Knowledge

Management and
Organizational
Design
This page intentionally left blank
Icnowledge
Management and
Organizational
Design
Edited by Paul S. Myers

Butterworth-Heinemann
Boston Oxford Johannesburg Melbourne New Delhi Singapore
 Copyright 0 1996 by Butterworth-Heinemann
-@ A member of the Reed Elsevier group
All rights reserved.

No part of this publication may be reproduced, stored in a retrieval system, or trans-

mitted in any form or by any means, electronic, mechanical, photocopying, record-
ing, or otherwise, without the prior written permission of the publisher.
Recognizing the importance of preserving what has been written, Butterworth-
@ Heinemann prints its books on acid-free paper whenever possible.
Library of Congress Cataloging-in-Publication Data
Knowledge management and organizational design / edited by Paul S.
Myers.
p. cm.-(Resources for the knowledge-based economy)
Includes bibliographical references and index.
ISBN 0-7506-9749-0 (pbk.)
1 . Organizational learning. 2. Organizational effectiveness.
I. Myers, Paul S., 1964- . 11. Series.
HD58.82.K66 1996
658.4’034~20 96-18403
CIP

British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the British Library.

The publisher offers discounts on bulk orders of this book.

For information, please contact:
Manager of Special Sales
Butterworth-Heinemann
313 Washington Street
Newton, MA 02158-1626
Tel: 617-928-2500
Fax: 617-928-2620

For information on all Business publications available, contact our World Wide
Web home page at: http:/lwww.bh.comlbh/bb

10 9 8 7 6 5 4 3

Printed in the United States of America

 Table of Contents

Acknowledgments vii

lntroduction to Series-Why Knowledge, Why Now? ix

1 Knowledge Management and Organizational Design:

An Introduction 1
Paul S . Myers

2 The Use of Knowledge in Society 7

Frederick A. Hayek

3 Specific and General Knowledge, and Organizational

Structure 17
Michael C. Jensen & William H . Meckling

4 The Rise and Fall of Bureaucracy 39

Gifford 6 Elizabeth Pinchot

5 The Emerging Flexible Organization:

Perspectives from Silicon Valley 55
H o m a Bahrami

6 The Organization of Innovation 77

Tom Burns 6 G.M. Stalker

V
vi Table of Contents

7 When a Thousand Flowers Bloom: Structural, Collective,

and Social Conditions for Innovation in
Organizations 93
Rosabeth Moss Kanter

8 Knowledge Links 133

Joseph Badaracco

9 Strategic Alliances, Organizational Learning, and

Competitive Advantage: The HRM Agenda 151
Vladimir Pucik

10 TheNumbers 167
Charles Handy

11 Motivating Knowledge Workers-The Challenge

for the 1990s 179
Mahen Tampoe

12 The Social System at the Shop Level 191

Michel Crozier

13 The Abstraction of Industrial Work 197

Shoshana Zuboff

14 Building Intelligence Networks 209

Wayne Baker
 Acknowledgements

The Ernst & Young Center For Business Innovation in Boston supported my work
on this anthology, and I thank Larry Prusak for inviting me to participate in this
project. Terrence Boyle provided valuable assistance in securing permission to re-
print the thirteen selections. I thank the authors and copyright holders for their
willingness to share their ideas with a diverse audience of practitioners, managers,
and academics.

vii
This page intentionally left blank
 Introduction to Series-
Why Knowledge, Why Now?

Why is there such an upsurge of interest in Knowledge? In 1996 there will be a t

least six major conferences on the subject; there are plans to add three new jour-
nals focusing on Knowledge, sometimes loosely called Intellectual Capital or Or-
ganizational Learning), and many major firms, in the United States and (more
slowly) Europe, are adding positions such as Chief Knowledge Officer, or Organ-
izational Learning, and even a few Vice Presidents for Intellectual Capital!
How come all this focus on a subject that, at some levels, has been around
since the pre-Socratic philosophers? Is it yet another one of the multitudinous
management enthusiasms that seem to come and go, with the frequency of some
random natural phenomena? We don’t think so! Many of us doing research on
this subject have seen the rise and fall of many of these varied nostrums-all of
which attempted to offer to firms a new road to achieving a sustainable competi-
tive advantage. However, when much of the shouting dies down, we have con-
cluded that, excluding monopolistic policies and other market irregularities, there
is no sustainable advantage other than what a firm knows, how it can utilize what
it knows, and how fast it can learn something new!
However, this still does not answer the question, why Knowledge, why now?
Let us try to list some very broad trends that seem to be playing a significant role
in the playing current in Knowledge.

A ) The globalization of the economy which is putting terrific pressure on

firms for increased adaptability, innovation, and process speed.
B) The awareness of the value of specialized Knowledge, as embedded in or-
ganizational processes and routines, in coping with above pressures
C ) The awareness of Knowledge as a distinct factor of production, and its
role in the growing book to market ratios within Knowledge based industries.
D) Cheap networked computing which is at last giving us a tool to work
and learn with each other.

ix
X Introductioii to Series- Why Knowledge, Why Now?

While many can argue for and against these trends we feel that the prepon-
derance of the evidence points to the increasing substitution of brain for brawn
within our organizations and social lives. Yet we have delayed few conceptional
tools to better work with “wetware.”
It is with these forces in mind that we offer the following volume to you.
While there are, as yet, few agreed upon standards and analytic frames and defini-
tions, there are enough serious articles and books to help managers get some real
traction in dealing with the crucial yet elusive subject of Knowledge.
After all, we have had about 500 years of thought concerning the other ma-
jor factors of production (e.g., land, labor, and capital). Let these volumes start
the process of codifying knowledge about knowledge in order for us to better
manage in the twenty-first century.

Laurence Prusak
 and Organizational Design:
An Introduction
Paul S. Myers

For more than a decade, management thinkers have heralded the arrival of the
new information economy characterized by globalization, increased complexity,
and rapid change. “Do more with less!,” “Don’t automate, obliterate!,” and “Get
innovative or get dead!”’ are but a few of the leading words of advice for meeting
the newly shaped competitive environment. Underlying many of these prescrip-
tions is the need to explicitly manage the intellectual capital and other knowledge
assets of a firm. Corporate success in today’s economy comes from being able to
acquire, codify, and transfer knowledge more effectively and with greater speed
than the competition.
But, managers and practitioners sensibly ask, how d o we get there from
here? Fortunately, for now we can draw on decades of research and ideas devel-
oped by social scientists and management consultants as a starting point for
understanding how to manage organizational knowledge. Research on decision-
making, innovation, and work design, to name but a few subjects, has a lot to say
about the contemporary concerns and practical problems of knowledge rnanage-
ment. This present volume is a collection of articles and book excerpts selected for
their insight into the relationship between organizational design and knowledge
management. It is the first anthology of its kind to draw together the work of
leading economists, sociologists, psychologists, management thinkers, and practi-
tioners, each with a unique contribution to our understanding of how the form
and management of organizations shapes their levels of knowledge transfer, inno-
vation, and learning.
While much of the work in the field of organization design done prior to the
1990s discusses “information” and “expertise” rather than “knowledge” per se,
its distinctive perspective is readily applicable to knowledge management issues.
These connections become clearer, of course, once we have a working definition of

‘Rosabeth Moss Kanter, When Gtants Learn To Dance, (NY: Simon & Schuster), 1989; Michael
Hammer, Harvard Business Review (July-August) 1990; Tom Peters, California Management Re-
w e w Winter & Spring, 1991.
1
2 KNOWLEDGE MANAGEMENT A N D OIIGANIZATIONAI, DESIGN

organizational knowledge. We can define organizational knowledge as processed

information embedded in routines and processes which enable action. At its core,
knowledge must be seen as tied to the personal or human element. Knowledge, as
we generally understand it, resides in peoples’ heads; for, after all, individuals must
identify, interpret, and internalize knowledge. The representation of knowledge,
however, can be mechanical, digital, visual, and so forth. For knowledge to provide
a company with sustainable competitive advantage, such knowledge must be inde-
pendent from any given individual. For this reason, we can identify-and then
manage-organizational knowledge only to the extent it has been captured by an
organization’s systems, processes, products, rules, and culture.

ORGANIZATION DESIGN AND KNOWLEDGE

MANAGEMENT
Organizational performance is the result of the interaction of strategy, or-
ganizational context, and individual behavior. At the risk of over-simplification,
this means managers need to choose the right approach to the right markets, cre-
ate processes to deliver quality goods and/or services to those markets, and moti-
vate people to act in line with the company’s objectives. Volumes have been
written about business strategy, including some recent books on the centrality of
knowledge management to successful strategies (e.g., Nonaka & Takeuchi, 1995;
Davis & Botkin, 1994). The high level of attention fueled by the “reengineering”
movement has added the language of business processes to every manager’s vo-
cabulary for structuring work and organizations (e.g., Hammer & Champy,
1993; Davenport, 1993). Many recent management books and articles emphasize
creating empowered teams and valuing diversity as ways to motivate high per-
formance in knowledge-intensive businesses (e.g., Nevis, Lancourt, & Vassallo,
1996; McGill & Slocum, 1994).
Organizational design takes into account all three critical performance fac-
tors: strategy, organization, and motivation. At its base, the approach presumes
that a person’s actions are influenced by his or her situation.’ Most practices
derived from this tradition are based on the belief that firms achieve effective per-
formance by aligning, or making consistent, various organizational features. Or-
ganization design interventions deal with modifying elements of an organization’s
structure, including the division of labor, allocation of decision rights, choice of
coordinating mechanisms, delineation of organizational boundaries, and net-
works of informal relationships (see Cash, et al, 1994).
The impetus for this collection of articles is the belief that businesses will
find it increasingly difficult to succeed in a knowledge-intensive economy without
leveraging the power of organizational design for effective knowledge manage-
ment. Organizational design is about enabling a group of people to combine, co-

’As this introduction is not intended as a comprehcnsive synthesis of the field, interested readers may
wish to look toward recent work by Galbraith ( 1 9 9 5 ) , Mohrman, Cohen, and Mohrman (1995),
and Nadler, at a! (1992) for additional insights.
ordinate, and control resources and activities in order to produce value, all in a way
appropriate to the environment in which the business competes. In this view, design is
more a process than a structure in that the resulting organization should be intended
as constantly adapting and evolving, not fixed forever in some pre- determined form.
Recent empirical work provides evidence for this view. Mohrman and her colleagues
concluded from their study of knowledge work in eleven Fortune 500 companies that
“appropriate organizational design enables an organization to execute better, learn
faster, and change more easily” (19957).The task for leaders is to implement (by us-
ing the various structural levers) the mix of factors which increases the likelihood that
individual and organizational knowledge acquisition, codification, and transfer will
occur regularly, appropriately, and productively.
The increased importance of organizational design for business performance
challenges the current paradigm and experience of many managers w h o may be
quick to propose a technological (“We need better databases!”) or a personnel-re-
lated solution (“Hire smarter people! Fire the laggards!”), rather than a structural
one, to address business performance issues. As part of their respective training,
doctors study the basic structures of the human body, engineers and architects un-
derstand the fundamentals of physics and materials science, auto mechanics learn
how a combustion engine works. Only when they understand the underlying con-
cepts, phenomenon, and variables i n their field and how they interrelate, can these
professionals deal with specific issues and identify solutions. Similarly, managers
need to raise to a conscious, actionable level their understanding of the fundamen-
tal behavior, processes, and dynamics of the organizations they lead.

THE SELECTIONS
This collection draws on fifty years of management thinking to present a
useful introduction to key issues facing knowledge-intensive organizations. The
articles have been selected from among classic works which have influenced the
field of organizational design as well as from more recent contributions which de-
scribe contemporary leading practices in knowledge management.
Even the most cursory review of writings on the subject of organizational
design during this century reveals an enormous array of approaches and perspec-
tives, topics of interest, underlying assumptions, research methodologies, and lev-
els of analysis. O n one side the field is anchored by social science discipline-based
researchers w h o strive to understand society and organizations as they are and
who treasure that knowledge for its own sake. O n the other end of the spectrum
are action-oriented consultants w h o are interested in providing solutions to prac-
tical problems of administration, but who pay scant attention to any higher-order
or more generally applicable learning. Located between these two groups are
those practitioners and management thinkers w h o seek to combine theory and
practice, to understanding social systems in order to change and improve them.
This volume, I hope, falls squarely in the middle of that spectrum. The selec-
tions are clearly written and concise and reflect the mix of empirical data and con-
ceptual richness which makes their message for practitioners both credible and
4 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

actionable while being suitable for further testing and refinement by academic re-
searchers. Arranged thematically, the chapters discuss decision-making, organiza-
tion structure, innovation, strategic alliances, managing knowledge workers, and
power relations.
The anthology begins with a pair of theoretical pieces to introduce a per-
spective that is useful when reading the remaining selections. Theories allow us to
identify critical variables and their relationships in a particular situation and then
to generalize those findings to wider applicability. Without a theory of action-
an explanation of cause and effect or a prediction about likely outcomes-a man-
ager’s anecdote or an office’s experiment cannot be a reliable and sustainable
source of leveragable knowledge or replicable learning.
Hayek’s classic essay provides an economic theory of the link between
knowledge and organizational structure. He provides a taxonomy which distin-
guishes between two kinds of knowledge by pointing out the importance of
context and interpretation to its value. While Hayek addressed issues at the
macro-economic level, fifty years later Jensen and Meckling have written a com-
panion piece to Hayek’s which places his ideas in an organizational context. They
expand on Hayek’s ideas about the cost of transferring knowledge and argue that
organizational design is about more than simply choosing between centralization
and decentralization in the allocation of resources. Rather, they posit a framework
for understanding the links between knowledge, decision making, and organiza-
tion design which identifies the choices managers must make about their organi-
zations in order to align individual behavior with corporate objectives.
Not surprisingly, the vast majority of management writing on organization
design has dealt with industrial age organizations. The dominant factor influenc-
ing design choices, as the contingency theory approach posits, is the external envi-
ronment of markets, technologies, and regulations. Pinchot and Pinchot describe
the familiar, traditional bureaucratic organization and explain why it is no longer
suited to the knowledge era. They focus their analysis on the changing nature of
work and argue that the old model is being replaced by an interdependent set of
structures and practices which better fits the demands of today’s environment. Ba-
hrami’s research in high technology companies suggests that flexible, agile firms
with “novel organization structures and management processes to accommodate
them” may be the leading organizational form of the next decade. He points out
that this new form presents managers with a fresh set of dilemmas and tensions to
be addressed on an on-going basis, rather than resolved ex ante through a single
design choice.
Innovation requires applied knowledge, and highly innovative organizations
are adept at knowledge acquisition, codification, and transfer (see Nonaka &
Takeuchi, 1995; Kanter, 1983). Burns and Stalker give an historical review of how
innovation became organized in this century. While the book from which this se-
lection has been excerpted, The Management of Innovation, is best known for in-
troducing the concepts of “organic” and “mechanistic” organizations and linking
their respective desirability to the stability of the external environment, this early
chapter helps us understand how conditions at the societal level influence levels of
innovation. They emphasize that invention is a social phenomenon relying in
Knowledge Marzagement aizd Orgaitizational Design: Ait Introdtiction 5

great part on the diffusion of information through professional relationships and

suggest that an organization’s design can hinder or facilitate that process. In a
thorough review of the field, Kanter fully fleshes out Burns and Stalker’s ideas
about the importance of organizational context in producing innovation. Break-
ing down the innovation process into four sets of tasks, she describes the struc-
tural, collective, and social conditions which enhance the flow of information and
knowledge, facilitate and strengthen relationships, and support individual and
group creativity.
Kanter’s article touches on the importance of inter-organizational relation-
ships as an important source of input for the innovation process, and the growing
numbers of business alliances and joint ventures over the past decade suggests that
companies are well aware of this benefit from partnerships (e.g., Lewis, 1995;
Yoshino & Rangan, 1995). Badaracco studied the variety of alliances created by
IBM and General Motors and found that among their main benefits were the crea-
tion of knowledge and learning new practices. This was particularly true regard-
ing “embedded” knowledge; for Badaracco, embedded knowledge is a more
social notion than Hayek’s “particular knowledge” for it “resides primarily in
specialized relationships among individuals and groups and in the particular
norms, attitudes, information flows, and ways of making decisions that shape
their dealings with each other” (1991: 79). Although such knowledge is difficult
and costly to transfer, alliances create an arena for building ties which will facili-
tate learning over time. Pucik provides some guidance for practitioners in how to
increase the chances t h a t such learning actually occurs. He outlines a role for the
human resource function in supporting organizational learning, which requires
transforming HR from an administrative unit to a valued business asset. Drawing
on the alliance experiences of firms which demonstrate high levels of learning, he
also offers specific advice for human resource professionals.
The need to manage knowledge more consciously raises more direct human re-
source implications: how can firms attract and retain knowledge workers? The rise of
the “intelligent organization,” in Pinchot’s phrase, has placed a premium on human
assets rather than natural resources or financial capital which reigned supreme in ear-
lier eras. Handy explores this phenomenon by reviewing the global demographic
shifts which are occurring simultaneously with the emergence of the information-in-
tensive economy. If Handy suggests that finding workers prepared for knowledge-era
businesses will be a limiting factor for corporate success in the years ahead, Tampoe
reminds us that motivating excellent performance presents challenges as well. More
and more, companies must meet the expectations of their knowledge workers in
terms of work environment and quality of life issues or else face lower productivity or
even turnover of their most highly valued assets. Addressing the personal and social
objectives of employees to build effective and “empowering” workplaces, he argues,
is the key to motivating knowledge workers.
The final group of articles deal with the nexus of knowledge, work, and
power. Organizational design shapes the flow of information, resources, and sup-
port within a firm, and in this way strongly determines the powerholders. Croz-
ier’s classic description of the French cigarette factory demonstrates the power
that can accrue to holders of technical knowledge, especially when few or no oth-
6 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

ers possess t h a t knowledge. T h e transformation of the s h o p floor t h r o u g h c o m -

puterization, however, threatens t o diminish or even eliminate such expert p o w e r
for s o m e individuals. T h e excerpt detailing t h e experience of paper mills w o r k e r s
f r o m Zuboff’s insightful ethnographic study of the effects of technological c h a n g e
illustrates how p o w e r is shifting f r o m w o r k e r s t o college-trained process engi-
neers. Clearly, s o m e employees gain while others lose p o w e r a s a result of t h e in-
creased value placed on organizational knowledge. Baker provides s o m e guidance
a b o u t how to exploit one’s position in t h e organization structure t o create o p p o r -
tunities a n d advantages for individuals a n d firms. H e points o u t t h a t n o t merely
knowledge, b u t social capital-who o n e k n o w s a n d how o n e handles those rela-
tionships-are critical assets for business a n d personal success.
Taken together, these thirteen selections connect organizational design con-
cepts a n d ideas t o the exigencies of knowledge m a n a g e m e n t for today’s c o r p o r a -
tions. T h e y a r e intended as a catalyst t o management action, whether by s p a r k i n g
a conversation a m o n g colleagues, providing a solution for a project t e a m , or in-
spiring a leader t o try a n innovative design. While this anthology represents only
a n introduction t o t h e w o r k of researchers a n d practitioners which focuses on t h e
role of organizational f o r m in business performance, it makes clear t h a t t h e or-
ganizational design a p p r o a c h offers valuable insights for those charged w i t h help-
ing their organizations acquire, codify, a n d transfer knowledge.

