. Organization Science, 10: 119-133.
ferent research agenda for the information tech
Conner, K. R., & Prahalad, C. K. 1996. A resource-based the
nology literature, since they suggest that tech
versus
ory of the firm: Knowledge opportunism. Organi
nology might be applied to leverage knowledge zation Science, 7: 477-501.
without codification. We have focused on the
deHeer, W. 2004. Nanotubes and the pursuit of applications.
need to leverage expertise across geographi 29: 281-286.
MRS Bulletin,
cally dispersed units. This opportunity may be
Dierickx, I., & Cool, K. 1989. Asset stock accumulation and
most promising for the types of telecommunica of competitive
sustainability advantage. Management
tions solutions that can convey tacit knowledge Science, 35: 1504-1511.
across great distances. The more traditional da
Gerhardus, D. 2003. Robot-assisted surgery: The future is
tabase-driven knowledge management systems here. Journal of Healthcare Management, 48: 242-251.

appear to require more codification than this

Gill, T. G. 1996. Expert systems usage: Task change and
and may ultimately be of somewhat less strate intrinsic motivation. MIS Quarterly, 20: 301-329.

gic value. We anticipate that, if such applica R. M. 1996. Toward a knowledge-based

Grant, theory of the
tions prove to be viable, this will greatly en firm. Sfrafegic Journal, 17: 109-122.
Management
hance the strategic role that technology can
Griffith, T. L., Sawyer, J.E., & Neale, M. A. 2003. Virtualness
play. In sum, this close alignment of the strat and in teams: Managing the love triangle of
knowledge
egy, knowledge, and information technology lit organizations, individuals, and information technology.
erature carries some very strong potential ben MIS Quarterly, 27: 265-287.

efits and offers fertile terrain for further Hall, R. 1993. A framework linking intangible resources and
to sustainable competitive Stra
exploration. capabilities advantage.
tegic Management Journal, 14: 607-618.

Hedlund, G. 1994. A model of knowledge management and

the N-form Journal,
REFERENCES corporation. Sfrafegic Management
15: 73-91.
Akerlof, G. A. 1970. The market for "lemons": Qualitative
Heracleous, L., & Barrett, M. 2001. Organizational change as
uncertainty and the market mechanism. Quarterly Jour discourse: Communicative actions and deep structures
nal of Economics, 84: 488-500.
in the context of information technology implementa
Alavi, M., & Leidner, D. E. 2001. Knowledge management and tion. Academy of Management Journal, 44: 755-778.

This content downloaded from 195.78.109.96 on Wed, 18 Jun 2014 15:13:38 PM

All use subject to JSTOR Terms and Conditions
 2006 Coff,Coff,and Eastvold 465

Kogut, B., & Zander, U. 1992. Knowledge of the firm, combi of intuition and emotion. Academy of Management Ex
native capabilities, and the replication of technology. ecutive, 1(1): 57-65.
Organization Science, 3: 383-397.
Smith, A. 1776. The wealth of nations. Middlesex, UK: Pen
Lado, A. A., & Zhang, M. J. 1998. Expert systems, knowledge guin Books.

development and utilization, and sustained competitive

Spender, J. C. 1989. Industry recipes: An enquiry into the
advantage: A resource-based model. Journal ofManage
nature and sources of managerial judgement. Oxford:
ment, 24: 489-509.
Blackwell.

Lippman, S. A., & Rumelt, R. P. 1982. Uncertain imitability: An

Stanley, D. 1991. Network links rural doctors to technology.
analysis of interfirm differences in efficiency under
Austin American-Statesman, October 6: Bl.
competition. Bell Journal of Economics, 13: 418-438.
Sykes, D.f & Mclntosh, W. 1999. Telemedicine, hospital via
Mata, F. J., Fuerst, W. L., & Barney, J. B. 1995. Information
and embeddedness: A case
bility, community study.
technology and sustained competitive advantage: A re
Journal of Healthcare Management, 44: 59-71.
source-based analysis. MIS Quarterly, 19: 487-505.
Szulanski, G.
1996. Exploring internal stickiness: Impedi
McEvily, S., & Chakravarthy, B. 2002. The persistence of
ments to the transfer of best practice within the firm.
knowledge-based advantage: An empirical test for
Sfrafegic Management Journal, 17: 27-43.
product performance and technological knowledge.
Teece, D. 1982. Toward an economic theory of the multiprod
Sfrafegic Management Journal, 23: 285-305.
uct firm. Journal of Economic Behavior and Organiza
Montgomery, C. A., & Wernerfeit, B. 1988. Diversification,
tion, 3: 39-63.
Ricardian rents, and Tobin's q. Rand Journal of Econom
ics, 19: 623-632. Teece, D.
J. 1977. Technology transfer by multinational firms:
The resource cost of transferring technological know
Nelson, R. R., & Winter, S. G. 1982. An evolutionary theory of how. Economic Journal, 87: 242-261.
economic change. Cambridge, MA: Belknap Press of
Harvard Press. Teece, D. J. 1986. Profiting from technological innovation:
University
Implications for integration, collaboration, licensing,
Nonaka, I. 1994. A dynamic theory of organizational knowl
and public policy, flesearch Policy, 15: 285-305.
edge creation. Organization Science, 5: 14-37.
Tsai, W. 2001. Knowledge transfer in intraorganizational net
Osterloh, M., & Frey, B. S. 2000. Motivation, knowledge trans
works: Effects of network position and absorptive capac
fer, and organizational forms. Organizafion Science, 11:
ity on business unit innovation and performance. Acad
538-550.
emy of Management Journal, 44: 996-1005.
Peteraf, M. A. 1993. The cornerstones of competitive advan
Vought, R. G., Grigsby, R. K., Adams, L. N., & Shevitz, S. A.
tage: A resource-based view. Sfrafegic Management 2000. Telepsychiatry: Addressing mental health needs in
Journal, 14: 179-191.
Georgia. Community Mental Health Journal, 36: 525-536.
Pitzer, J.,Kumar, S., & Chin, B. 2004. Company updafe: KLA Williamson, O. E. 1975. Markets and hierarchies: Analysis
Tencor Corp. New York: Credit Suisse First Boston.
and antitrust implications. New York: Free Press.

Polanyi, M. 1966. The tacit dimension. New York: Anchor Day Zander, U., & Kogut, B. 1995. Knowledge and the speed of the
Books.
transfer and imitation of organizational capabilities?
Rivkin, W.
J. 2001. Reproducing knowledge: Replication with an empirical test. Organization Science, 6: 76-92.
out imitation at moderate complexity. Organizafion Sci
Zoilo, M., & Winter, S. 2002. Deliberate learning and the
ence, 12: 274-293.
evolution of dynamic capabilities. Organization Sci
Simon, H. A. 1987. Making management decisions: The role ence, 13: 339-354.

Russell W. Coff (www.bus.emory.edu/RCoff/) is an associate professor of organization

and management at Emory University's Goizueta Business School. He received his
Ph.D. from UCLA. His research interests include the management of human assets,
competitive advantage, and value capture from strategic capabilities.

David C. Coff is a client business manager at Xtelesis, where he helps clients improve
network efficiency. Previously, he worked at AT&T for twenty-six years, where he
served as a senior data/IP sales executive and IP specialist.

Roger Eastvold is the president of e-Diagnostic Network, where he helps clients apply

technology to improve service operations. Previously, he served as vice president of

technology at Brooks Automation and at KLA-Tencor.

This content downloaded from 195.78.109.96 on Wed, 18 Jun 2014 15:13:38 PM

All use subject to JSTOR Terms and Conditions