REFERENCES
Cash, James I., Jr., Robert G. Eccles, Nitin Nohria, and Richard L. Nolan. 1994. Building
the Information-Age Organization: Structure, Control, and Information Technologies.
Homewood, IL: Irwin.
Davenport, Thomas H. 1993. Process Innovation. Boston: Harvard Business School Press.
Davis, Stan and Jim Botkin. 1994. The Monster Under The Bed. New York: Simon & Schuster.
Galbraith, Jay R. 1995. Designing Organizations. San Francisco: Jossey-Bass.
Hammer, Michael and James Champy. 1993. Reengineering the Corporation. New York:
Harper Business.
Kanter, Rosabeth Moss. 1983. The Change Masters. New York: Simon & Schuster.
Lewis, Jordan D. 1995. The Connected Corporation. New York: The Free Press.
McGill, Michael E. and John W. Slocum, J . 1994. The Smarter Organization. New York:
John Wiley & Sons.
Mohrrnan, Susan A,, Susan Cohen, and Allan M. Mohrman, Jr. 1995. Designing Team-
Based Organizations: New F o m s for Knowledge Work. San Francisco: Jossey-Bass.
Nadler, David A., Marc S. Gerstein, Robert B. Shaw, and associates. 1992. Organizational
Architecture. San Francisco: Jossey-Bass.
Nevis, Edwin C., Joan Lancourt, and Helen Vassallo. 1996. International Revolrrtions: A
Seven Point Strategy For Moving Beyond Resistance. San Francisco: Jossey-Bass.
Nonaka, Ikujiro and Hirotaka Takeuchi. 1995. The Knowledge-Creating Company. New
York: Oxford University Press.
Yoshino, Michael Y. and U. Srinivasa Rangan. 1995. Strategic Alliances. Boston: Harvard
Business School Press.
 in Society
Frederick A. Hayek

Many of the current disputes with regard to both economic theory and economic
policy have their common origin, it seems to me, in a misconception about the na-
ture of the economic problem of society. This misconception in turn is due to an
erroneous transfer to social phenomena of the habits of thought we have devel-
oped in dealing with the phenomena of nature.
What is the problem we wish to solve when we try to construct a rational
economic order? On certain familiar assumptions the answer is simple enough. I f
we possess all the relevant information, if we can start o u t from a given system of
preferences, and if we command complete knowledge of available means, the
problem which remains is purely one of logic. That is, the answer to the question
of what is the best use of the available means is implicit in our assumptions. Stated
briefly in mathematical form, it is that the marginal rates of substitution between
any two commodities or factors must be the same in all their different uses.
This, however, is emphatically not the economic problem which society faces.
And the economic calculus which we have developed to solve it, though an impor-
tant step toward the solution of the economic problem of society, does not yet pro-
vide an answer to it. The reason for this is that the “data” for the whole society,
from which the economic calculus starts, are never “given” to a single mind.
The peculiar character of the problem of a rational economic order is deter-
mined precisely by the fact that the knowledge of the circumstances which we
must use never exists in concentrated or integrated form but solely as the dis-
persed bits of incomplete and frequently contradictory knowledge which separate
individuals possess. The economic problem of society is thus not merely a prob-
lem of how to allocate “given” resources-if “given” is taken to mean given to a
single mind which deliberately solves the problem set by these “data.” It is rather
a problem of how to secure the best use of resources known to any of the mem-
bers of society, for ends whose relative importance only these individuals know.

From “The Use of Knowledge in Society,” by Frederick A. Hayek. Reprinted with permission from The
American Economic Review, 35:4 (September). Copyright 0 1945 American Economic Association.
8 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

Or, to put it briefly, it is a problem of the utilization of knowledge which is not

given to anyone in its totality.
This character of the fundamental problem has, I am afraid, been obscured
rather than illuminated by many of the recent refinements of economic theory,
particularly by many of the uses made of mathematics. Though the problem with
which I want primarily to deal in this paper is the problem of a rational economic
organization, I shall be led again and again to point to its close connections with
certain methodological questions. Many of the points I wish to make are indeed
conclusions toward which diverse paths of reasoning have unexpectedly con-
verged. But, as I now see these problems, this is no accident.
In ordinary language, we describe by the word “planning” the complex of
interrelated decisions about the allocation of our available resources. All eco-
nomic activity is in this sense planning; and in any society in which many people
collaborate, this planning, whoever does it, will in some measure have to be based
on knowledge which, in the first instance, is not given to the planner but to some-
body else, yet somehow will have to be conveyed to the planner. The various ways
in which this knowledge on which people base their plans is communicated to
them is the crucial problem for any theory explaining the economic process, and
the problem of what is the best way of utilizing knowledge initially dispersed
among all the people is at least one of the main problems of economic policy-or
of designing a n efficient economic system.
The answer to this question is closely connected with that other question which
arises here, that of who is to do the planning. It is about this question that all the dis-
pute about “economic planning” centers. This is not a dispute about whether plan-
ning is to be done or not. It is a dispute as to whether planning is to be done centrally,
by one authority for the whole economic system, or is to be divided among many in-
dividuals. Planning in the specific sense in which the term is used in contemporary
controversy necessarily means central planningdirection of the whole economic
system according to one unified plan. Competition, on the other hand, means decen-
tralized planning by many separate persons. The half-way house between the two,
about which many people talk but which few like when they see it, is the delegation
of planning to privileged industries, or, in other words, monopolies.
Which of these systems is likely to be more efficient depends mainly on the
question under which of them we can expect that fuller use will be made of the
existing knowledge. This, in turn, depends on whether we are more likely to suc-
ceed in putting at the disposal of a single central authority all the knowledge
which ought to be used but which is initially dispersed among many different in-
dividuals, or in conveying to the individuals such additional knowledge as they
need in order to enable them to dovetail their plans with those of others.

UNCOMMON KNOWLEDGE
It will at once be evident that on this point the position will be different with
respect to different kinds of knowledge. The answer to our question will therefore
largely turn on the relative importance of the different kinds of knowledge: those
The Use of Knowledge in Society 9

more likely to be at the disposal of particular individuals and those which we

should with greater confidence expect to find in the possession of an authority
made up of suitably chosen experts. If it is today so widely assumed that the latter
will be in a better position, this is because one kind of knowledge, namely, scien-
tific knowledge, occupies now so prominent a place in public imagination that we
tend to forget that it is not the only kind that is relevant. I t may be admitted that,
as far as scientific knowledge is concerned, a body of suitably chosen experts may
be in the best position to command all the best knowledge available-though this
is of course merely shifting the difficulty to the problem of selecting the experts.
What I wish to point out is that, even assuming t h a t this problem can be readily
solved, it is only a small part of the wider problem.
Today it is almost heresy to suggest t h a t scientific knowledge is not the sum
of all knowledge. But a little reflection will show that there is beyond question a
body of very important but unorganized knowledge which cannot possibly be
called scientific in the sense of knowledge of general rules: the knowledge of the
particular circumstances of time and place. It is with respect to this that practi-
cally every individual has some advantage over all others because he possesses
unique information of which beneficial use might be made only if the decisions de-
pending o n it are left to him or are made with his active co-operation.
We need only to remember how much we have to learn in any occupation
after we have completed o u r theoretical training, how big a part of o u r working
life we spend learning particular jobs, and how valuable an asset in all walks of
life is knowledge of people, of local conditions, and of special circumstances. To
know of and put to use 3 machine not fully employed, or somebody’s skill which
could be better utilized, or to be aware of a surplus stock which can be drawn
upon during an interruption of supplies, is socially quite as useful as the knowl-
edge of better alternative techniques. The shipper who earns his living from using
otherwise empty or half-filled journeys of tramp-stearners, or the estate agent
whose whole knowledge is almost exclusively one of temporary opportunities, o r
the arbitrageur who gains from local differences of commodity prices-all are per-
forming eminently useful functions based on special knowledge of Circumstances
of the fleeting moment not known to others.
It is a curious fact that this sort of knowledge should today be generally re-
garded with a kind of contempt, that anyone who, by such knowledge, gains an
advantage over somebody better equipped with theoretical or technical knowl-
edge is thought to have acted almost disreputably. To gain an advantage from bet-
ter knowledge of facilities of communication or transport is sometimes regarded
as almost dishonest, although it is quite as important that society make use of the
best opportunities in this respect as in using the latest scientific discoveries. This
prejudice has in a considerable measure affected the attitude toward commerce in
general compared with that toward production. Even economists who regard
themselves as immune to the crude materialist fallacies of the past constantly com-
mit the same mistake where activities directed toward the acquisition of such
practical knowledge are concerned-apparently because i n their scheme of things
all such knowledge is supposed to be “given.” The common idea now seems to be
that all such knowledge should as a matter of course be readily a t the command of
10 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

everybody, and the reproach of irrationality leveled against the existing economic
order is frequently based on the fact that it is not so available. This view disre-
gards the fact that the method by which such knowledge can be made as widely
available as possible is precisely the problem to which we have to find an answer.

THE PLANNER’S DILEMMA

If it is fashionable today to minimize the importance of the knowledge of the

particular circumstances of time and place, this is closely connected with the
smaller importance which is now attached to change as such. Indeed, there are
few points on which the assumptions made (usually only implicitly) by the “plan-
ners’’ differ from those of their opponents as much as with regard to the signi-
ficance and frequency of changes which will make substantial alterations of
production plans necessary. Of course, if detailed economic plans could be laid
down for fairly long periods in advance and then closely adhered to so that no
further economic decisions of importance would be required, the task of drawing
up a comprehensive plan governing all economic activity would be much less
formidable.
It is, perhaps, worth stressing that economic problems arise always and only
in consequence of change. As long as things continue as before, or a t least as they
were expected to, there arise no new problems requiring a decision, no need to
form a new plan. The belief that changes, or at least day-to-day adjustments, have
become less important in modern times implies the contention that economic
problems also have become less important. This belief in the decreasing irnpor-
tance of change is, for that reason, usually held by the same people who argue that
the importance of economic considerations has been driven into the background
by the growing importance of technological knowledge.
Is it true that, with the elaborate apparatus of modern production, economic
decisions are required only at long intervals, as when a new factory is to be
erected or a new process to be introduced? Is it true that, once a plant has been
built, the rest is all more or less mechanical, determined by the character of the
plant, and leaving little to be changed in adapting to the ever changing circum-
stances of the moment?
The fairly widespread belief in the affirmative is not, as far as I can ascer-
tain, borne out by the practical experience of the businessman. In a competitive
industry a t any rate-and such an industry alone can serve as a test-the task of
keeping costs from rising requires constant struggle, absorbing a great part of the
energy of the manager. How easy it is for an inefficient manager to dissipate the
differentials on which profitability rests. A great variety in costs of production,
even when using the same technical facilities, is commonplace in business experi-
ence but does not seem to be equally familiar to many economists. The very
strength of the desire, constantly voiced by producers and engineers, to be allowed
to proceed untrammeled by considerations of money costs, is eloquent testimony
to the extent to which these factors enter into their daily work.
One reason why economists are increasingly apt to forget about the con-
The Use of Knowledge in Society 11

stant small changes which make up the whole economic picture is probably their
growing preoccupation with statistical aggregates, which show a very much
greater stability than the movements of the detail. The comparative stability of the
aggregates cannot, however, be accounted for-as the statisticians occasionally
seem to be inclined to do-by the “law of large numbers” or the mutual compen-
sation of random changes. The number of elements with which we have to deal is
not large enough for such accidental forces to produce stability. The continuous
flow of goods and services is maintained by constant deliberate adjustments, by
new dispositions made every day in the light of circumstances not known the day
before, by B stepping in a t once when A fails to deliver. Even the large and highly
mechanized plant keeps going largely because of an environment upon which it
can draw for all sorts of unexpected needs: tiles for its roof, stationery or its
forms, and all the thousand and one kinds of equipment in which it cannot be self-
contained and which the plans for the operation of the plant require to be readily
available in the market.
I should also briefly mention the fact that the sort of knowledge with which I
have been concerned is knowledge of the kind which by its nature cannot enter into
statistics and therefore cannot be conveyed to any central authority in statistical
form. The statistics which such a central authority would have to use would have
to be arrived at precisely by abstracting and then lumping together items which dif-
fer as regards location, quality, and other particulars in ways that may be very sig-
nificant for the specific decision. It follows from this that central planning which is
based on statistical information cannot by its nature take direct account of these
circumstances of time and place. The central planner will have to find some way to
make decisions by leaving them to be made by the “man on the spot.”
If we can agree that the economic problem of society is mainly one of rapid
adaptation to changes in the particular circumstances of time and place, it would
seem to follow that the ultimate decisions must be left to the people who are fa-
miliar with these circumstances, who know directly of the relevant changes and of
the resources immediately available to meet them. We cannot expect that this
problem will be solved by first communicating all this knowledge to a central
board which, after integrating all knowledge, issues its orders. We must solve it by
some form of decentralization.
But this answers only part of our problem. We need decentralization because
only thus can we insure that the knowledge of the particular circumstances of time
and place will be promptly used. The “man on the spot” cannot, however, decide
solely on the basis of his limited but intimate knowledge of the facts of his imme-
diate surroundings. There still remains the problem of communicating to him
such further information as he needs to fit his decisions into the whole pattern of
changes of the larger economic system.

USEFUL INDIVIDUAL KNOWLEDGE

How much knowledge does the individual need in order to make these deci-
sions successfully? Which of the events beyond his own horizon of immediate
12 K N 0W L ED G E MAN A G E hl ENT AND 0 I<C; AN I Z AT10N A L DESIGN

knowledge are of relevance to his immediate decision, and how much of them
need he know?
There is hardly anything that happens anywhere in the world that might not
have an effect on the decision he ought to make. But he need not know of these
events as such, nor of all their effects. It does not matter for him why more screws
of one size than of another are wanted a t the particular moment, why paper bags
are more readily available than canvas bags, or why skilled labor, or particular
machine tools, have for the moment become more difficult to obtain. All that is
significant for him is how much more or less difficult to procure they have become
compared with other things with which he is also concerned, or how much more
or less urgently wanted are the alternative things he produces or uses. It is always
a question of the relative importance of the particular things with which he is con-
cerned, and the causes which alter their relative importance are of no interest to
him beyond the effect on those concrete things of his own environment.
It is in this connection that what I have called the “economic calculus” (or
the Pure Logic of Choice) helps us, at least by analogy, to see how this problem
can be solved and is being solved by the price system. Even the single controlling
mind, in possession of all the data for some small and self-contained economic
system, would not-every time some small adjustment in the allocation of re-
sources had to be made-go explicitly through all the relations between ends and
means which might possibly be affected. It is indeed the great contribution of the
Pure Logic of Choice to have demonstrated conclusively that even such a single
mind could solve this kind of problem only by constructing and constantly using
rates of equivalence (or “values,” or “marginal rates of substitution”); that is, he
would have to attach to each kind of scarce resource a numerical index which
cannot be derived from any property possessed by t h a t particular thing, but which
reflects, or in which is condensed, its significance in view of the whole means-end
structure, In any small change he will have to consider only these quantitative in-
dices (or “values”) in which all the relevant information is concentrated; and, by
adjusting the quantities one by one, he can appropriately rearrange his disposi-
tions without having to solve the whole puzzle a6 inilio o r without needing a t any
stage to survey it a t once in all its ramifications.
Fundamentally, in ;i system in which the knowledge of the relevant facts is
dispersed among many people, prices can act to co-ordinate the separate actions
of different people in the same way as subjective values help the individual to co-
ordinate the parts of his plan.

“MIRACLE” OF T H E PRICE SYSTEM

It is worth contemplating for a moment a very simple and commonplace in-

stance of the action of the price system to see what precisely it accomplishes. As-
sume that somewhere in the world a new opportunity for the use of some raw
material, say, tin, has arisen, or that one of the sources of supply of tin has been
eliminated. It does not matter for o u r purpose-and it is significant that it does
The Use of Knowledge in Society 13

not matter-which of these two causes has made tin more scarce. All that the us-
ers of tin need to know is that some of the tin they used to consume is now more
profitably employed elsewhere and that, in consequence, they must economize tin.
There is no need for the great majority of them even to know where the more ur-
gent need has arisen, or in favor of what other needs they ought to husband the
supply. If only some of them know directly of the new demand and switch re-
sources over to it, and if the people who are aware of the new gap thus created in
turn fill it from still other sources, the effect will rapidly spread throughout the
whole economic system. This influences not only all the uses of tin but also those
of its substitutes and the substitutes of these substitutes, the supply of all things
made of tin, and their substitutes, and so on. All this takes place without the great
majority of those instrumental in bringing about these substitutions knowing any-
thing at all about the original cause of these changes. The whole acts as one mar-
ket, not because any of its members surveys the whole field, but because their
limited individual fields of vision sufficiently overlap so that through many inter-
mediaries the relevant information is communicated to all. The mere fact that
there is one price for any commodity-or rather that local prices are connected in
a manner determined by the cost of transport, etc.-brings about the solution
which (if conceptually possible) might have been arrived a t by one single mind
possessing all the information which is in fact dispersed among all the people in-
volved in the process.
We must look at the price system as such a mechanism for communicating
information if we want to understand its real function-a function which it fulfills
less perfectly as prices grow more rigid. (Even when quoted prices have become
quite rigid, however, the forces which would operate through changes in price still
operate to a considerable extent through changes in the other terms of the con-
tract.) The most significant fact about this system is the economy of knowledge
with which it operates, or how little the individual participants need to know in
order to be able to take the right action. In abbreviated form, by a kind of symbol,
only the most essential information is passed on, and this is passed on only to
those concerned. It is more than a metaphor to describe the price system as a kind
of machinery for registering change, or a system of telecommunications which en-
ables individual producers to watch merely the movement of a few pointers, as an
engineer might watch the hands of a few dials, in order to adjust their activities to
changes of which they may never know more than their reflection in the price
movement.
Of course, these adjustments are probably never “perfect” in the sense in
which the economist conceives of them in his equilibrium analysis. But I fear that
our theoretical habits of approaching the problem with the assumption of more or
less perfect knowledge on the part of almost everyone has made us somewhat
blind to the true function of the price mechanism and led us to apply rather mis-
leading standards in judging its efficiency. The marvel is that in a case like that of
a scarcity of one raw material, without an order being issued, without more than
perhaps a handful of people knowing the cause, tens of thousands of people
whose identity could not be ascertained by months of investigation, are made to
14 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

use the material or its products more sparingly; that is, they move in the right di-
rection. This is enough of a marvel even if, in a constantly changing world, not all
will react so perfectly that their profit rates will always be maintained at the same
“normal” level.
I have deliberately used the word “marvel” to shock the reader out of the
complacency with which we often take the working of the price mechanism for
granted. I a m convinced that if it were the result of deliberate human design, and
if the people guided by the price changes understood that their decisions have sig-
nificance far beyond their immediate aim, this mechanism would have been ac-
claimed as one of the greatest triumphs of the human mind. Its misfortune is the
double one that it is not the product of human design and that the people guided
by it usually d o not know why they are made to do what they do. But those who
clamor for “conscious direction”-and who cannot believe that anything which
has evolved without design (and even without our understanding it) can solve
problems which we cannot solve consciously-should remember this: the problem
is precisely how to extend our utilization of resources beyond the span of the con-
trol of any one mind; and, therefore, how to dispense with the need of conscious
control, how to provide inducements which will make the individuals d o the de-
sirable things without anyone having to tell them what to do.
The problem which we meet here is by no means peculiar to economics but
arises in connection with nearly all truly social phenomena, including language
and most of our cultural inheritance, and constitutes really the central theoretical
problem of all social science. As Alfred Whitehead has said in another connection,
“It is a profoundly erroneous truism, repeated by all copy-books and by eminent
people when they are making speeches, that we should cultivate the habit of
thinking what we are doing. The precise opposite is the case. Civilization ad-
vances by extending the number of important operations which we can perform
without thinking about them.” This is of profound significance in the social field.
We make constant use of formulas, symbols, and rules whose meaning we d o not
understand and through the use of which we avail ourselves of the assistance of
knowledge which individually we d o not possess. We have developed these prac-
tices by building upon habits and institutions which have proved successful in
their own sphere and which have in turn become the foundation of the civilization
we have built up,
The price system is just one of those formations which man has learned to
use (though he is still very far from having learned to make the best use of it) after
he had stumbled upon it without understanding it. Through it, not only a division
of labor but also a co-ordinated utilization of resources based on a similarly di-
vided knowledge has become possible. The people who like to deride any sugges-
tion that this may be so usually distort the argument by insinuating that it asserts
that by some miracle just that sort of system has spontaneously grown up which
is best suited to modern civilization. It is the other way round: man has been able
to develop that division of labor on which our civilization is based because he
happened to stumble upon a method which made it possible. Had he not done so,
he might still have developed some other, altogether different type of civilization,
The Use of Knowledge in Society 15

something like the “state” of the termite ants, or some other altogether unimagin-
able type. All that we can say is that nobody has yet succeeded in designing an al-
ternative system in which certain features of the existing one can be preserved
which are dear even to those who most violently assail it-such as particularly the
extent to which the individual can choose his pursuits and consequently freely use
his own knowledge and skill.
It is in many ways fortunate that the dispute about the indispensability of
the price system for any rational calculation in a complex society is now no longer
conducted entirely between camps holding different political views. The thesis
that without the price system we could not preserve a society based on such exten-
sive division of labor as ours was greeted with a howl of derision when it was first
advanced by von Mises in the early 1920’s. Today the difficulties which some still
find in accepting it are no longer mainly political, and this makes for an atmos-
phere much more conducive to reasonable discussion. When we find Leon Trotsky
arguing that “economic accounting is unthinkable without market relations”;
when Professor Oscar Lange promises Professor von Mises a statue in the marble
halls of the future Central Planning Board; and when Professor Abba P. Lerner re-
discovers Adam Smith and emphasizes that the essential utility of the price system
consists in inducing the individual, while seeking his own interest, to do what is in
the general interest, the differences can indeed no longer be ascribed to political
prejudice. The remaining dissent seems clearly to be due to purely intellectual, and
more particularly methodological, differences.
This page intentionally left blank
 Knowledge, and
Organizational Structure
Michael C. Jensen and William H. Meckling”

Specific knowledge is knowledge that is costly to transfer among agents and gen-
eral knowledge is knowledge that is inexpensive to transmit. Because specific
knowledge is costly to transfer, getting it used in decision-making requires decen-
tralizing many decision rights in both the economy and in firms. Such delegation
in turn creates two problems: the rights assignment problem (determining who
should exercise a decision right), and the control or agency problem (ensuring that
self-interested decision agents exercise their rights in a way that contributes to the
organizational objective).
Capitalist economic systems solve the rights assignment and control prob-
lems by granting alienability of decision rights to decision agents. A right is alien-
able if its owner has the right to sell it and capture the proceeds offered in the
exchange. Indeed, we define “ownership” to mean possession of a decision right
along with the right to alienate that right, and we believe that when people use the
word ownership that is what is they mean. This combination of a decision right
with the right to alienate that right is also what is generally meant by the term
“property right” that is so often used in economics.’
In contrast to markets, organizations generally d o not delegate both decision
rights and the alienability of those rights to the agent. A machine operator might
be delegated the rights to operate and maintain a machine, but not the rights to
sell it and pocket the proceeds. In the absence of alienability, organizations must
solve both the rights assignment and control problems by setting up alternative
systems and procedures. We discuss the critical role that alienability plays in the
market system and some of the substitute control mechanisms used in firms.

‘This article is a slightly revised version of an article by the same title that was published as Chapter
9 in Contract Economics, Lars Werin and Hans Wijkander, eds. (Oxford: Basil Blackwell Ltd.,
1992). Our research has been supported by the Managerial Economics Research Center, University
of Rochester, and the Division of Research, Harvard Business School. We are grateful for the
comments and criticisms of George Baker, Robert Eccles, Lars Werin, and Karen Wruck.

From Journal ofApplied Corporate Finance, Vol. 8, No. 2 (Summer): 4-18. Copyright 0 1995 Jour-
nal of Applied Corporate Finance. Reprinted with permission.
17
18 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

COLLOCATION OF KNOWLEDGE AND DECISION

AUTHORITY

E A. Hayek was among the first economists to note the importance of

knowledge and its distribution to a well-functioning economy. In his seminal arti-
cle, “The Use of Knowledge in Society,” Hayek argues that most economists, as
well as advocates of centralized planning, misunderstand the nature of the eco-
nomic problem. “The economic problem of society . . . is not merely a problem of
how to allocate ‘given’ resources-if ‘given’ is taken to mean given to a single
mind. . . . It is rather a problem of how to secure the best use of resources known
to any of the members of society. . . a problem of the utilization of knowledge
which is not given to anyone in its totality.”2 Hayek’s insight was that an organi-
zation’s performance depends on the collocation of decision-making authority
with the knowledge important to those decision^.^ He argues that the distribution
of knowledge in society calls for decentralization:

I f we . . . agree that the economic problem of society is mainly one of rapid

adaptation to changes in the particular circumstances of time and place. . .
decisions must be left to the people who are familiar with these circum-
stances, who know directly of the relevant changes and of the resources im-
mediately available to meet them. We cannot expect that this problem will
be solved by first communicating all this knowledge to a central board
which, after integrating all knowledge, issues its orders. We must solve it by
some form of de~entralization.~

Hayek’s pioneering work provides a point of departure for analyzing how

the distribution of knowledge affects organizational structure and its critical role
in the development of a theory of organization. Hayek assumes that markets
automatically move decision rights to the agents with the relevant knowledge, and
that those agents will use the decision rights properly. Unfortunately, he never dis-
cusses how this occurs. We show how understanding this issue provides insights
into the organizational and managerial problems of firms.
In the second section of this paper, we discuss the limits of human mental ca-
pacities and their implications for the costs of transferring knowledge. The third
section defines the characteristics of decision rights and the different systems by
which such rights are allocated. The fourth discusses the role of alienability in
solving the rights assignment and control problems in markets, and the implica-
tions of this market solution for the internal problems faced by firms and other
large organizations that cannot use alienability to solve the rights assignment and
control problems. The fifth section discusses the problems of the firm in collocat-
ing decision rights and specific knowledge, the sixth discusses the technology for
partitioning decision rights within the firm, and the seventh discusses corporate
internal control systems.
Specific and General Knowledge, and Organizational Structure 19

KNOWLEDGE

The opportunity set confronting an individual or a firm is a function of the

individual’s knowledge. Decision-makers confront the limits of their knowledge at
two levels. One is “technological feasibility,” by which we mean the limits to hu-
man knowledge about physical laws. Most economic analysis effectively sidesteps
this issue by assuming a given, fixed level of technological knowledge.
The second limitation on knowledge, and the one of primary concern here,
arises from those limitations that are specific to each i n d i v i d ~ a l Humans
.~ have
limited mental capability. The computers and sensory systems with which we are
individually endowed are a scarce resource with limited storage and processing
capability, as well as limited input and output channels. The limitations on human
mental and sensory faculties mean that storing, processing, transmitting, and re-
ceiving knowledge are costly activities.
This limited capacity of the brain means that knowledge possessed by any
individual decision-maker or group of decision-makers is thereby limited to a
minuscule subset of the knowledge known to humanity. While decision-makers
seldom, if ever, possess all available knowledge, they are constantly creating new
knowledge. In maximizing their own objective functions, decision-makers deli-
berately seek out knowledge (including knowledge about what decisions to
consider).
When knowledge is valuable in decision-making, there are benefits to collo-
cating decision authority with the knowledge that is valuable in making those de-
cisions. There are two basic ways to accomplish such a collocation of knowledge
and decision rights. One is by moving the knowledge to those with the decision
rights; the other is by moving the decision rights to those with the knowledge. The
process for moving knowledge to those with decision rights has received much at-
tention from researchers and designers of management information systems. But
the process for moving decision rights to those with the relevant knowledge has
received relatively little attention in either economics or management science.
In a market system, collocation of decision rights and knowledge is accom-
plished either when those with decision rights spend time and resources to acquire
the knowledge or when those with knowledge buy the decision rights. When the
cost of moving knowledge is higher than the cost of moving decision rights,
knowledge holders will value the decision rights more highly and will thus tend to
purchase them. In this way, optimizing behavior on the part of individuals causes
the distribution of decision-making rights in the economy to reflect the limitations
of human mental and sensory systems.

Knowledge and the Cost of Transfer

The cost of transferring knowledge depends on factors such as the nature of

the knowledge, the organizational environment, and technology. We use the terms
“specific” and “general” knowledge to distinguish between knowledge a t the ex-
20 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

tremes of the continuum measuring transfer costs. The more costly is knowledge
to transfer, the more specific it is; the less costly to transfer, the more general.
“Transfer,” as we use it, means effective transfer, not merely “communica-
tion.” The recipient of knowledge is assumed to understand the message well
enough to act on it. The simple purchase of a physics book is not sufficient to
transfer the knowledge to the purchaser (as evidenced by students who regularly
pay thousands of dollars for help in acquiring such knowledge). Thus, transfer in-
volves the use of storage and processing capacity as well as inpudoutput channels
of the human brain. Moreover, knowledge transfers are not instantaneous: it takes
people time to absorb information. These delays are costly; and for some deci-
sions such costs can be high, including even the complete loss of opportunities.
Hayek’s 1945 article takes the distribution of knowledge in the economy as
given and thus never mentions the costs of transferring or producing knowledge.
Nevertheless, the significance of such information costs are the logical foundation
of his a n a l y k 6 Writing during the 1940s’ debate in Britain over central planning,
Hayek attacks central planners on the grounds that they will make bad decisions
because they will not (indeed they cannot) have the knowledge of “particular cir-
cumstances of time and place” necessary to make the best decisions. As examples
of such “idiosyncratic” knowledge, he cites knowledge of the existence and loca-
tion of the following: a not-fully-employed machine, someone’s particular skills,
surplus stock, empty or half-filled freighters, temporary opportunities in real es-
tate, and commodity price differences.
As Hayek points out, conveying knowledge of such particular circumstances
to a central authority in statistical form is impossible. Aggregating or lumping to-
gether items such as location or quality destroys their usefulness for specific deci-
sions. Adding up the quantity of empty spaces in steamers or logs in widely
scattered wood piles, for example, eliminates the information about time and lo-
cation that is so valuable in periods of transportation or energy shortages.
Specific knowledge-of which such “idiosyncratic” knowledge of particular
circumstances is an example-is often acquired jointly with the production of
other goods. When knowledge is a by-product of activities that will be performed
anyway, the cost of that knowledge to the acquirer is nil. Other examples of idi-
osyncratic knowledge include knowledge of the specific skills or preferences of in-
dividuals, of the peculiarities of specific machines, of particular unemployed
resources or inventories, and of arbitrage opportunities. Such knowledge, almost
by definition, is difficult or impossible to aggregate and summarize.
Thus, while the initial costs of acquiring idiosyncratic knowledge tend to be
modest, the costs of transferring such information are likely to be high relative to
the benefits. Because time is often important in taking advantage of opportunities
for arbitrage or for exploiting knowledge of unemployed resources, delays in ac-
tions are costly.
Uncertainty about what specific piece of idiosyncratic knowledge will prove
valuable also enlarges transfer costs in a subtle way. After the fact, it is often obvi-
ous that a specific piece of knowledge critical to a decision could have been trans-
ferred a t low cost (for example, particular quirks of an organization, person, legal
Specific and Geiieral Knowledge, and Organizational Structwe 21

rule, or custom). But transferring this specific piece of knowledge in advance re-
quires knowing in advance that it will be critical. Without such clairvoyance,
transfer of the fact must occur as part of a larger and more costly-to-transfer body
of knowledge, most of which will never be used. The expected cost of transferring
that larger body of data, not the particular fact, is the relevant transfer cost.
Although knowledge of particular circumstances of time and place and idi-
osyncratic knowledge cannot be summarized in statistics, they can be transmitted
to other locations in the decision-making structure. The question is not whether
knowledge can be transferred, but at what cost it can be transferred, and whether
it is worth it to d o so. Transfers yield benefits when the additional knowledge en-
ables the decision-maker to make better choices. The issue is whether decisions
will be improved enough to warrant the transfer costs.
Quantities and prices are good examples of general knowledge. Unlike idi-
osyncratic or other specific knowledge, quantities are easily aggregated and trans-
ferred among agents at low cost. Prices, which are also easily communicated
among agents, are signals that communicate a large amount of information inex-
pensively. When a price rises, people know it is appropriate to conserve the com-
modity-and they need not know why its relative supply has shrunk.
Of course, we d o observe situations in which collocation of knowledge and
decision rights is achieved by transferring knowledge. Formal educational pro-
grams and the collection, analysis, and dissemination of data are obvious exam-
ples. United and American Airlines achieved a major competitive advantage with
computerized reservation and pricing systems that reduced the cost of transferring
knowledge about prices, empty seats, and schedules. Particularly challenging in-
formation transfer problems arise in situations where optimal decision-making re-
quires integration of specific knowledge possessed by different individuals
performing traditionally quite separate corporate functions. One good example is
integration of the specific knowledge of marketing, manufacturing, and R&D
personnel required to design and bring a new product to market. The fairly recent
move to cross-functional teams by many large corporations is a response to such
high information-transfer costs.
While the general applicability of scientific knowledge distinguishes it from
idiosyncratic knowledge, scientific knowledge is also costly to transfer and thus it
too falls in the category of specific knowledge. Science creates order out of chaos
by abstracting from particulars and providing general rules of cause and effect.
Scientific knowledge is an essential ingredient in decisions because it provides the
basis for predicting the outcomes of alternative courses of action.
At the level of the firm, scientific knowledge plays a central role in resolving
the key questions that economists address from a macro or economy-wide per-
spective-notably, what to produce and how to produce it. For example, the de-
sign and development of products from machinery and buildings to household
appliances and drugs depend critically on scientific knowledge.
In addition to scientific and idiosyncratic knowledge, knowledge produced
by assembling and analyzing knowledge of particular circumstances (through time
and/or across circumstances such as location, income, education, age) is a signifi-
22 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

cant input to decision-making. For example, the entrepreneur who wants to capi-
talize on a particular half-filled freighter must be able to identify the freighter, its
location, and its cargo capability. On the other hand, someone deciding whether
to found a business to increase the utilization of freighters will want to assemble
knowledge about how many partially-filled freighters there are, what routes they
follow, what kinds of cargo capacity they have, and so on-knowledge that ab-
stracts from the particular circumstances crucial to making full use of a particular
freighter. Assembled knowledge includes, but is not limited to, that generated by
formal statistical methods.
Assembled knowledge also includes knowledge gained from experience. The
exercise of skills such as machine operation, writing, mathematics, or statistics are
examples. Knowledge of law, of accounting practices, of contracting practices,
and of the rules that govern the operation of organized exchanges are all poten-
tially important inputs to decision-making. Assembled knowledge can be either
general (as is likely to be true of the output of statistical manipulation of basic
data) or specific (as is likely to be true of knowledge from experience).

RIGHTS SYSTEMS
A decision right is the right to decide on and take an action. Decision rights
are the basis for saying that individuals have the “power” to make decisions and
to take actions with resources. Power means that a decision made by a party will
be operative, In modern societies, the ultimate source of this power is the police
powers-the threat of physical violence by the state. An entity or person has the
right to take an action with a specific object if the police powers of the state will
be used to help ensure its ability to take the action. The right to choose what ac-
tion will be taken is an important part of possessing a right. (The word “right” in
this context, incidentally, has no normative content.)
In any developed social system, the right to take actions with specific physi-
cal objects, including our persons, is assigned to specific individuals or organiza-
tions. In a private property capitalist system, most of these rights are assigned to
private individuals or organizations. In a socialist or communist system, most of
these rights are assigned to the state or the governing party.
Although rarely emphasized,’ the usual economic analysis of the price sys-
tem is founded on the existence of a system of privately “owned” rights. There are
two actions of special importance that are an integral part of ownership of a right
in a resource: the right to sell the resource (more accurately, to sell rights in the
resource) and the right to capture the proceeds of the sale.* Thus, the objects of
exchange in markets are not physical articles per se, but bundles of rights attached
to those article^.^ It is this system of alienable rights (almost universally misla-
belled “the price system” in our profession) that extends the efficient utilization of
resources beyond the capacity of any single mind. It provides incentives to make
individuals take appropriate actions without anyone having to direct them.1° This
Specific and General Knowledge, and Organizational Structure 23

is what Adam Smith meant by the “invisible hand,” and his fundamental insight
was that control of human behavior is inherent in the operation of markets.
The assignment of decision-making rights in modern societies is largely a
matter of law.” But once assigned, rights are regularly reshuffled by contracts, by
purchase and sale, and by managerial assignment within firms. In the United
States, the body of law that spells out the assignment of rights is the product of
hundreds of years of lawmaking of three sorts: court decisions (common law),
legislative enactments (statutory law, including constitutions), and administrative
decrees (administrative law).
The private-property capitalist mechanism is the product of thousands of
years of evolution. It is highly complex and embraces a multitude of actions, ob-
jects, and individuals. Most important, however, it functions as a free-standing
system. It is automatic; there is no central direction. With minor exceptions, rights
to take almost all conceivable actions with virtually all physical objects are fixed
on identifiable individuals or firms at every instant of time. The books are kept up
to date despite the burden imposed by dynamic forces such as births and deaths,
dissolutions, and new technology. Disputes arise, but evolution has provided a so-
phisticated arbitration service-namely, the courts-to deal with that problem as
well. The extent to which the legal system enforces property rights (that is, once
again, the security of decision rights and the right to alienate them) is a major de-
terminant of the effectiveness of markets.
The failure of socialist and communist economies (whose distinguishing
characteristic is the absence of private property rights) is now the topic of head-
lines throughout the world. The difficulties that Eastern bloc countries are having
in attempting to establish capitalist market systems to replace their failed systems
is testimony to the complexity and value of market systerns.l2 These economies
provide vivid evidence of the inefficiency and poverty that result from the waste of
specific knowledge and the lack of control in the absence of alienable decision
rights. Without the assignment of private alienable rights, there can be no true
market system. Thus, given the failure of most Eastern bloc countries to establish
alienable private rights in resources, it is not surprising that many of them are fail-
ing in their attempt to create effective market systems.

THE FUNCTIONS OF ALIENABILITY

Alienability is the effective combination of two rights: the right to sell or
transfer rights and the right to capture the proceeds of exchange.13 Alienability is
not only a necessary condition for exchange, it is the foundation of markets and
the institutional device by which markets both (1)collocate knowledge with deci-
sion rights and (2)exercise control over decision makers. The alienability of rights
deserves special attention in analyzing both markets and organizations because
understanding the function of alienability in markets clarifies several critical func-
tions that must be performed inside organizations.
24 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

Alienability Solves the Rights Assignment Problem. When decision rights

are alienable, voluntary exchange creates a process in which the purchase and sale
of rights by maximizing individuals collocates knowledge and decision rights. It
does so by conveying decision rights to the site of knowledge.
In a market system, decision rights are acquired through exchange by those
who have knowledge. Voluntary exchange ensures that decision rights will tend to
be acquired by those who value them most highly, and this will be those who have
specific knowledge and abilities that are most valuable to the exercise of the right.
Alienability Solves the Control Problem. By the word “control” we mean
the process and the rules that govern both the measures of performance used to
evaluate individuals’ actions, and the rewards and punishments meted out to
those individuals as a consequence. Control and knowledge are complements in
the analysis of organizations. The knowledge and decision rights possessed by the
individual, together with the state of the world, define the opportunity set from
which individual decision-makers can choose. The control system plays a major
role in determining which choices individuals make from their opportunity sets.
By collocating decision rights with rights to their capital value, alienability
provides both a measure of performance for individual decision-makers and the
rewards and punishments to motivate them to use those decision rights efficiently.
Market prices for alienable rights reveal the value of assets in alternative uses to
current as well as potential holders of those rights. In cases where resources pro-
duce future flows of revenue or consumption services, and rights to those flows
are alienable, prices represent the present value of claims to those future flows.
These capitalized values perform two important functions in controlling human
behavior.

0 They provide a measure of the performance of the parties who have the
rights to decide how the asset or assets will be used.
They provide the reward or punishment that accrues to the owners of the
rights as a result of their decisions.

The collocation of decision rights with rights to their capital value accom-
plished by alienability thus both measures the performance of individuals and
brings the (capitalized) wealth consequences of an individual’s decisions to bear
upon that person. The decision-maker who chooses an action that lowers the
value of rights assigned to him or her bears the costs of so doing. When the deci-
sion-maker chooses actions that enhance the value of the rights, he or she captures
the increased value.
The major problems with the market control system occur when the legal or
technological environments create “externalities” by not allowing for the defini-
tion and assignment of rights that cause an individual to bear the full costs or to
capture the full rewards of his or her actions. Pollution and non-patentable inven-
tions are two good examples of situations in which decision-makers d o not bear
the full costs or receive the full benefits of their actions.
The problems that arose in organizing production in Eastern bloc countries
without alienability highlight the importance of alienability to issues of organiza-
Specific and General Knowledge, and Organizational Structure 25

tional design and efficiency. But the internal organization of the capitalist firm is also
an example of the suppression of alienable decision rights. Indeed, we distinguish ac-
tivities within the firm from activities between the firm and the rest of the world by
whether alienability is transferred to agents along with the decision rights.
In this view, transfers of decision rights without the right to alienate those
rights are intra-firm transactions. While firms can sell assets, workers in firms
generally d o not receive the rights to alienate their positions or any other assets or
decision rights under their control. They cannot pocket the proceeds. This means
there is no automatic decentralized process that tends to ensure that decision
rights in the firm migrate to the agents who have the specific knowledge relevant
to their exercise. It also means there is no automatic performance measurement
and reward system that motivates agents to use their decision rights in ways that
promote the interests of the organization. Explicit managerial direction and the
creation of mechanisms are required to substitute for alienability.

The Existence of Firms

Pushed to its logical extreme, our focus on specific knowledge implies more
or less complete atomization of the economy. There is no room for the firm. Firms
as we know them would not exist if alienability of all decision rights were granted
to each agent along with the rights. There would be nothing left over for the resid-
ual claimants in the enterprise, be they entrepreneurs, partners, or stockholders.
Firms must obtain advantages from the suppression of alienability that are
large enough to offset the costs associated with its absence, or they could not sur-
vive open competition with independent agents. Such advantages could come
from economies of scale or scope, or from the reduction of transaction costs that
could not be obtained by independent contracting agents.
Knowledge considerations are one reason for the emergence of firms.I4 By
bringing diverse kinds and sources of knowledge to bear on decisions, the exist-
ence of a firm significantly expands the collective opportunity set for all because
no one person is likely to possess the entire set of knowledge relevant to a particu-
lar decision.
In principle, it's true, an entrepreneur could assemble the relevant knowl-
edge by individual exchanges, and knowledge transfer on a quid p r o quo basis is
not an uncommon phenomenon. Consulting and legal services provide obvious
examples of such outsourcing, and so d o the network organizations growing in
the U.S. that contract out most internal functions common to organizations.'s But
where the production, transfer, and application of knowledge are the primary
goods being offered, exchanges tend to take the form of long-term relationships.
The most common of these is employment contracts. Such contracts tend to be
general in nature-the contents of the exchange are not precisely specified-and
they are seldom alienable. Transaction costs are one reason for the prevalence of
such contracts.I6 Single proprietors who contract on a case-by-case basis for pro-
duction and application of all knowledge would soon find themselves swamped
by transaction costs in all but the smallest-scale firms.
26 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

The value of proprietary knowledge to competitors or potential competitors

is another reason for long-term employment relationships. Longer-term contracts
reduce the costs of restricting the flow of valuable knowledge to outsiders. Finally,
longer-run relationships encourage individual participants to invest in firm-specific
knowledge that has little or no value except within the particular organization.
The suppression of alienability, while necessary for the existence of a firm,
does impose costs. Nevertheless, we believe that those costs can be reduced by
thorough understanding and analysis of the functions performed by alienability.
The franchise organization, a rapidly growing sector of the American econ-
omy, is a good example of a mixture of firm and market systems that uses alien-
ability of rights as part of the control system. A franchise contract sells the right to
manage a divisional profit center to a manager for a franchise fee. The manager
receives the capital value right to the residual cash flows, subject to an annual roy-
alty payment and contractual provisions limiting his decision rights in various ar-
eas.” Most important for our purposes, the manager receives the right to alienate
the franchise contract by sale to others. The contract often restricts alienation
rights in various ways-for example, by the right of the franchiser to approve the
purchaser.
The advantage of alienability as a control device is that it rewards and pun-
ishes agents by imposing on them the capitalized value of the future costs and
benefits of their decisions. In the absence of arm’s-length transactions, this is diffi-
cult to accomplish inside a firm. Nevertheless, there are mechanisms for providing
within companies the functions that alienability normally provides in markets. We
turn now to a discussion of these substitute mechanisms and how they help to
solve the organizational problems of the firm.

THE ORGANIZATIONAL PROBLEMS OF THE FIRM:

TRADE-OFFS BETWEEN INFORMATION COSTS AND
AGENCY COSTS
We have seen how alienability solves the rights assignment and control
problems in the economy. Recognizing that firms, by definition, can make rela-
tively little internal use of alienability enables us to see clearly the problems faced
by every firm in constructing substitute mechanisms. The assignment and enforce-
ment of decision rights in organizations are a matter of organizational policy and
practice, not voluntary exchange among agents.
In principle, the modern corporation vests all decision rights in the board of
directors and the chief executive’s offices. Decision rights are partitioned out to in-
dividuals and to organizational units by the rules established by top-level manage-
ment and the board of directors. The chief executive’s office enforces the rules by
rewarding those who follow them and punishing those who violate them. These
assignment and enforcement powers are constrained in important ways by the
laws and regulations of the state and by social custom.
Every CEO, including a benevolent despot with the power to direct the
economy, confronts the rights assignment and control problems of organizational
Specific and General Knowledge, and Organizational Strirctzrre 27

structure discussed above. The limitations of his or her own mental and commu-
nication abilities make it impossible for the CEO to gather the requisite informa-
tion to make every detailed decision personally. Any CEO attempting to d o so in
a large, complex organization will commit major errors. In delegating authority to
maximize survival, the CEO wants to partition the decision rights out among
agents in the organization so as to maximize their aggregate value.
Ideally, as we have seen, assigning decision rights to maximize value means
collocating decision responsibility with the knowledge that is valuable in making
particular decisions. In practice, however, accomplishing such collocation of deci-
sion rights with knowledge within organizations is more difficult than in markets.
In markets, as we have seen, those with the most knowledge tend to acquire the
decision rights by purchasing them. In organizations, by contrast, assigning deci-
sion rights requires consideration of the costs of generating and transferring
knowledge in the organization, and how the assignment of decision rights affects
incentives to acquire information.
Because they are ultimately self-interested, the agents to whom the CEO
delegates authority have objective functions that diverge from his or her own. The
costs resulting from such conflicts of interest in cooperative behavior are com-
monly called “agency costs.” Because agency costs inevitably result from the dele-
gation of decision rights, the CEO must devise a control system (a set of rules)
that fosters desirable behavior.
I t is generally impossible, however, to structure an incentive and control sys-
tem that will cause agents to behave exactly as the CEO wishes. In addition, con-
trol and incentive systems are costly to design and implement. Agency costs are
the sum of the costs of designing, implementing, and maintaining appropriate in-
centive and control systems as well as the residual loss resulting from the difficulty
of solving these problems completely.’8
Figure 3-1 provides an intuitive way to think about the trade-offs associated
with assigning a particular decision right to different levels in the organization’s
hierarchy. The vertical axis measures costs and the horizontal axis measures the
distance of the decision right from the CEO’s office (measured by levels of hierar-
chy) in a simple, hierarchically-structured organization. For simplicity, Figure 3-1
abstracts from the decision regarding where the right is assigned within a given
level of the hierar~hy,’~ and thus deals with the age-old debate over centralization
versus decentralization in organizations.
Determining the optimal level of decentralization requires balancing the
costs of bad decisions due to poor information and those due to inconsistent ob-
jectives. The costs attributable to poor information plotted in Figure 3-1 measure
the costs of acquiring information plus the costs of poor decisions made because it
is too expensive to acquire all relevant information. In the extreme case of a com-
pletely centralized organization (located at the origin on the horizontal axis), the
costs owing to poor information are the high while the agency costs owing to in-
consistent objectives are zero.2o
The costs owing to poor information fall as the CEO delegates the decision
right to lower levels in the organization. They fall because the decision right is ex-
ercised by agents that have more specific knowledge relevant to the decision. We
28 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

Increasing Decentralization >

Decision Right and

Specific Knowledge
are collocated

Distance of Decision Right from CEO’s Office

FIGURE 3-1 Determining the optimal level of decentralization.

assume for simplicity that the hierarchy and both cost functions are continuous.
We assume the costs owing to inconsistent objectives increase monotonically and
at an increasing rate as the right is assigned to lower levels, and that these costs
are conditioned on optimal controls at each alternative rights assignment. We also
assume that the cost owing to poor information has a unique minimum. By defi-
nition this minimum must occur where the right is collocated with the specific
knowledge relevant to the decision.
Total organizational costs plotted in Figure 3-1 are the sum of the costs ow-
ing to poor information and the costs owing to inconsistent objectives. They are
high at the completely centralized allocation and decline as the right is moved
down in the hierarchy to where more relevant specific knowledge is located. In
Figure 3-1 the vertical line marks the optimal location of the decision right. It oc-
curs where the decrease in the cost owing to poor information just offsets the in-
crease in the cost owing to inconsistent objectives (the point where the absolute
values of the slopes of the two curves are equal).
Specific knowledge exists at all levels of the organization, not just at lower
levels. For example, a machine operator often has specific knowledge of a particu-
lar machine’s operating idiosyncrasies, but the chief financial officer is likely to
have the specific knowledge relevant to the capital structure decision. The CEO
may often have the best specific knowledge of the strategic challenges and opportu-
nities facing the firm. The key to efficiency is to assign decision rights to each agent
at each level in such a way that minimizes the sum of the costs owing to poor infor-
mation and the costs owing to inconsistent objectives. Figure 3-1 illustrates that,
even at the optimum, an organization will be making poor decisions due to both
poor information and the conflicts that arise from inconsistent objectives.
Specific and General Knowledge, and Organizational Structure 29

Factors Affecting the Degree of Decentralization

The optimal degree of decentralization depends on factors iike the size of the
organization, information technology (including computers, communications,
and travel), the rate of change in the environment, government regulation, and the
control technology. In general, as the size of a firm increases, the sum of the cost
owing to poor information and the cost owing to inconsistent objectives rises.
When the marginal costs owing to poor information rise more rapidly with size
than the marginal costs owing to inconsistent objectives, the optimal degree of de-
centralization rises.
Changes in information technology have an ambiguous effect on the optimal
degree of decentralization. The direction of the effect depends on which kinds of
information are most affected. When improved technology makes it easier to
transfer specific knowledge effectively from lower to higher levels in the organiza-
tion, there will be a shift toward centralization. Mrs. Fields Cookies is an example
of a firm where technological development made it possible for headquarters to
obtain detailed and timely information on store operations and to provide very
detailed day-by-day, even hour-by-hour, directions on operating decisions in its
company-owned stores.21
Conversely, when improved technology makes it easier to transfer to lower
levels in the organization information that was formerly specific to higher levels in
the organization, there will be a shift toward decentralization. J. C. Penney’s in-
vestment in satellite communications provided the firm with closed circuit TV
that made it possible to decentralize much of the store purchasing decisions from
corporate headquarters to the local store managers. The TV system made it possi-
ble for central buyers in New York to display and “market” the goods to local
store managers, who could then use their specific knowledge of local tastes and
fashions in stocking their stores.22
Increased governmental regulation tends to increase centralization. It does
so by increasing the amount of specific knowledge in the headquarters office deal-
ing with the regulatory agency. Improvements in control technology-such as
communication and measurement techniques that reduce the marginal agency
costs associated with delegating decision rights-will tend to increase decentrali-
zation in an organization.
Our characterization of decision rights so far has been overly simple. It is
relatively uncommon in large organizations for agents to have the total rights to
make any major decision in the way we normally think about decisions. Instead,
decisions are normally made by a process in which decision management and de-
cision control rights are assigned to different individuals within the firm. Decision
management rights are the rights to initiate and implement recommendations for
resource allocations. Decision control rights are the rights to ratify initiatives and
to monitor the implementation of resource ~ o m m i t m e n t s Although
.~~ we d o not
have space to pursue the issue here, the analysis portrayed in Figure 3-1 can be ap-
plied to the assignment of both decision management and decision control rights.
For example, when the relevant specific knowledge for decision control lies at a
30 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

lower level in the organization (such as the knowledge that would be used in the
performance evaluation and bonus-setting process for lower-level managers),
some decentralization of control rights is optimal.
In sum, the CEO in the typical firm cannot generally use alienability to solve
the firm’s organizational problems. He cannot delegate the alienability of decision
rights to decision agents without thereby converting them into independent firms.
Organizational problems within the firm must therefore be solved by substitute
means. This is accomplished by devising a set of internal “rules of the game” for
the firm that:

partition out the decision-making rights to agents throughout the organi-

zation;
create a control system that
provides measures of performance;
specifies the relationship between rewards and punishments and the
measures of performance.

This is a simple but remarkably powerful list. While there are many factors
that determine the behavior of any individual organization, our empirical obser-
vations indicate that knowledge of these “rules of the game” enables one to make
good predictions about an organization’s behavior and effectiveness. We now con-
sider common organizational devices for implementing these organizational rules
of the game.

THE TECHNOLOGY FOR PARTITIONING DECISION

RIGHTS IN THE FIRM
The techniques available for structuring activities within the firm are a prod-
uct of evolution, as is the system of rights for the economy as a whole. What has
evolved is a complex body of managerial technology that is employed in assigning
decision rights and in controlling behavior within the firm. Scientific under-
standing of that technology is rudimentary, but we can describe some of its major
components and their use.

Job Descriptions and Internal Common Law

Decision rights are allocated to agents within firms in various ways. Many
are allocated directly to individuals or positions through job descriptions (and
these descriptions are often the best source of written documentation of the as-
signment of decision rights in an organization). Examples include the right to
make pricing, hiring, or promotion decisions, the rights to initiate recommenda-
tions for resource allocation, to ratify or monitor the initiatives of others, or to
implement particular programs.24The allocations of decision rights to individuals
Specific and General Knowledge, and Organizational Structure 31

change over time as the organization and individuals change. These rights assign-
ments occur both formally and informally, and are associated with committee
memberships and project assignments as well as the organization’s internal “regu-
latory’’ and “common law” traditions.

Budgeting

Physical and monetary budgets are common techniques for partitioning de-
cision rights in firms. Agents can be given decision rights over the use of physical
resources, such as capital equipment or building space. The rights allocated
through such physical budgets are less complete and therefore more constraining
than are decision rights allocated by the grant of monetary budgets. Dollar budget
authorizations tend to be used when the intent is to grant some discretion in the
choice of inputs. When rights are allocated through monetary budgets without
side constraints, decision agents have the opportunity to sell or exchange assets,
and therefore to substitute among them. The organization is better off to the ex-
tent that managers use their specific knowledge to make substitutions that in-
crease the efficiency of the organization.
Nevertheless, budgets denominated in money terms are frequently con-
strained in ways that deny managers the opportunity to substitute. These ‘‘line’’
budgets, which are commonly used in government as well as industry, are broken
down in great detail and the recipient is specifically forbidden from transferring
funds from one category to another. Under such budgets the manager’s ability to
use his or her specific knowledge to increase efficiency is obviously restricted.
Such restrictions can be optimal if the specific knowledge relevant to making these
substitutions lies a t a higher level in the organization. This occurs, for example,
when there are external effects on other parts of the organization that cannot be
incorporated in the manager’s performance measure, but can be incorporated in
the performance measure at a higher level of the organization.
Budgets can be fixed or variable. They are fixed if the amount of authorized
spending is independent of the level of activity or of performance. Under a vari-
able performance budget, spending authority is a specified function of perform-
ance or activity levels-for example a fraction of revenues. (The “each tub on its
own bottom” budgeting systems of some universities are examples of variable
performance-related budgets.) While variable budget allocations have substantial
incentive effects (because most agents prefer to have control over more resources),
these incentive effects often seem to be ignored in practice.
Budgets are usually accompanied by side constraints. Physical resource
budgets, for example, are commonly restricted to use rights; that is, the recipient
is not allowed to sell the resources and retain the proceeds. Diversion of dollars or
physical resources to personal use (except that specified as compensation) is also
prohibited. Manpower or head count limitations that are independent of the dol-
lars available are another example of a separate constraint.
32 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

Rules, Regulations or Fiat

The rules and regulations that accompany budgets are examples of regula-
tory constraints on behavior that exist because employees are self-interested. Such
constraints imposed by fiat are the most primitive form of control technology.
Like line budgets, they control behavior by circumscribing in advance the oppor-
tunity set from which a decision-maker can choose. But unless the regulator is om-
niscient, such rules will eliminate superior as well as inferior courses of action
because they are made without the specific knowledge that lies at the local level.
In this sense, control by regulation tends to disregard the advantage of collocating
knowledge and decision rights at the local level. Regulations are efficient control
devices when the budget office has the relevant specific knowledge or where the
prohibited behavior-for example, theft or embezzlement-is clearly not consis-
tent with the objectives of the CEO.

THE CONTROL SYSTEM

Because all individuals in a firm are self-interested, simply delegating deci-
sion rights to them and dictating the objective function each is to maximize is not
sufficient to accomplish the objective. A control system that ties the individual’s
interest more closely to that of the organization is required. The control system
specifies (1)the performance measurement and evaluation system for each subdi-
vision of the firm and each decision agent, and (2) the reward and punishment
system that relates individuals’ rewards to their performance.
In a real sense, specification of the performance measurement and evalu-
ation system is specification of the objective function, although a surprising num-
ber of organizations seem to fail to recognize this point. Self-interest motivates
individuals to discover and understand the performance measures and evaluation
system on which their rewards and punishments depend. It does not take them
long to discover when the rewarded objective is different from that which is
stated.

Cost Centers and Profit Centers as Performance Measurement

Systems

Cost centers and profit centers embody two widely used divisional perform-
ance measurement rules. Cost centers are subdivisions that are directed to mini-
mize the total cost of providing a specified quantity of service. Manufacturing
divisions are frequently organized as cost centers. Mathematically (and in the ab-
sence of information or agency problems), minimizing total cost for a given quan-
tity of output is equivalent to maximizing output for a given total cost. In
addition, both are consistent with maximizing the value of the firm if the correct
output constraint is chosen.
Specific and General Knowledge, and Organizational Structure 33

In the presence of information and agency problems, however, the two for-
mulations are not equivalent. Minimizing cost for given total output often seems
to degenerate into a system where managers are rewarded for minimizing average
cost per unit of output. And, in the absence of a quantity constraint, measuring
performance by average cost per unit of output will virtually never be consistent
with firm value maximization. A decision manager with such an objective will
strive to achieve the output quantity that minimizes average cost even though it
bears no relation to the value-maximizing quantity.
The tendency of firms to divisionalize along product lines appears to be in-
fluenced by control considerations. Product subdivisions are often operated as
profit centers, where the measure of performance is the difference between some
measure of revenues and costs. Profit centers are more independent than cost cen-
ters: their budgets are more likely to be variable than those of cost centers, and
this generally means fewer knowledge demands on the CEO. The scale of opera-
tions of the center then varies directly with revenues, and does not require the
same forecasting accuracy as a fixed-dollar budget would require.
The reduction in knowledge required to monitor a division organized as a
profit center is particularly evident where the products are sold in outside mar-
kets. Here the CEO can use competition in outside markets as a part of the control
system. Competition and the ability of the division’s customers to purchase from
others provide the CEO with a performance measure for the product division-
namely, profits-that incorporates consumers’ assessment of quality, timeliness,
and value. Internal transfer pricing systems in which buyers have the right to pur-
chase from any source also allow the CEO to decentralize to the buyers an impor-
tant part of the control system. Such decentralization is optimal to the extent that
specific knowledge of product and service quality lies with the buyers and is costly
to observe from higher in the hierarchy.
But neither profit centers nor cost centers are panaceas for the CEO’s organ-
izational problems. Cost centers, for example, tend to lead to problems of quan-
tity and quality control. Measured on the cost of output for a fixed quantity,
division managers are motivated to reduce cost by reducing quality. Preventing
this requires quality to be cheaply observable from higher in the hierarchy. To the
extent that quality is easily observable, cost centers will tend to be more desirable.
Divisions where quantity is difficult or impossible to measure (such as computer
services) are difficult to run as cost centers because the manager can simply reduce
the quantity of service to lower cost.

The Role of Budgets in Performance Measurement

Budgets are related to performance measurement in several ways. Budgets

are sometimes used to delegate decision rights, but they are also used as targets in
the performance measurement system-for example, as expenditure or revenue
targets. In these cases, the amount by which expenditures are less than the targets
and by which revenues exceed targets are favorable performance measures.
34 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

We have two major points about the use of budgets in performance evalu-
ation. The first is fairly straightforward: When budgets are used to delegate deci-
sion rights, measures of violations of budgeted expenditures must be part of
performance measurement if expenditure limits are to have meaning. Indeed, vio-
lations of any rules, regulations, or fiat must affect performance measures and re-
wards and punishments if the constraints are to affect behavior.
Our second point is that the use of some kinds of budgets can cause major
problems for large organizations. Take the case of the budget-target system
known as strategic business pfunning. In this widely-used system, performance is
measured by how close the results are to a plan whose targets are typically “nego-
tiated” between corporate headquarters and division heads.
Such a budget-target system poses problems because its success depends
critically on setting correc‘i plans or targets for each division and decision agent.
This in turn imposes enormous knowledge requirements on the central staff that
must d o the planning. When much of the required specific knowledge is located at
lower levels in the organization and involves high cost to transfer to the central
planning staff, strategic business planning will be inefficient. When such knowl-
edge is important, the result of centrally devised targets will be poor plans and
strategic business planning will generate large organizational costs.
In short, strategic business planning is the private organizational version of
central planning in the market system. And much as central planning has failed in
most countries, the practice of strategic business planning (at least as defined as a
system that measures performance against pre-set targets) has contributed to the
failure of many large American corporations over the past two decades.25

Measuring, Rewarding, and Punishing Individual Performance

The performance measurements discussed previously are all grolrp meas-

ures. But the CEO’s measurement problem is not simply one of measuring group
performance. In the end, he or she must reward and punish individuals.
For a sizable organization, the CEO cannot literally either review the per-
formance of every individual or decide on his or her specific rewards. Inevitably,
the CEO will delegate much of the responsibility for measuring and rewarding
performance and will promulgate rules or policies that control the decisions of
those to whom authority is delegated. The CEO can, for example, tie individual
rewards to individual performance by direct pay-for-performance systems (and
here the sensitivity of the relation between pay and performance is a major deci-
sion variable), or by promotions that depend on performance. Individual rewards
can be tied to group performance by creating bonus pools that are a function of
group performance or by setting up profit-sharing plans, employees stock owner-
ship plans, stock option plans, or phantom stock plans.
We have observed a longstanding tendency for large organizations to avoid
pay-for-performance incentive plans and to rely instead on promotion-based re-
wards.26Although this phenomenon is as yet poorly understood by economists,
Specific a n d General Knowledge, and Organizational Striictrire 35

o u r belief is that promotion-based systems will turn o u t to be a relic of an older

era of centralization and technological stability. Factors such as more rapid tech-
nological change and more intense global competition-when coupled with the
long-term tendency of most organizations to become even larger (even as the need
increases for downsizing and exit in maturing industries)-have made decentrali-
zation of corporate decision-making a more valuable strategy in the past decade
or so. And as decentralization of corporate decision rights becomes more valu-
able, our prediction is that many firms will choose (or be forced) to replace their
promotion-based systems with significantly greater use of the incentives held out
by profit-sharing and stock ownership.

CONCLUSIONS
This paper analyzes the relations between knowledge, control, and organ-
izational structure, both in the market system as a whole and in private organiza-
tions. The limited capacity of the human mind and the costs of producing and
transferring knowledge mean that knowledge relevant to all decisions can never
be located in a single individual or body of experts. Thus, if knowledge valuable
to a particular decision is to be used in making that decision, there must be a sys-
tem for assigning decision rights to individuals who have the knowledge and abili-
ties or who can acquire or produce them a t low cost. In addition, self-interest on
the part of individual decision-makers means that a control system is required to
motivate individuals to use their specific knowledge and decision rights properly.
The rights assignment and control problems are solved in a capitalist econ-
omy by a system of voluntary exchange founded on a system of alienable decision
rights. Voluntary exchange of alienable decision rights tends to ensure that those
agents with the relevant knowledge and abilities will place the greatest value on a
decision right, and will therefore acquire it. This solves the rights assignment
problem of collocating decision rights and specific knowledge.
In the absence of externalities, alienable decision rights also solve the control
problem; they motivate individual decision agents to use their decision rights effi-
ciently. Alienability does this by providing an effective system in which the market
price or capital value of the right measures the effectiveness with which any indi-
vidual uses a decision right. Alienability also means that the individual can cap-
ture the value of the right in exchange. In this sense, alienability provides an
effective reward and punishment system that places the capitalized value of the
costs and benefits of an individual’s actions on his or her own shoulders.

Why Firms Are Different from Markets

Alienable rights cannot generally solve the control problem inside firms be-
cause firms cannot generally assign alienability along with the decision rights
without turning each individual agent into an independent firm. Indeed, the char-
36 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

acteristic that distinguishes such organizations from markets is the fact that
alienability of the rights is not delegated to individual decision agents in the
organization.
Because of the limited computational capacity, storage, and input/output
channels of the human mind, it is often desirable for groups of individuals to ex-
ercise decision rights jointly. Private organizations are widespread examples of
such joint exercise of decision rights. In such organizations, independent individu-
als coordinate their actions through contracts with the legal fiction that serves as
the firm’s nexus. The bundle of decision rights owned in the name of such an or-
ganization is vested nominally in its board of directors and CEO, and the rights
are then partitioned out among decision agents in the organization. Those organi-
zations that accomplish this partitioning in a fashion that maximizes their value
will tend to win out in the competition for survival.
The inalienability of decision rights within an organization means that the
exchange mechanisms that serve to collocate decision rights with the relevant
knowledge and skill are not operative. Furthermore, the inalienability of rights
within an organization means that the control problems must be solved by alter-
native means. Organizations solve these problems by establishing internal rules of
the game that provide:

a system for partitioning decision rights out to agents in the organization;

and
a control system that provides:
a performance measurement and evaluation system; and
a reward and punishment system.

In general, because of their inability to simulate true capital value claims,

these substitute organization “rules of the game” will not perform as effectively as
alienable rights in a market system. Therefore, survival requires that the firm must
realize benefits from the joint exercise of rights that are large enough to offset the
disadvantages incurred by sacrificing alienability. Economies of scale and scope,
information advantages, and specialization are potential sources of such benefits.
The creation of a science of organization is still in its infancy. We believe that
the structure outlined in this paper provides a view of organization that yields im-
portant insights for both social scientists and managers. Knowledge of an organi-
zation’s rules of the game, along with a surprisingly small amount about its
technology or opportunity set, enables one to make accurate predictions about its
behavior.
Specific and General Knowledge, and Organizationnl Structure 37

REFERENCES
1. See, for example, Ronald H. Coase, “The Problem of Social Cost,” ]ournal of Law
and Economics, 3, (1960),pp. 1-44; and Armen Alchlan and William Allen, Exchange
and Production Competition, Coordination & Control (Wadsworth, Belmont, CA.,
1983), p. 91.
2. E A. Hayek, “The Use of Knowledge in Society,” American Economic Review, N o 35
(September, 1945), pp. 1-18.
3. This principle is also recognized in Milton Harris, C. H. Kriebel, and Artur Raviv,
“Asymmetric Information, Incentives and Intrafirm Resource Allocation.” Manage-
ment Science, No. 28, (June, 1982), pp. 604-620.
4. Hayek (1945).
5. A condition described as “bounded rationality” by J. March and Herbert Simon in
their book. Organizations (New York: John Wiley and Sons, Inc. 1958). See also H. A.
Simon, “A Behavioral Model of Rational Choice,” Quarterly Journul of Economics,
No. 69 (1955), pp. 99-118; and H. A. Simon, “Theories of Decision Making in Eco-
nomics and Behavioral Science,” American Economic Review, Vol XLIX, No. 3 (June,
1959), pp. 253-283.
6. Like Hayek, economists have generally taken the costs of information transfer to be
prohibitively large, and, therefore, taken the distribution of knowledge as given. They
have analyzed extensively the effects of “information asymmetry” (as it is known in
the principal/agent literature) on contracting relations. In his study of institutions,
Oliver Williamson defines the concept of “information impactedness” to deal with the
organizational implications of transactions where information is “known to one or
more parties but cannot be costlessly discerned by or displayed for others.” (See Oliver
E. Williamson, Markets and Hierarchies: Analysis and Antitrust Implications (New
York: The Free Press, 1975), p. 31. Explicitly recognizing the costs of transferring
knowledge is more useful analytically.
7. A notable exception to this generalization is Alchian and Allen (1983, and earlier edi-
tions dating back to 1969).
8. Including the right to sell the rights in output that an individual or firm creates with
the resource.
9. It follows that the values established in exchanges are values of bundles of rights, not
prices of physical objects. Property whose use is restricted by regulatory constraints or
private covenants will sell at different prices from identical property with full-use
rights. Goods are sometimes alienated illegally, e.g. theft, black markets, drugs and
prostitution. When the police powers are not 100% effective, rights are not 100% se-
cure, and the lower value of such rights will reflect the probability that the rights will
be taken (either illegally, or legally through political action such as confiscation or na-
tionalization).
10. In the absence of externalities or monopoly, of course. But externalities are themselves
a result of an incomplete definition and assignment of rights. See Coase (1960).
11. Customs and mores not embodied in law also confer decision-making powers and
constraints on individuals or groups, especially in primitive societies. The social sanc-
tions imposed on those who take actions in violation of social or group norms can
have substantial impact on the decision rights of individuals-an impact that is dis-
tinct from that of the formal legal sanctions of the state. Alternatively, individuals
38 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

sometimes possess decision-making powers without having legal rights in those re-
sources, e.g., possessors of stolen goods. Those engaged in illegal activities themselves
employ threats of physical violence to preserve powers.
12. For an excellent survey, see Clive Crook, “Perestroika: And Now for the Hard Part.
The Economist (April 28, 1990), pp. 1-22.
13. Alienability includes the right to sell or transfer alienability itself.
14. Indeed, one economist has argued that “conservation of expenditures on knowledge”
determines the vertical boundaries of the firm. See Harold Demsetz “The Theory of the
Firm Revisited,”]ournul of Law, Economics, & Organization, No. 4 (1988),pp. 159.
15. See John W Kensinger and John D. Martin, “Financing Network Organizations,”
]ournu1 of Applied Corporate Finance, Vol. 4 No. 1 (Spring, 1991), pp. 66-76.
16. This point is emphasized by Ronald H. Coase in “The Nature of the Firm,”
Economica (1937).New Series, IV, 386-405. See also Williamson (1975).
17. For a description and analysis of the nature of the franchise contract, see Paul H. Ru-
bin, “The Theory of the Firm and the Structure of the Franchise Contract,” Iournal of
Law and Economics, No. 21 (April, 1978), pp. 223-233. Like so much of the litera-
ture on franchises, this analysis ignores the critical role of alienability in the function-
ing of this organizational form.
18. For our original formulation of agency theory, see Michael C. Jensen and William H.
Meckling, “Theory of the Firm: Managerial Behavior, Agency Costs and Ownership
Structure,” Journal of Financial Economics, No. 3 (1976) pp. 305-60.
19. We can assume for simplicity that the right is optimally assigned within each level in
the hierarchy.
20. Assuming the CEO does not have agency problems with himself or herself. See Rich-
ard H. Thaler and H. M. Schefrin, “An Economic Theory of Self-Control,” ]ournu1 of
Political Economy, Vol. 89, No. 2 (April, 1981), pp. 392-406.
21. See Tom Richman, “Mrs. Fields’ Secret Ingredient,” Itzc. Magazine, (October, 1987).
22. See Hank Gilman, “J. C. Penney Decentralizes its Purchasing: Individual Stores Can
Tailor Buying to Needs,” Wall Street]ournal, May 8, 1987.
23. See Eugene E Fama, and Michael C. Jensen, (1983a). “Agency Problems and Residual
Claims,” lournal of Law and Economics, No. 26 (June 1983), pp. 327-349; and
Eugene E Fama and Michael C. Jensen, (1983b), “Separation of Ownership and Con-
tro1,~’]ournalof Law and Economics, No. 26. pp. 301-325.
24. For further discussion of the breakdown of the decision process into initiation, ratifi-
cation, implementation, and monitoring rights, see Fama and Jensen (1983b).
25. See Walter Kiechel, “Corporate Strategists Under Fire,” Fortune (December 27, 1982);
and Robert H. Hayes, “Strategic Planning-Forward in Reverse?,” Harvurd Business
Review, Vol. 63, No. 6 (November-December, 1985), pp. 111-119.
26. See George P. Baker, Michael C. Jensen, and Kevin J. Murphy, “Compensation and In-
centives: Practice vs. Theory,”]ournal of Finance, Vol. 43, No. 3 (July, 1988), pp. 593-
616; and Kevin J. Murphy, “Performance Measurement and Appraisal: Motivating
Managers to Identify and Reward Performance,” in William J. Bruns, Jr., ed., Perfom-
ance Measurement, Evaluation, and Incentives, Harvard Business School Press, 1992.
 Bureaucracy
Gifford & Elizabeth Pinchot

Given that bureaucracy is in such ill repute today, it is hard to remember that it
once was considered a great organizational innovation. By organizing the division
of labor, by making management and decision making a profession, and by pro-
viding an order and a set of rules that allowed many different kinds of specialists
to work in coordination toward a common end, bureaucracy greatly extended the
breadth and depth of intelligence that organizations could achieve. Begun as a sys-
tem of organizing government activities, it has spread to big businesses and large
organizations of all kinds.
Max Weber, who launched the systematic study of bureaucracy as its role in
Western society began to explode in the late nineteenth century, saw bureaucracy
as both the most efficient possible system and a threat to the basic liberties he held
dear. Weber predicted the triumph of bureaucracy because of its greater efficiency:
“The purely bureaucratic form of administrative organization, that is the mono-
cratic variety of bureaucracy, is, as regards the precision, constancy, stringency
and reliability of its operations, superior to all other forms of administrative
organization. ”’
Weber would have been surprised (even frightened) by how accurate his pre-
diction of bureaucracy’s triumph has proven. During the last hundred years, the
landscape of society has changed dramatically as large bureaucratic organizations
replaced small family enterprises in retailing, manufacturing, and services. Many
not-for-profits, from Blue Cross to the Audubon Society, have adopted the bu-
reaucratic form. Even family entrepreneurship has taken a step toward bureauc-
racy with the shift from hosting a hometown diner to owning a franchise.
Bureaucracy created a system capable of effectively managing the massive
investments, division of labor, and large-scale mechanized production of capital-
ism. Its organizational power drove the initial rapid growth of the steel, chemical,
and automobile industries. Bureaucracy united AT&T as it established a peerless
national communication network with rank on rank of managers structured by

Reprinted with permission of the publisher. From Chapter 2 of The End of Bureaucracy & t h e Rise of
the Intelligent Organization, copyright 0 1993 by Elizabeth & Gifford Pinchot, Berrett-Koehler Pub-
lishers, Inc., San Francisco, CA. All rights reserved.
39
40 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

the Bell System Practices-a set of policy manuals that provided detailed and ex-
plicit instructions for every task. IBM added customer focus to bureaucracy and
created an organization effective enough to give it forty years of preeminence in
the new computer industry.
Despite all these successes, respect for bureaucracy is declining. As in so
many other areas of life, what brought great success in the past has become the
limitation of today. Suddenly everyone knows that bureaucracy is slowing us
down and keeping our organizations internally focused and uncreative. It is time
to question bureaucracy. What is the basis of its success? Why is it suddenly
less useful than it was? What can we d o about it? What are the alternatives to
bureaucracy?

WHAT BUREAUCRACY IS AND WHY IT CONQUERED

ALL
Bureaucracy gained preeminence because it worked for many of the needs of
the industrial age. It increased the effectiveness of hierarchy by reducing some of
the worst abuses of power and by providing a rational way to manage tasks too
complex for any one person to comprehend. Let us look more closely at why it
worked so well. There is consensus among social scientists that the six charac-
teristics of bureaucracy, all part of Weber’s original description, are roughly as
follows:

A hierarchical chain of command

Specialization by function
Uniform policies covering rights and duties
Standardized procedures for each job
A career based on promotions for technical competence
Impersonal relations2

To this list we add an operating principle of bureaucracy suggested by Fred

Emery:

All coordination done from a level or more above the work being coordinated3

A Hierarchical Chain of Command

The bureaucratic organization is structured as a pyramid with an absolute

boss on top who divides up the overall task of the organization and gives respon-
sibility for each subtask to subbosses who divide responsibility yet more finely
and so on through an unbroken chain of sub-subbosses that stretches down to
every employee. In the 1980s, huge organizations such as General Motors, Sears,
IBM, and the U.S. government had as many as twelve layers of management be-
tween the CEO and the worker-too many, as it has turned out.
The Rise and Fall of Bureaticracy 41

The establishment of a clear chain of command was a powerful way to bring

order to large groups in a common enterprise. The chain of command resolved
potential conflicts by granting clear responsibility, authority, and accountability
for each potential decision. Each boss and subboss in the chain of command was
given an absolute monopoly of power over a task or function and then held ac-
countable for it. This greatly simplified the boss’s task of making sure the organi-
zation executed commands.
Limitations of Prehureaucratic Autocracy. Autocratic organizations with-
out a clear chain of command run out of steam at about a hundred persons. Many
entrepreneurs fall into this trap by assuming the role of a “craftsman entrepre-
neur,” a person who maintains control of a growing organization like a fine
craftsman with many assistants. Rather than establish an effective chain of com-
mand, he or she tries to be everywhere making all decisions throughout the or-
ganization for a big group of helpers. The result is the classic growth curve of the

AA DIGGING
DEPARTMENT
PAPER SHUFFLING
DEPARTMENT
METER READING
DEPARTMENT

FIGURE 4-1 Bureaucracy.

42 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

Time

FIGURE 4-2 The craftsman entrepreneur.

craftsman entrepreneur: rocky start, smooth expansions, rocky leveling out as if

bumping against a ceiling.
Such entrepreneurs might explain all the ups and downs of their business as
changes in the market, but in fact they are suffering from the limitations of the
prebureaucratic form of autocracy. To the extent the entrepreneur continues to be
an absolute ruler, he or she cannot go beyond the scope of business that can be
understood and managed by a single omnipresent manager. Some entrepreneurs
caught in this craftsman role appoint subordinate managers, but by continually
countermanding the orders of those managers, they fail to respect the chain of
command they establish and thus disempower their managers.
Delegation and Empowerment. Those entrepreneurs who succeed in ex-
pansion commonly introduce a chain-of-command structure, which by its nature
delegates power and increases the thinking power of the organization by empow-
ering more brains to take action. The business may grow when supervisors and
middle managers are empowered in a limited but significant way to make deci-
sions about their areas and to establish procedures and issue orders. Postbureau-
cratic entrepreneurships are growing just fine with decentralized teams and lateral
networking taking the place of chain of command. Nonetheless, the innovations
of bureaucracy, including the divestiture of some of an owner’s power to a hierar-
chy beneath, served the goal of growth in earlier eras.

Specialization by Function

Bureaucracy achieves efficiency through specialization of labor. In fact, the

organizational structure of a bureaucracy is created by dividing the overall task
The Rise and Fall of Bureaucracy 43

into a series of well-defined specialties or functions. Each function is given respon-

sibility for a defined set of tasks and given the tools needed to accomplish that
task. The boss gives orders and assigns tasks in such a way that all the parts add
up to a coherent whole.
With specialization, different varieties of engineers study exactly why effi-
ciency is lost in each of many steps in the production process and then design
equipment and procedures to raise yields. Salespeople perfect their selling skills,
and financial professionals manage the liquidity and profitability of the business
with increasingly sophisticated tools. In general, specialization leads to more ef-
fective ways of doing each aspect of the organization’s overall task.
Before the specialization of bureaucracy, each craftsperson learned all of
blacksmithing or all of barrel making and performed all aspects of the job from
start to finish. Craft production can often be satisfying and have artistic merit, but
in the Industrial Revolution it worked against the mechanization and economies
of scale that specialization and division of labor made possible. As organizations
moved from craft production to division of labor, the strict hierarchy of bureauc-
racy provided the clout to set aside the traditions and concerns of craftsmen and
to make each new innovation part of the rules and procedures of the organization.
Specialization can contribute to organizational intelligence by allowing peo-
ple to concentrate on each little aspect of what the organization does. With many
specialists, each good at his or her special area, the organization can bring great
intellectual pressure and ingenuity to bear on each of the many different aspects of
the business.

Uniform Written Rules and Policies

A bureaucracy is governed by uniform written rules and policies that in a

corporation, profit or not-for-profit, are set by the board and the management.
These rules define the rights and duties of employees and managers. The most ba-
sic rules concern who can give orders to whom.
In a bureaucracy, the boss is responsible for the actions of all the people un-
der him or her and has the right to give them orders that they must dutifully obey.
The employee’s primary responsibility is not to do what is right or what needs to
be done but only to follow exactly the orders of his or her immediate boss.
The written policies of a bureaucracy also guarantee employees regular
wages as long as they are employed and, in some cases, even a pension for long-
term service. These forms of compensation are quite different from those of the
feudal systems, in which each level in the hierarchy, from serf on up to the local
lord, often took a piece of the action, however paltry, in their domain.
Written rules concerning rights and duties partially offset some of the worst
aspects of chains of command by reducing the potential power of any petty ty-
rants at the supervisory and middle management levels. Supervisors disciplining
employees have more precisely defined powers, which both empower them to a
degree and limit arbitrary behavior.
44 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

Standardized Procedures Defining Each Job

In a bureaucracy, fixed procedures govern how employees are to perform

their tasks, sometimes to an astonishing degree. Frederick Taylor, an engineer who
became known as the father of scientific management for his work in the early
part of this century, recorded and then taught the exact motions of the most pro-
ductive workers in a factory so that everyone else doing that task could make the
same motion^.^ This reliance on established procedures is in stark contrast to the
system of “make it up as you go along,” characteristic of an entrepreneurial start-
up or, a t worst, the arbitrary personal whims of a feudal lord and his powerful
minions.
Uniform rules and procedures written down and stored as official docu-
ments increased the intelligence expressed in organizations by instituting a crude
“memory” of lessons learned. Written rules and procedures extended the power
of the commands, standardizing the actions to be learned through frequent turn-
over of employees. Change could be accommodated if it could be written down
and not bump into an existing rule. Standardized procedures could serve to make
lessons learned in one part of the organization more broadly effective and to over-
come irrational resistance to more effective ways of doing things.

The Professional Career

Success in the bureaucratic organization is defined as a lifetime career of ad-
vancing to higher levels in the chain of command. Rising in the ranks provides
both power and symbols of status. Promotion is achieved through technical com-
petence in one’s specialty and efficiency in carrying out orders.
The professional career provides a “contract” between employee and or-
ganization: In its simplest form, a person devotes him- or herself to the organiza-
tion in exchange for secure work and wages. The full-time professional manager
was married to the organization for life. In return, the organization promised a
stable or rising salary, a pension, lifetime employment, and a chance to rise in the
hierarchy.
Before bureaucracy, favoritism and nepotism destroyed the efficiency of or-
ganizations more than they do today. Even today, there are many cultures in
which a boss confronted with a choice between promoting an incompetent rela-
tive or another employee has no culturally acceptable alternative but to opt for
the incompetent relative. In its ideal form, bureaucracy subordinates these family
loyalties and other sympathies to the goals of the organization through a policy of
promotion for measurable technical competence. In a government bureaucracy, in
the civil service and the police and fire departments, for example, this policy is
often manifest in exams that are prerequisites for moving to higher level positions.
The promise of a good bureaucratic career allowed organizations to recruit, train,
and retain highly skilled specialists.
The Rise and Fall of Bureaucracy 45

The lure of rising in the hierarchy and the security of a professional career
was an important element in bureaucracy’s success, providing a strong motivation
for long-term loyalty to the organization. Yet most will not make it in a bureauc-
racy, since the only success is moving upward. This carries the seeds of disappoint-
ment later in one’s career when the pyramid has narrowed and only a few can
move up to the next level.

Impersonal Relations

In a bureaucracy, relationships are from role to role rather than from person
to person. The organizational structure and job description define what is ex-
pected of an individual in each role, and the holder of a particular role is expected
to carry out its responsibilities in a rational and unemotional manner. Therefore,
emotions are not to be displayed: The coolly analytical win, and the open and car-
ing lose.
Impersonal relations helped move bureaucracy beyond nepotism and favor-
itism by preventing family feeling or friendship from getting in the way of enforc-
ing rules and making tough decisions. It kept managers’ sentiments from getting
in the way as they wrenched workers away from the satisfactions of craft produc-
tion and toward the bureaucratic routines and unthinking work of the assembly
line.

All Coordination from a Level or More Above

In a bureaucracy, workers do not figure out how to coordinate their work

with their peers. The boss divides up the work and defines each person’s job so
that added together those jobs produce the output that is the boss’s responsibility
to manage. The boss’s boss then provides coordination between units, and the
units, therefore, do not need to coordinate with one another. All coordination
must rise up and pass through the next higher boss.
Employees are not paid to think broadly; their job is to stay within the
“boxes” defined by their job descriptions and the standardized procedures. Above
all, they are forbidden to coordinate with their peers, who are either subordinates
of the same boss or of another one elsewhere i n the organization. To do so would
rob the bosses of their authority.
Coordination from above worked well during the early Industrial Revolu-
tion when huge numbers of employees unskilled in the mechanical arts had to be
quickly fitted to a job in the “satanic mill^."^ Turnover could be extraordinary.
Near the turn of the century, Ford Motor Company’s Highland Park, Michigan,
plant had to hire fifty-four thousand workers a year just to keep thirteen thousand
working6 With such rapid turnover, workers had little time to understand the
whole and needed a simple and clearly defined assignment.
46 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

WHY BUREAUCRACY N O LONGER WORKS

The world no longer needs the machinelike organizations bureaucracy pro-

duces. The challenges of o u r times call for lively, intelligent organizations. Bu-
reaucracy was efficient for certain kinds of repetitive tasks that characterized the
early Industrial Revolution. It no longer works so well, because its rules and pro-
cedures are often diametrically opposed to the principles needed for workers to
take the next step toward greater organizational intelligence. These principles in-
clude more responsibility to define and direct one’s own job, more responsibility
to coordinate with others, and a shift in authority from one’s boss to one’s
“customers.”

From Unskilled Work to Knowledge Work

Peter Drucker has been telling us for decades that more and more of work,
both technical and nontechnical, is knowledge-based. We no longer need many
unskilled assembly-line workers; most of the jobs in factories involve technical
knowledge and training. What is more, few of the jobs in a manufacturing organi-
zation are in the factory. Most “manufacturing” jobs are in functions such as
marketing, design, process engineering, technical analysis, accounting, and man-
agement, which require professional expertise and mastery of a large body of
knowledge. This same trend toward more knowledge workers is present in service
industries, not-for-profits, and government. Drucker estimates that one-third of
all jobs are already filled by the highly paid and productive group he calls knowl-
edge workers.’
The very nature of knowledge work, which involves information gathering,
imagination, experiment, discovery, and integration of new knowledge with larger

Unskilled work > Knowledge work

Meaningless repetitive tasks > Innovation and caring

lndividual work Teamwork

Functional-based work > Project-based work

Single-skilled > Multiskilled

Power of bosses > Power of customers

Coordination from above > Coordination among peers

FIGURE 4-3 The changing nature of work.

The Rise and Fall of Bureaucracy 47

systems, means that bosses cannot order about knowledge workers like the ditch
diggers or assembly-line bolt turners of yore. If knowledge workers are any good
a t all, they soon learn more about what they are doing on a specific project than
their boss. Knowledge work inherently has a large component of self-direction
and teamwork and is hampered by remote control from distant bosses. As we
move beyond bureaucracy we will find ways to organize so that all work is
knowledge work, bringing everyone’s native intelligence and collaborative abili-
ties to bear on constantly changing ways of achieving shared goals.

From Repetitive Tasks to Innovation and Caring

Since the passing of craft production, management has been responsible for
organizing people to work efficiently at narrow, boring jobs. This has meant that
the managerial role was as much to limit the intelligence and potential of employ-
ees as it was to elicit talent. Now the mindless repetitive jobs that bureaucracies
were designed to manage are rapidly disappearing. Machines d o more of the rou-
tine work, and the work that is left requires initiative and flexibility. As a result,
the job of leaders is more nearly to bring out people’s talents around a common
vision.
What sort of work will be left as machines get smarter? What d o people d o
so much better than machines that it will provide human work for the foreseeable
future?
People are much better than machines at innovating, a t seeing new possibili-
ties within fluid and imperfectly defined systems and knowing what to do. Inno-
vation in this sense includes the creative salesperson who sees what the customer
really wants and bends the system to get it. It includes the member of a quality ac-
tion team who makes an intuitive leap that exposes the real root cause of a prob-
lem to measurement and analysis. It also includes the intrapreneur who sees how
to use company assets to generate more revenues and thus create more jobs.
Another apparently irreplaceable human talent is caring. As more work be-
comes service, caring about and for others becomes increasingly important. Peo-
ple do not generally sue doctors just because they make a mistake. They sue them
because they make a mistake and relate to patients in a way that says they d o not
care. Good salespeople keep customers because the customers can sense that they
genuinely care. Good intrapreneurs are able to break through barriers within the
organization when others sense that they care more about the result than about
personal success. Good leaders spread intrapreneurial zeal when it comes from in-
ner values that all can get behind. Leaders elicit commitment when their people
sense that they care about them, the group’s success, and their mutual contribu-
tions.
The rules of bureaucracy forbid caring and, in particular, acting on the basis
of the inner values one holds dear rather than out of strict obedience and loyalty
to the boss. We find no examples of innovation where the intrapreneur did not
break some bureaucratic rules. Most often the intrapreneurs and team members
48 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

were carried away by a passion for an idea that aligned with deeper values-that
promised at least in some small way a better world. We know few con artists who
can long fool the alert into thinking they care when they d o not.
Caring, like innovation, must come from the inside: We cannot order people
to innovate or to care. We also cannot order people to use their intelligence; peo-
ple engage their intelligence when they have reason to care, when they are part of
something bigger than themselves and see that their wider interests are served by
the work at hand. Bureaucracy is too autocratic and rule-driven to motivate and
manage the intelligence that is brought to innovation and caring. Creativity and
connecting with others require engaged relationships, personal responsibility, and
flexible thinking and acting. Thus, as the rules of bureaucracy block both innova-
tion and caring, they block the essence of modern work.

Education, Innovation, and Caring

The Tofflers pointed out in The Third Wave that universal public education
had the purpose of teaching obedience, punctuality, and the ability to sit still a
long time and d o mindless, repetitive work.8 In the early industrial era the ability
to endure boredom was a key survival skill. Although education has improved a
bit, bureaucracies have done little to prepare the average worker for the innova-
tion, teamwork, and caring that constitute much of modern work.
For years corporations have used effective training in creativity and innova-
tion (for a chosen few), but these lessons are generally remedial. They seek to re-
store what was destroyed by education. We need educational systems today that
preserve “childlike” curiosity and give practice in teamwork, initiative, and col-
laborative big-picture responsibility.
Many of our current practices in education not only block innovation, they
also blunt one’s ability to care, to engage heart and mind in one’s work. People
who act on what they care about jump out of their seats. They fail to follow the
lesson plan and ask too many questions. They help their fellow students rather
than maximizing their own grades. Many schools are getting better at teaching
children to care about one another and to treat one another with respect, but still
follow the bureaucratic model in the way both teachers and students are treated-
forced to measure up to defined procedures rather than pursue goals with creative
innovations, evaluated on individual performance instead of teamwork and col-
laboration, taught compliance rather than participative self-management and
democratic processes.

From Individual Work to Teamwork

Bureaucracy replaces the natural ability of humans to find ways to work to-
gether with the more sterile discipline of the chain of command. It is not rich and
lively enough for today’s fast-paced changes and challenges. Virtually every recent
management innovation that works relies in part on the power of teams. A “Total
The Rise and Fall of Btrreaucracy 49

Quality” program gives power to teams to examine processes and make them
work better, a task that until recently belonged exclusively to managers. Because
knowledge workers cannot produce much of value alone, their work takes them
across organizational boundaries to search for integrated information. In reengi-
neering, case teams replace isolated functions. In lean manufacturing, ordinary
workers take responsibility for the whole and run for help whenever trouble
shows up.
When three members of a thirty-two-person work team a t Hoechst
Celanese’s Salisbury plant left the team, the remaining team members had authori-
zation to replace them but decided not to do SO.^ As is so often the case, the people
doing the work knew more than the bosses about where work and expense could
be saved. Organizations become more intelligent when they find ways to bring
the intelligence of every member into supporting the purpose and goals of the
organization.

From Functional Work to Project Work

As knowledge workers shift from static jobs to solving a series of problems

or seizing opportunities, they d o so in work organized as projects. Each project in
this complex world generally requires a cross-disciplinary team. These teams then
learn together as the project evolves. Soon, their bosses in the functions they “re-
port” to become too distant from the work to manage the decisions for the teams.
As a consequence, control shifts from the functional organization of bureaucracy
to project teams.
Specialization will continue to be a critical part of every complex organiza-
tion. But because of the interconnection of issues in a complex world, more and
more work will involve integrating the viewpoints and activities of specialists, and
less and less will be performing tasks completely within those specialties. As a re-
sult, each employee will have to be both a specialist and a generalist.
The 1956 Pontiac was designed with a concave sculpted panel with many
vertical ribs indented between the two taillights. The door to the gas cap was
neatly hidden between two of the vertical ribs. Because the cracks a t the edge of
the gas door fell where the eye already expected a vertical line, the door did not
break up the uniform sweep of the design.
The design worked aesthetically, but a senior manager reviewing it was
afraid car owners would be unable to figure out where to find the gas cap. Rather
than raising the concern with the designers and asking them to deal with it in a
thoughtful way, he ordered the gas cap door chromed, ruining the whole sweep of
the design with an anomalous chrome square.
Managers cannot bring out the intelligence of everyone in the organization
if they pretend they can d o better thinking in a few hours than a project team that
has wrestled with the problem for months. Instead of issuing arbitrary orders,
they need to raise concerns and trust the project team to find a way of handling
them that integrates with all the other issues guiding the design. Paradoxically, as
50 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

issues become more complex and specialties more differentiated, it becomes in-
creasingly necessary for teams of diverse specialists to themselves integrate their
work with the work of other teams. Management can never understand all the
trade-offs and creative solutions that get the team where it is. Heavy-handed inter-
vention leads to inconsistencies-or worse. In an intelligent organization, partici-
pation is widespread to help expose all the issues as early as possible. Individuals
with multiple skills are brought together to cover more viewpoints in a team of
manageable size, and the team does its work guided by feedback, not commands.

From Single-Skilled to Multiskilled

As Fred Emery has pointed out, no system can exist without redundancy to
provide reserve capacity when something does not exactly follow the plan.I0 Bu-
reaucracy gets its margin of safety from extra bodies. If extra work of one kind
appears because customers ordered a different mix of products than expected, a
bureaucracy has extra workers of that exact type waiting in the wings, or it falls
short of meeting the orders. The same situation arises if someone is sick: Another
“identical” worker needs to be waiting to d o the job. This system of narrowly de-
fined skills and extra bodies is expensive and inflexible.
In a typical multiskilling program, responsibility shifts to teams, and em-
ployees get raises for each new skill they acquire. At Lechmere, Inc., a twenty-
seven-store retailer, cashiers at the Sarasota, Florida, outlet get pay raises by
learning to sell products, and sporting goods staff get raises by learning to operate
the forklift. With a multiskilled workforce, when bottlenecks appear, whether
through absenteeism or a sudden rush of one kind of work, someone can step in
and get things moving.”
Bureaucratic relationships between organized labor and management pre-
vent multiskilling by adherence to numerous contractually defined job classifica-
tions. Unions today d o well to negotiate for more training and education to make
members more widely employable. Unionized companies as entrenched as Na-
tional Steel and General Motors “have improved morale, speed, and efficiency by
loosening job classifications and developing a broader more flexible workforce
through cross-training employees.”12

From the Power of Bosses to the Power of Customers

For an organization to be responsive, customers’ wishes have to have a

strong influence on the people doing the work. Relaying this sort of information
through bosses is too slow-and besides, they may not be there to hear what cus-
tomers want.
This sort of thinking applies to internal customers or “users” of a unit’s out-
put as much as to external customers. In a rapidly changing world, if internal cus-
tomers cannot get what they need promptly and flexibly, the system will not be
able to serve external clients promptly and flexibly. Freedom of choice between al-
The Rise and Fall of Bzueaiwacy 51

ternative suppliers gives users of internal services the power enjoyed by real cus-
tomers-the power to say no to one and yes to another. Once internal customers
have this power, the attention of those internal suppliers shifts from pleasing their
bosses to winning customers. If they have customers, the boss can be pleased;
without customers, they had better find new work.

From Coordination from Above to Coordination among Peers

Clearly, new systems of coordination and control are needed. In a bureau-

cratic system, employees are not responsible for coordinating their work with oth-
ers at their level; that is their boss’s job. They need not think about the big picture
beyond doing their specialty well-to d o so would be presumptuous. It is the job
of senior management to figure out how it all fits together, so cross-functional
concerns are referred up to a level of management that can resolve them. When
coordination is the boss’s job, cross-functional, or horizontal, communication
with one’s peers is frowned upon as either a waste of time or a usurpation of the
boss’s authority.
In postbureaucratic organizations, most of the coordination between func-
tions and even businesses is done by teams. In 1988, John Hanley, vice president
for product development at AT&T, needed to cut in half the product development
time for cordless phones. The old product development system was a series of
handoffs from R&D to Manufacturing to Marketing to Sales. Hanley formed
teams that included people from each of these functions and gave the teams
authority to make decisions about almost everything except their deadline: They
would be finished in one year. Rather than wrestle with the bureaucracy, the teams
worked together as intrapreneurial generalists. They did market research, decided
how much each product should cost, what its features would be, what it should
look like, and how it should work. The result: half the development time, better
quality, lower cost.
Reality has become so complex and multidimensional that there is no way
of dividing the organization into chains of command that will work for all aspects
of the challenges faced. As a result, integration is achieved through peer-level
cross-organizational communication rather than through the hierarchy. Huge vol-
umes of cross-functional communication are needed because every important
process crosses the boundaries of the organization. The general manager does not
have time enough in the day just to relay communications; the process is not fast
enough. Besides, as you may remember from the childhood game of “telephone,”
in which a verbal message is whispered from person to person down a long line of
kids, communications relayed through too many humans get garbled. In the intel-
ligent organization, communications whenever possible are direct, without inter-
mediaries.
In the industrial era, the large-scale but stable means of production pushed
us toward distant, formal, and unequal relationships at work. Today, our complex
and intelligence-intensive tasks push us toward relationships that are close, open,
honest, and more nearly equal. Because “organization” is about how we structure
52 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

TABLE 4-1 Revolutionary Change in the Structure of Our Relationships

What
Bureaucracy Is Why It Once Triumphed Why It Fails NOW What Replaces It
Hierarchical Brought simple large- Cannot handle Visions and values
chain of scale order complexity Teams (self-managing)
command Bosses brought order Domination not best Lateral coordination
by dominating way to get orga- Informal networks
subordinates nization intelligence Choice
Free intraprise
Specialization Produced efficiency Does not provide Multiskilling
Organization through division of intensive cross- specialists and
by function labor functional intrapreneuring
Focused intelligence communication and Organization in
continual peer-level market-mediated
coordination networks
Uniform rules Created a sense of Still need rules, but Guaranteed rights
fairness need different rules Institutions of free-
Clearly established dom and community
power of bosses
Standard Provided crude Responds slowly to Self-direction and
procedures organizational change self-management
memory Does not deal well Force of the market
Able to use unskilled with complexity and ethical
workers Does not foster community
Overcame old ways interconnection
A career of Bought loyalty Fewer managers A career of growing
advancing up Furnished continuity needed and more competence
the ladder of elite class of educated workforce A growing network
managers and expects promotions; to get more done
professionals therefore, not More pay for more
enough room for capabilities
advancement
Impersonal Reduced force of Informa tion-intensive Strong whole-person
relations nepotism jobs require in-depth relationships
Helped leaders relationships Options and
enforce tough alternatives
discipline and make Strong drive for
tough decisions results
Coordination Provided direction for Educated employees Self-managing teams
from above unskilled workers are ready for self- Lateral commun-
Furnished strong management ications and
supervision required collaboration
by rapid turnover in
boring jobs
The Rise and Fall of Bureaucracy 53

our relationships, these new realities will completely change our ideas about
methods and patterns of organization.
The nature of work in modern high-tech workplaces calls on people in many
positions in the organization to take responsibility for processes and services that
intimately affect the customer and the wider community. Even in small service
businesses and government agencies, the goods and services produced are knowl-
edge- and information-intensive by virtue of the skills and intelligence of the peo-
ple with their hands on the work processes. When a medical unit delivers
life-saving help to patients, its members must intelligently apply hundreds of tech-
nical instruments, drugs, and procedures to a variety of unique customers-and
learn anew as the knowledge and technology are continually updated. This is as
true of the technicians as the physicians. What works in a society of knowledge
workers will be completely different from what worked before.

REFERENCES
1. Wolfgang J. Mommsen, Political and Social Theory of Max Weber, p. 112.
2. R. H. Hall, “The Concept of Bureaucracy: An Empirical Assessment,” Americanlour-
nu1 of Sociology 69 (1963): 33. See also Warren Bennis, Beyond Bureaucracy (San
Francisco: Jossey-Bass, 1993), p. 5; and Gerald Zaltman, Robert Duncan, and Jonny
Holbeck, Innovations and Organizations (Malabar, Fla.: Robert E. Krieger Publishing,
1984), pp. 122-123.
3. Fred Emery, “The Management of Self-Managing Groups,” unpublished paper, No-
vember 1989.
4. Weisbord, Productive Workplaces, p. 24.
5. William Blake (1757-1827).
6. John Case, “A Company of Business People,” Inc., April 1993, p. 81.
7. Peter Drucker, Post-Capitalist Society (New York: Harper Business, 1993), p. 64.
8. Alvin Toffler, The Third Wave (New York: William Morrow, 1980), p. 45.
9. D. Keith Denton, “Multiskilled Teams Replace Old Work Systems,” H R Magazine,
September 1992, p. 49.
10. As Fred Emery has pointed out, no system can exist without redundancy to provide re-
serve capacity when something does not exactly follow the plan.
11. Denton, “Multiskilled Teams Replace Old Work Systems,” p. 49.
12. Ibid., p. 48.
This page intentionally left blank
 Organization: Perspectives
from Silicon Valley
Homa Bahrami

Many enterprises are in the midst of fundamental changes in organizational de-

signs and management practices. Pioneering and traditional companies alike are
experimenting with novel organizational structures and management processes in
order to accommodate the fast pace of technological change, global competition,
and the emergence of a knowledge-based economy. These developments are col-
lectively precipitating a move away from monolithic and rigid organizational de-
signs which were geared for repetitive transactions and routine activities. The
resulting impetus is toward flexible and agile organizational forms which can ac-
commodate novelty, innovation, and change.'
This article describes some of the organizational features of the emerging
flexible enterprise and is based on field studies of 37 high-technology firms in
California's Silicon Valley.2 These firms are experimenting with new organiza-
tional arrangements and are a t the forefront of experiencing the challenges of the
information era. Their business foundations are anchored in knowledge-based in-
dustries. Many compete in global markets and face global competition. They em-
ploy educated, young, and mobile professionals with high expectations. Some
enter, or even, create pioneering markets and develop as yet untested products
without the benefit of existing role models and blueprints for success. Moreover,
they must manage novelty and continuous changes in products designs, competi-
tive positions, and market dynamics. As Bill Joy, a co-founder of Sun Micro-
systems, observes:

"High-technology obeys the iron law of revolution. . . the more you

change, the more you have to change. . . you have to be willing to accept
the fact that in this game the rules keep changing. "3

Copyright 1992 by The Regents of the University of California. Reprinted from the California Man-
agement Review, Vol. 34, No. 4. By permission of The Regents.

55
56 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

THE CHANGING ORGANIZATIONAL LANDSCAPE

An extensive array of organizational experiments have been under way in
many firms during the past decade. Some of these developments have turned out
to be transient fads, whereas others point to fundamental shifts in organizational
design and management practice. Some of the more prevalent developments in-
clude delayering, team-based networks, alliances and partnerships, and a new em-
ployer-employee covenant.
The delayering and down-sizing trend was initially triggered by the need to
reduce costs. However, it also reflects the administrative impact of information
and communication technologies. Increased use of technologies, such as electronic
mail, voice mail, and shared databases, has, over time, reduced the need for tradi-
tional middle management, whose role was to supervise others and to collect, ana-
lyze, evaluate, and transmit information up, down, and across the organizational
hierarchy. The potential consequences of delayering are intended to be, in part,
faster response to competitive and market changes, larger spans of control, in-
creased workloads, and a broader range of assignments and roles for individuals
and groups. One of the expected benefits of flatter hierarchies is the organization’s
ability to become flexible and responsive by reducing the time lag between deci-
sion and action-enabling faster response to market and competitive dynamic^.^
In an attempt to manage cross-unit projects and to reduce time-to-market,
many firms are increasingly relying on multi-functional, multi-unit teams. Indeed,
during the last decade “teams” and “groups” have become part of our managerial
vocabulary and are now viewed as a central organizational building b1ock.I A key
advantage of teams is their intrinsic flexibility. They can be formed, re-formed,
and disbanded with relative ease; they can by-pass the traditional hierarchy; and
their composition can evolve over time in order to blend different skills and ad-
dress changing priorities.
Reliance on sub-contracting has been prevalent in a number of industries for
some time.6 Recently, however, there has been a substantial increase in alliances
which affect core business activities-such as product development, distribution,
and financing. This trend is giving rise to complex organizational forms and busi-
ness relationships, A number of reasons have been put forward to explain the
rapid diffusion of such “hybrid” organizational forms. These include “changing
environmental conditions, the limits of large-scale organization, and the impor-
tance of speed and inf~rmation.’’~ As Evans suggests, collaborative partnerships
are a flexible mode of blending capabilities, sharing risks, and generating op-
tioms
Recently, we have also witnessed a major re-assessment of the implicit life-
time contract between employers and employees. Many firms have reexamined
their employment policies-initiating early retirement programs and other incen-
tives to reduce the size of their workforce. As pointed out in other studies, the
critical tradeoff in this context is between “corporate flexibility and individual se-
~urity.’’~ Many corporations rely on temporary workers, specialized vendors, and
consultants in order to flexibly deal with unique contingencies. Additionally, this
The Emerging Flexible Organization: Perspectives from Silicofi Valley 57

trend points to a fundamental shift in the foundation of employer-employee rela-

tionship, away from the traditional patriarchal orientation toward what may be
characterized as a peer-to-peer relationship. This sentiment is echoed in the fol-
lowing comment which encapsulates the implicit relationship between Apple
Computer and its employees: “You own your own careers; we provide you with
the opportunities.”’”
Collectively, these and other changes point to a somewhat radical reshaping
of the traditional organizational landscape.” As current trends indicate, contem-
porary firms need flexible and agile organizations that can effectively function in
environments of continuous and kaleidoscopic, rather than periodic and paradig-
matic, change.

FLEXIBILITY: THE EMERGING IMPERATIVE

Historically, the term “flexibility” has been used rather loosely-referring to
a blend of capabilities and attributes that facilitate adjustments to change. How-
ever, as suggested in previous studies, flexibility is a polymorphous concept whose
meaning varies according to the situational context.’2 For example, flexibility
means “being agile”-fast on one’s feet, able to move rapidly, change course to
take advantage of an opportunity or to side-step a threat. This capability is critical
for enabling “time-based” competition, facilitating rapid response, and reducing
product development cycles. It also refers to the ability to quickly redefine a posi-
tion and re-focus in the midst of a dynamic engagement-such as an acquisition,
new product introduction, or legal proceedings.
Flexibility, however, is not just synonymous with agility. It also implies the
ability to be “versatile”-able to d o different things and apply different capabili-
ties depending on the needs of a particular situation. For example, employees with
diverse capabilities are versatile in that they can readily switch between different
assignments.
On the “defensive” side of the spectrum, flexibility also refers to qualities
which enable an enterprise to endure when negatively affected by change. This
attribute is reflected in concepts such as “robustness” or “resilience.” The former
characterizes the capability to absorb shocks and withstand perturbations-for
example, by having excess slack or liquid assets. The latter refers to the ability to
come back from the brink of disaster without bearing permanent scars or disabili-
ties.13 Sometimes the events which trigger the need to change can be anticipated
ahead of time. More often than not, however, firms need to respond to changes
which are typically unexpected.
The point is that all these different attributes-spanning both offensive and
defensive qualities-are needed in a truly flexible enterprise. The concept of flexi-
bility, in an organizational context, refers to the ability to precipitate intentional
changes, to continuously respond to unanticipated changes, and to adjust to the
unexpected consequences of predictable changes. Put simply, strategic flexibility
58 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

“is the ability to d o things differently or d o something else should the need
arise.”l4
All the different senses of flexibility are critical for the survival and success
of high-technology companies. Indeed many of these firms are at the forefront of
both, inflicting and responding to continuous change. Such environments exhibit
a high propensity for what the economist Shackle termed “kaleidoscopic” change,
where a small, apparently insignificant, change can dramatically alter the entire
c~ntext.’~
Due to short product life-cycles, technology firms have to quickly capitalize
on narrow windows of market opportunity, introduce new products in rapid suc-
cession, and respond, in real time, to competitive and market dynamics. Organ-
izational problems are further exacerbated by rapid and volatile growth patterns.
Early success is no guarantee of long-term survival. An incumbent pioneer can be
quickly eclipsed by a technological breakthrough, an unexpected spin-off, or a
sudden shift in market conditions. Managing kaleidoscopic change is an everyday
fact of life and a criteria for survival; it is not a one-time, periodic adjustment, or
simply a corrective move following a crisis.
In view of these challenges, a number of innovative organizational experi-
ments have been under way in many high-technology firms in Silicon Valley and
elsewhere. Some pioneering moves have also been initiated by established corpo-
rations in the process of metamorphosis and transformation.’6

BUILDING BLOCKS OF FLEXIBLE ORGANIZATIONAL

DESIGNS
High-technology firms face significant organizational tensions in spite of
their relative youth. Irrespective of their size or stage of development, they need to
remain disciplined, lean, and focused, requiring minimal duplication of effort,
stringent accountability, and effective control and coordination. However, a loose,
hands-off management style is needed to manage expectant professionals, main-
tain a conducive environment for creative thinking, and provide the capability for
rapid response to competitive and market developments. As depicted in Table 5-1,
they need flexible organizational systems which can balance dialectical forces-
facilitating creativity, innovation, and speed, while instilling co-ordination, focus
and control, and the staying power to withstand periods of adversity.
The following comments capture the essence of some of the dialectical ten-
sions facing these firms: “We want an environment that enhances individual crea-
tivity, but we d o not want chaos. . . we want people involved in decisions that
affect their work and we want teamwork, yet we want our employees to have a
bias toward action. . . we want small groups of dedicated workers (decentraliza-
tion) but such groups may feel aimless or may be charging in the wrong direction
with hidden agendas. . . we want people to stretch to reach tough goals, so our
real emphasis is on easily-measured short-term growth and profits-but we
The Emerging Flexible Organization: Perspectives from Silicon Valley 59

TABLE 5-1 Organizational Dilemmas

Managing Opposing Tensions
Control Autonomy
Focus Innovation
Global Products Local Recipes
Less Duplication Rapid Response
Time-to-Market Future Products
Today’s Performance Long Term Vision

should also have time to develop our employees for the longer haul, to promote
from within, to monitor the atmosphere for creativity.””
Similar tensions also seem to confront many established entities. Percy
Barnevik, the CEO of ABB, described his firm’s critical organizational challenge as
dealing with three internal contradictions: “We want to be global and local, big
and small, decentralized with centralized reporting.”18 The challenge facing Brit-
ish Petroleum is depicted in terms of a critical paradox: “How to reinforce its
strengths as a corporation while allowing its constituent businesses much greater
flexibility and speed of r e s p ~ n s e . ” ’Similarly,
~ the modern “transnational” must
simultaneously address the need for scale efficiency, local responsiveness, and con-
tinuous learning.20

A Multi-Polar Organization
The traditional model of the industrial enterprise has been one of an all-
powerful center with various subsidiaries. The center has historically formulated
the strategic direction, consolidated and integrated divisional plans, allocated re-
sources, and monitored performance. For example, in the classic multi-divisional
structure, senior corporate management-assisted by their staff-have set the
long term direction while the divisions have implemented the plans.21
This model of the omnipotent center which functions as the enterprise’s
brain has been subjected to much pressure as business enterprises have had to
think and act quickly, re-calibrating their strategies continuously in fast-moving
conditions. Under these circumstances, the traditional approach has several draw-
backs:

Rapid change demands quick reactions and continuous re-calibration.

Separating the brain (the center-which plans a response) from the
muscles (the line units-which enact the response) can lead to slow re-
sponse and result in information distortion through hierarchical filtering
processes.
60 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

The executives with the most up-to-date understanding of evolving mar-

ket realities are typically in the trenches. They are thus best-positioned to
strategize and execute the necessary actions in real time as new impera-
tives unfold.
Line managers in knowledge-based companies have the professional ex-
pertise and the educational background to undertake much of the
strategizing and analytical work; assisted by new technologies, they can
minimize their reliance on corporate support groups.**

The emerging organizational system of high-technology firms is more akin

to a “federation” or a “constellation” of business units that are typically interde-
pendent, relying on one another for critical expertise and know-how. Moreover,
they have a peer-to-peer relationship with the center. The center’s role is to orches-
trate the broad strategic vision, develop the shared organizational and administra-
tive infrastructure, and create the cultural glue which can create synergies, and
ensure unity of mission and purpose. However, these tasks are undertaken to-
gether with the line units, rather than for them. This sentiment is reflected in the
following: “[The center’s] mission is to support our business units in fulfilling
their business goals, and perform the truly corporate services in an effective and
cost efficient manner.”23
Apple Computer is a case in point. Its main line units-although varying in
size, scope, and style-have a peer-to-peer relationship with one another and with
the center. The heads of the line units-Apple Products, Apple USA, Apple
Europe, and Apple Pacific-are represented on its top management team together
with the leaders of the corporate functions-finance, human resources, and legal
and administrative services. Members of the different units collectively participate
in setting and implementing the corporate direction; worldwide meetings (held
twice a year) of the top 400 or so executives provide focused opportunities for dis-
cussing critical challenges; and the extensive movement of people between the
units ensures that personal relationships are forged to enhance inter-unit co-
operation.

Dualistic Systems

Many observers may have the impression that the organizational systems of
high-technology companies are in a continuous state of flux; that formal struc-
tures-in the sense of clear reporting relationships, grouping of skills, and concise
assignment of responsibility, authority, and accountability-do not exist in their
organic setting. Such an impression, however, only reflects one dimension of the
organizational reality. Many firms we observed were both structured and yet cha-
otic; they had evolved dualistic organizational systems, designed to strike a dy-
namic balance between stability on the one hand, and flexibility on the other.
The first component is a substrate of the formal structure which only peri-
odically undergoes major transformation. This provides a formal mechanism for
T h e Emerging Flexible Organization: Perspectives f rom Silicon Valley 61

grouping skills, clustering activities, and assigning reporting relationships, as well

as a base unit which gives many employees an anchor of stability.24However, due
to inertial forces, these bedrock structures can not be changed as frequently as
may be warranted by internal and external changes. Many firms compensate for
the relative inflexibility of the bedrock structure by using overlays of temporary
project teams and multi-functional groups whose members are drawn from vari-
ous operating units. These enable a firm to focus on critical assignments without
causing major disruptions.
A good case in point is the structural evolution of ROLM Corporation, a
pioneering telecommunications company which was acquired by IBM in 1984.
During its 15 years as an independent company, ROLM went through 4 major re-
organizations of its bedrock structure, although it formed and disbanded many
temporary groups and project teams. As depicted in Table 5 - 2 , the first major
structural change was initiated in 1973 (four years after its founding) when it en-
tered the telecommunications business. This involved a fine-tuning of its func-
tional structure to embrace the new venture. The second re-organization occurred
in 1977 when 3 autonomous divisions were set up to focus on different busi-
nesses: mil-spec computers (its original business), telecommunications products,
and a new venture (later discontinued) in the energy management field. The third
re-organization (which was largely confined to the telecommunications business)
was initiated in 1981 when a hybrid structure was created to consolidate its end-
user sales and service organization and to focus on the new initiative in office sys-
tems. A further re-organization was completed in February 1984, prior to the
IBM acquisition. It resulted in a partly functional superstructure and divisional
substructures devised to ensure effective co-ordination of its telecommunications
and office products.

TABLE 5-2 Organizational Evolution of ROLM Corporation: 1973-83

Year Revenue (m$) Business Organization
1973 3.6 Million Mil-Spec Computers Functional
New Venture: PBX

1977 30.0 Million Mil-Spec Computers 3 Stand-Alone Divisions

Digital PBX
New Venture: Energy
Management

1981 294.5 Million Mil-Spec Computers Hybrid: Partly Functional/

PBX Systems Partly Divisional
New Venture: Office Systems

1983 502.6 Million Mil-Spec Computers Functional Superstructure;

PBX Office System Divisional Sub-structure
62 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

In many of the observed firms, such fundamental re-organizations of the

bedrock structure were typically undertaken in response to, or in anticipation of,
metamorphic events-such as changes in the composition of top management
teams, strategic re-orientations, shifting priorities, performance setbacks, and re-
source constraints. Temporary teams, on the other hand, were used for a wide
range of activities-including new product development, strategic assessments,
and the formation of management processes. For example, in early 1984 ROLM
formed a five-person team in the System Development Group (the product devel-
opment arm of its telecommunications business) to set up a company-wide busi-
ness planning process. Team members were drawn from various product
divisions, they made their recommendations within 6 months, and the teams were
subsequently disbanded.
Such dualistic systems enable high-technology firms to deal with a widely
felt tension: how to create a relatively stable organizational setting within whose
boundaries people and resources can be flexibly deployed. Bedrock structures are
the relatively stable base units. Temporary teams are the flexible, rapid deploy-
ment overlay. They enable the organization to pool together different individuals
at short notice, put them to work on diverse projects, and disband them once their
task has been accomplished.

Front-Line Orientation

Historically, organizational roles and departmental activities have been di-

vided into staff and line positions. The first category comprise functions whose
power and influence are based on advisory or monitoring roles, with “the right to
advise, rather than the power to decide.”25Typically, these groups have limited di-
rect control over line operations, and hence over revenues and profits. Functions
such as personnel, planning, and MIS, among others, have historically belonged
to this category. By contrast, line functions, such as sales, manufacturing, or prod-
uct development, have the “power to decide” with direct control over, and ac-
countability for, revenues and profits. Critics have long argued that as a result,
staff functions have been cushioned from the harsh realities of the “market.”
This instrumental distinction between staff and line functions is becoming
increasingly blurred, not just in high-tech companies, but also in many traditional
organizations.26 The impetus for change has largely come from competitive pres-
sures to reduce costs. Many staff functions are becoming directly exposed to the
“front-line” realities of their internal customers-funding their operations by sell-
ing their services to the line units.27
In many high-technology companies, support groups are also typically re-
sponsible for undertaking what would have traditionally been viewed as advisory
assignments and are held directly accountable for the results. For example, an em-
ployee relations expert may deal with a disgruntled employee, and the training
staff may actually design and deliver many courses. Other staff functions, such as
strategic planning and business development, are more support-oriented, rather
The Emerging Flexible Orgmiizatioti:Perspectives from Silicon Valley 63

than control-oriented. These groups typically view their role as facilitators, con-
sultants, and process managers, rather than as formulators of strategies and over-
seers of line activities.28
This front-line orientation has re-assigned power and influence to those in
direct contact with the market and competitive realities. This trend is further re-
flected in the fact that CEOs of a number of technology firms have dual roles and
are directly accountable for specific line operations. For example, John Sculley,
Apple’s Chairman and CEO, has also been partially responsible for the company’s
product development group. In a recent interview, he made the following observa-
tion:

“As 1 look back over the last eight and a half y e l m and say, what things
would 1 have done differently, the one that really stands out is that 1
should’ve gotten involved in product development a lot sooner than 1 did.
To lead a high-technology company, you really hauc to lead it through the
technology and through the products. ’’29

This orientation fuses the strategic and operational roles of senior execu-
tives-enabling them to re-calibrate strategies based on real-time information and
realistic action plans.30

Cosmopolitan Mindset

Many technology firms become global very early in their development. For
example, it is not unusual to find young companies-less than 10 years old-with
manufacturing, research, and distribution facilities in the U.S., Europe, Japan, and
the Pacific Rim. Moreover, many generate more than half of their sales outside the
U.S., and have a large population of non-American employees.
Such a rapid process of globalization makes it necessary to develop a cosnio-
politan mindset that incorporates different cultural assumptions and premises.
This is a significant challenge since it requires balancing strong corporate values
(which typically reflect the “home” culture) with a broad perspective (which ac-
commodates the diverse viewpoints of global customers, employees, and competi-
tors). Despite the inherent challenges, however, a pluralistic culture can provide
considerable versatility by drawing on diverse perspectives, approaches, and
solutions.
Apple Computer is a good case in point. Its executives have attempted to
manage Apple not as an American entity, but as a global company: “we want to
look and feel like a local company to our customers while successfully competing
with worldwide corporations that rapidly leverage expertise and resources wher-
ever they are l ~ c a t e d . ” ~Apple
’ strives to create a cosmopolitan organization-not
with one heart rooted in U.S. culture, but with “multiple hearts which beat as
one” reflecting the diversity of its markets and employees.32 It has attempted to
create a pluralistic organization and a cosmopolitan culture in a number of ways:
64 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

Its top management team is composed of different nationalities. Until

1990, a French-born executive was in charge of worldwide product de-
velopment, manufacturing and R&D. A German-born national is its cur-
rent President and an Australian has been in charge of the Asia-Pacific
group. The composition of this team sends a strong symbolic message to
its employees, partners and customers, reinforcing the value of cultural
diversity.
Workforce diversity is an important part of Apple’s human resource strat-
egy. It is a key component of its recruiting plans, promotion policies, and
management training and development programs.
Apple is also focusing on other initiatives to further strengthen its global
orientation. These include “dispersed expertise to leverage unique local
talent, global dissemination of knowledge and skills partly through com-
munication forums which bring together groups with similar interests,
consistent treatment of global accounts with local look and feel, global
account management information systems, integrated databases and net-
works, and global telecommunications facilities.””
Simultaneous product launch in key global markets is another goal. For
example, Claris Corporation, Apple’s software subsidiary, has set out to
develop the U.S. and international versions of its products at the same
time, so they can be distributed in its global markets soon after their U.S.
introdu~tion.’~

In summary, Apple’s strong corporate culture provides a few bedrock values,

which provide “sameness” and give cohesion to its global operations. However,
each region can exercise discretion in evolving its structure and style to accommo-
date different market conditions and cultural values. Apple’s genetic code perme-
ates every unit, yet each has its own distinctive identity. In striking an effective
balance, a key challenge is “figuring out what has to be the same so that every-
thing else can be different.”3s

Capability-Based Organizations and Multi-Talented

Employees
Andrew Grove, the President and CEO of Intel Corporation coined the ex-
pression: “Our assets have legs; they walk home every day.”36Indeed, the core ca-
pability of high-technology companies is their know-how, which resides in people.
The organization can thus be characterized as a montage of individual capabilities
and informal networks and relationships, rather than a series of pre-determined
roles and positions and formal hierarchical relationships.
The pivotal importance of informal networks in high-technology companies
is due to the fact that the productivity of knowledge-based entities depend on em-
ployees’ capabilities, commitments, motivations, and relationships. They can not
The Emerging Flexible Organization: Perspectives from Silicon Valley 65

be programmed around pre-determined roles and positions in a machine-like hier-

archy, Moreover, continuous change typically renders institutionalized roles and
positions somewhat obsolete. An individual’s effectiveness in getting things done
is based on results and credibility, perceived reputation, and network of relation-
ships, rather than on formal authority, job descriptions, and position in the hier-
archy, In this context, titles, seniority, spans of control, formal power, and
hierarchical position are not necessarily significant determinants of individual suc-
cess and organizational power.
Moreover, in contrast to the specialized orientation of traditional entities,
many high-technology companies build versatility into their organizations by lev-
eraging their employees in different capacities, depending on their situational
needs, This is reflected in the following comments which were made by the foun-
der of a medical electronics firm: “I want to recruit people who are absolute ex-
perts in a given area but who can also apply their talents to other areas; “A” class
players in their field, but also “B” and “C” class players in other field^."^' Effec-
tive employees have the flexibility and the confidence to leverage their knowledge
and capabilities across different areas as and when conditions change and new
needs arise.
Despite the inherent difficulties, many firms try to make their employees
more versatile by putting them through different experiences and rotating them
through various assignments. For example, the chief financial officer of one com-
pany took over the responsibility for building and managing its direct sales and
service organization, despite the fact that he had no prior sales experience. The as-
signment made sense because he was both negotiating with and acquiring a num-
ber of its existing distributors-requiring an understanding of the company’s
strategy, coupled with financial acumen, and personal trust relationships forged
over a number of years. The chief administrative officer of a network-server com-
pany was given the additional responsibility for co-ordinating its major accounts
programs in Japan, despite the fact that he had no previous sales experience or fa-
miliarity with the Japanese market; in this case the critical requirement was the
coordination of the different functional groups in addressing the client’s needs.
The executive in question was ideally positioned to d o this because he continu-
ously interacted with the various groups as part of his on-going administrative
responsibilities.
Developing versatile employees by exposing them to different experiences is
not new or unique to the high-tech sector. As early as the 1970s, Royal Dutch
Shell used its corporate planning group as a vehicle for broadening its line manag-
ers’ perspective and giving them a bird’s eye or “helicopter” view of Shell’s global
operation^.^^ Similarly, job rotation programs at companies such as IBM, Hewlett
Packard, and many Japanese corporations have been a key component of their ca-
reer planning systems for some time. What is different in the emerging high-tech
sector is that employees need to possess a flexible mindset and the ability to adjust
unexpectedly and quickly to the demands of a new assignment, without going
through extensive training or being assigned the responsibility as part of a system-
atically planned career management program.39
66 KNOWLEDGE iMANAGEMENT AND ORGANIZATIONAL DESIGN

Semi-permeable Boundaries

Much has been written in recent years about the rise of strategic alliances
and collaborative partnerships. The consensus seems to suggest that such alliances
are a novel form of “hybrid” organizational arrangement, provide a mechanism
for pooling complementary capabilities, addressing rapid product development
cycles, reducing risks, and providing strategic f l e ~ i b i l i t y Moreover,
.~~ in recent
years they have proliferated into various forms, and are continuously evolving.
High-technology companies have been at the forefront of initiating and
managing many types of strategic partnerships. These vary in form, scope, and
longevity. Many companies have forged their fundamental business proposition
and organizational infrastructure around partnerships. Apple, for example, col-
laborates with third-party software developers, dealers, distributors and resellers,
and sub-system and component suppliers.
While such ‘‘leverage’’ models of business partnership are at the extreme end
of the alliance spectrum, others may have a more limited objective. They may be
used for financing purposes-as is the case with many Japanese investments in
new start-up^.^' They may give the parties reciprocal access to geographic mar-
kets, or they may provide an effective way of pooling know-how and sharing risks
in developing technologically advanced products. In many instances, they are an
extension of the traditional supplier-customer relationships. Irrespective of their
purpose, scope, or form, their continuous formation has broken down the solid
walls which have historically separated the firm from its external stakeholders.
The emergence of these semi-permeable boundaries in the high-technology
sector is organizationally apparent in a number of ways. Many firms have access
to their partners’ internal information systems through electronic mail networks.
For example, Apple gives its partners-including software developers, consult-
ants, dealers and resellers, and sub-system suppliers-access to its internal elec-
tronic mail system. This facilitates communication between the different groups
and gives them timely information on new product releases, press an-
nouncements, and re-organizations, among other Additionally, it is a
common practice for engineers working on joint development projects to be as-
signed to a strategic partner. The employee in question becomes a temporary em-
ployee of the partner for a limited period of time-forging crucial relationships
and gaining access to vital information about the partner’s culture and modus op-
erar~di.~~
In summary, the key organizational challenge facing many high-technology
firms is balancing several opposing tensions: selling and servicing existing prod-
ucts while developing and bringing new ones on stream; remaining, disciplined,
focused, and frugal, while continuously learning, experimenting, and re-calibrat-
ing; generating consensus, yet ensuring timely decisions; balancing individual con-
tribution and teamwork; ensuring short-term profitability in the context of a
long-term vision. The modern high-technology enterprise needs diverse capabili-
ties and multi-faceted organizational arrangements to flexibly deal with these
complex tensions. As depicted in Table 5 - 3 , their organizational building blocks
The Emerging Flexible Orgmization: Perspectives from Silicoii Valley 67

TABLE 5-3 Organizational Attributes: A Comparison

Traditional Model Emerging Model
-
Single Center Multiple Centers
Self Contained Steeples of Expertise
Independent Activities Interdependent Units
Vertically Integrated Multiple Alliances
Uniform Structure Diverse Structures
Parochial Mindset Cosmopolitan Mindset
Emphasis on Efficiency Emphasis on Flexibility

have evolved in order to address these tensions, and to provide different forms of
flexibility.

THE EMERGENCE OF A BI-MODAL ORGANIZATION

Many firms appear to have walked a tightrope between these tensions with-
out having allowed any one imperative to dominate the strategic and organiza-
tional context. These attempts cannot be described in monolithic, unidimensional
terms, as simple recipes and “eithedor” solutions. Their organizational systems
were by no means chaotic, but neither were they in total control. They were not
frugal although a cost-conscious mentality pervaded their style. The management
teams were not mavericks, yet an entrepreneurial zeal and anti-bureaucratic senti-
ments were frequently observed. They focused on generating short-term results
but did not lose sight of their long-term mission. The resulting organizational sys-
tems can be best depicted as “bi-modal”-in that they could accommodate oppos-
ing tendencies and yet function as coherent and cohesive concerns. Signs of
bi-modality were commonly observed in broaching three types of tension: Cen-
tralization versus decentralization, stability versus change, and uniformity versus
diversity.

Centralization and Decentralization

The organizational system of many high-technology firms clearly transcends

the centralization-decentralization spectrum.44 O n the one hand, it needs to re-
main loose, decentralized, and differentiated in order to provide the capability for
creative initiatives and rapid responsiveness. On the other hand, tight centralized
direction is needed to maintain strategic cohesion, manage interdependencies, and
reduce the time lag between decision and action. This imperative is reflected in the
following: “we like the idea of small, decentralized units. . . with focused ac-
countability. . . but our products have to play together. . . our customers buy an
68 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

integrated system. . . there is a major element of success that depends on co-ordi-

nation and close co-operation between the units.”45
Centralizing tendencies can be observed in visible and involved leaders
whose passion, vision, and charisma are critical in charting the direction, generat-
ing cohesion, defining the boundaries, and motivating the troops.46 Moreover,
top management teams are typically involved in new, risky projects during the for-
mative stages and participate side-by-side with the troops in the development
process. For example, a co-founder of ROLM was directly involved in the devel-
opment of its office systems products during the early 1980s, even though he was
an executive vice president and a member of ROLM’s top management team.47
However, strong leadership and directed moves d o not imply that leaders are
the sole source of the corporate vision, or that strategies and decisions are im-
posed from the top. The scenario portrayed by individual contributors is one of a
“great deal of autonomy,” a “lot of room for initiatives,” and “doing whatever it
takes t o get the job done.” Indeed, those who are promoted and rewarded are
typically champions of major initiatives and doers who have made things happen.
Such levels of autonomy have historically been associated with decentralized
structures.
The resulting organization can be best characterized as both centralized and
decentralized. It is centralized in that top management teams are a critical force
behind charting the strategic direction and defining the boundaries for individual
and team initiatives. It is decentralized in that front-line personnel can exercise
discretion in dealing with new imperatives as they arise-within broad, yet well-
defined, strategic and cultural parameter^.^^ The critical catalyst in creating this
alignment is reliance on formal and informal bridging mechanisms which estab-
lish direct communication channels between the leaders and the doers. These in-
clude electronic-based communication, planning sessions and review meetings,
informal opportunities for interaction, educational forums, and open access pro-
tocolsa49Regular communication ensures that impending changes in market reali-
ties and strategic priorities can be quickly discussed, evaluated, and implemented.

Stability and Dynamism

Bimodality is also manifest in the tradeoffs made between stability and

change as reflected in the following remark: “we want to be flexible and respond
to market changes without creating chaos and confusion amongst our people. ”50
Indeed, the priorities facing many high-technology firms are in a state of flux, re-
sulting in continuous change and frequent re-calibrations. For example, one
month the focus may be on launching a new product; another month it may shift
over to volume manufacturing and procurement; and in the third month, product
re-design, based on lead-users’ feedback, may be on top of the business agenda.
Dynamism and change are accommodated through extensive reliance on
project teams, micro re-organizations, and re-deployment of core employees in
various capacities. Moreover, many high-technology firms seek to improve their
The Emerging Flexible Orgonizatiorr: Perspectives from Silicon Volley 69

flexibility by relying on temporary workers, specialist vendors, and consultants

and contractors. Reliance on such a variable talent pool enables them to under-
take different assignments without incurring the fixed cost and the long-term
commitment expected by core employees.
However, constant change can also be threatening and de-motivating for in-
dividuals, and disruptive and unproductive for the organization. It is not surpris-
ing to find that many firms strive to create anchors of stability around which
everything else can change. Some attempt to clarify and articulate a clear sense of
purpose and a few overarching values which define the broad boundaries within
which changes take shape.” For example, the mission of Conner Peripherals, a
disk drive manufacturer and one of the fastest growing companies in U.S. corpo-
rate history, is described as follows: “Identify customers’ needs sooner and fill
them faster than the competition.”s2 Moreover, their recruiting practices and ori-
entation programs help set the employees’ expectations and thereby ensure an ef-
fective fit between personal and organizational goals.

Uniformity and Diversity

There is a clear sense of corporate purpose and cultural identity associated

with pioneering high-technology companies, yet their style professes to value di-
versity. Inculcating diversity enables these firms to become versatile, pool together
different capabilities, and nurture the ability to address different contingencies.
Many high-technology companies attempt to become “diverse” by blending
various management styles and cultural perspectives. For example, they may re-
cruit inexperienced college graduates as well as experienced professionals with ex-
tensive track record. They also recruit people from different cultures and ethnic
backgrounds to blend together different cognitive orientations. A young company
in the network server business, for example, consciously sought to recruit a
woman chief financial officer from a different cultural background in order to
provide a role model for its women professionals and develop the capability base
to deal with clients and partners from other cultures. In this case, after a period of
extensive search, they recruited an Asian woman as their chief financial officer.
Composition of top management teams can also send an important sym-
bolic message and further reinforce the importance of diversity. A well-known
case is the complementarity between David Packard’s business style and Bill
Hewlett’s technical orientation. Other famous examples include the late Noyce-
Moore-Grove troika at Intel, and Oshman, Maxfield, and Chamberlain at
ROLM. These teams represent unity through their shared values and overarching
sense of purpose. Diversity is promoted in that they have complementary skills
and management styles.
Recently, a number of high-technology firms have also set out to sensitize
their employees to cultural diversity through in-house training and educational
programs. A few companies have made strong commitments to internal training
programs that prepare executives for global assignments and strive to build cul-
70 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

tural awareness in all employees. The crucial value of diversity further highlights
the importance of distinctive corporate values. These spell out a few boundary
conditions within which everything else is free to operate. They define the limits
and set the constraints for individual and team initiatives.

SUMMARY AND CONCLUSION

Developing flexible organizations is critical for business enterprises in the
1990s. Flexibility is a multi-dimensional concept-demanding agility and versatil-
ity; associated with change, innovation, and novelty; coupled with robustness and
resilience, implying stability, sustainable advantage, and capabilities that may
evolve over time.
A critical challenge facing many business entities is how to transform their
traditional organizational systems and management practices in order to become
more flexible. This task requires identifying and implementing those approaches,
processes, and tools that can be used to manage a bimodal-rather than a mono-
lithic-organization. This poses a major challenge because our existing organiza-
tional systems and managerial mindsets have evolved to address uni-dimensional
imperatives, rather than the new, rampant multi-dimensional tensions.
Moreover, our expectations, norms of behavior, vocabularies, and frames of
reference have evolved around the traditional themes of stability rather than
change, uniformity rather than diversity, and optimality rather than flexibility. We
need to forge new attitudes and behavior patterns by deploying educational pro-
grams, incentive systems, and communication protocols, among others, to sup-
port and reinforce the importance of flexibility, diversity, and dynamism. If the
experience of the high-technology sector is indicative of broader trends, the 1990s
is likely to be a decade of organizational experimentation and managerial innova-
tion, and one likely to bring forth novel organizational systems and management
approaches. This challenge requires focused attention, a readiness to experiment,
and the willingness to share ideas and learn from different corporate experiences.

REFERENCES
1. The need for flexibility is discussed in a number of recent books. For example, see C.
Handy, The Age of Uweason (Boston, MA: Harvard Business School Press, 1989);
R.M. Kanter, When Giants Learn to Dance (New York, NY: Simon and Schuster,
1989);T. Peters, Thriving on Chaos (New York, NY Harper and Row, 1987);R. Pas-
cale, Managing on the Edge (New York, NY: Simon and Schuster, 1990).
2. These findings are derived, in part, from a study of 37 firms in the electronics sector,
conducted during 1982-91. The firms varied in size, ranging from start-ups to
multi-billion dollar corporations. Their businesses spanned several high technology in-
dustries-including computer hardware and software, networking and telecommuni-
cations, computer peripherals, semiconductors, and scientific instruments. Sources of
The Emerging Flexible Organization: Perspectives from Silicon Valley 71

data included structured and semi-structured interviews with a cross-section of senior

executives and rank and file employees, analysis of internal documents, attendance at
company meetings, and use of published data.
3. Speech given by Bill Joy, a co-founder of Sun Microsystems, at the Churchill Club,
Palo Alto, California, 1990.
4. In an earlier article, it was argued that alignment between the apex and the core, or the
generals and the troops, is critical in this context. See H. Bahrami and S. Evans “Stra-
tocracy in High Technology Firms,” California Management Review (Fall 1987).
5. Self-managing, multi-functional teams are considered to be the backbone of “fast cy-
cle” companies. See J.L. Bower and T. Hout, “Fast Cycle Capability for Competitive
Power,” Harvard Business Review (NovemberDecember 1988). A “project orienta-
tion” enacted through “co-located, joint function teams” is viewed as the hallmark of
innovative companies. See T. Peters, “Get Innovative or Get Dead,” California Man-
agement Review (Fall 1990), p. 23. A recent British study concludes that in today’s
business environment career success often hinges on project management (and hence
team management) skills. The business section in London’s Sunday Times, March 11,
1990. Additionally, since the Hawthrone studies we have known that groups are a
critical source of individual commitment, loyalty, and identity-essential ingredients
for managing professional and expectant knowledge-workers. For an insightful per-
spective on the Hawthorne Studies, see EJ. Roethlisberger, The Elusive Phenomena
(Boston, MA: Harvard University Press, 1977).
6. For example, general contractors in the construction industry have relied on their
long-standing relationships with specialist sub-contractors. The textile sector has his-
torically evolved around an intricate web of relationships between different specialists;
and the automobile industry has traditionally benefited from the diverse capabilities of
its subcontractors-each possessing a unique skill in a distinctive area. For further dis-
cussion see W. Powell, “Hybrid Organizational Arrangements,” California Manage-
ment Review (Fall 1987).
7. Ibid.
8. J.S. Evans. “Strategic Flexibility in Business,” SRI Business Intelligence Program. Re-
port #678, Menlo Park, California, December 1982.
9. As Kanter points out the broad implication is that “people’s careers are more depend-
ent on their own resources and less dependent on the fate of a particular company . . .
no longer counting on the corporation to provide security and stature requires people
to build those resources in themselves.” See Kanter, op. cit., p. 357.
10. Speech given by K. Sullivan, Senior Vice President, Apple Computer, to the author’s
MBA class on Organizational Behavior a t U.C. Berkeley, April 1991.
11. Miles and Snow characterize the firm of the future as a “network” in which “brokers”
play significant roles as intermediaries. See R. Miles and C. Snow, “Organizations:
New Concepts for New Forms,” Califoriiia Managemerit Review (Spring 1986).
Hedlund describes the emergence of “heterarchy” based on observations of Swedish
multinationals. By way of contrast from their hierarchical predecessors, these firms are
characterized by “diffusion of strategic roles, a wide range of governance modes, and
holographic attributes.” G. Hedlund, “The Hypermodern MNC-A Heterarchy?”
Human Resource Management lournal (Spring 1986). Child discusses how informa-
tion and communication technologies have enabled many firms to “externalize” inter-
nal activities to contractors, and yet retain a certain measure of operational
72 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

control-giving rise to novel organizational arrangements. See J. Child, “Information

Tech nology and Response to Strategic C ha Ilenges, ’‘ California Management Review
(Fall 1987). Handy describes the firm of the future as resembling a “shamrock”-
comprising partners, consultants and part-time workers, as well as full-time employ-
ees. See Handy, op.cit.
12. J.S. Evans, “Strategic Flexibility and High Technology Maneuvers: A Conceptual
Framework,” ]ortmal of Management Strrdies (January 1991).
13. Ibid.
14. A. G. Hart, “Anticipations, Business Planning, and the Cycle,” Quarterly lorrrnal of
Economics (February 1937), p. 272.
15. G.L.S. Shackle, Expectations, Investment, and Income (Oxford University Press,
1938).
16. Bell Atlantic and ABB are cases in point. See R. M. Kanter’s interview with R. Smith of
Bell Atlantic entitled “Championing Change,” Harvard Business Review (Janu-
arylFebruary 1991), p. 26; and W. Taylor interview with P Barnevik of ABB in “The
Logic of a Global Business,” Harvard Birsiriess Review (MarcMApril 1991), p. 95.
17. Internal memorandum from the Executive Vice President of a $700 million high-tech-
nology firm on its business philosophy, May 26, 1981.
18. See Taylor, op. cit., p. 95.
19. O n changes at B.P., see C. Lorenz, “ A drama behind closed doors that paved the way
for a corporate metamorphosis,” Fimncial Times, March 13, 1990, p.13.
20. See C. Bartlett and S. Ghoshal, “Organizing for Worldwide Effectiveness,” California
Management Review (Fall 1988).
21. See A.D. Chandler, Strdegy and Structure (Cambridge, MA: MIT Press, 1962) on the
historical evolution of the multi-divisional structure.
22. J. Child, “New Technology and Developments in Management Organization,” Omega,
12/3 (1984).
23. Source: internal board presentation at a software company. The firm employs about
100 professionals in roles which are classified as “corporate.” These include “finance
and administration” (corporate reporting, taxes, treasury, planning, investor relations,
and internal audit), infrastructure support (facilities and safety, information systems,
corporate purchasing, and risk management), legal services, human resources, and
corporate communications.
24. See H. Bahrami and S. Evans “Emerging Organizational Regimes in High Technology
Firms,” H u m a n Resource Management ]ozrrnal (Spring 1989) for an account of the
evolution of the formal structure in high technology firms.
25. H. Mintzberg, Structure in Fives: Designing Effective Organizations (Englewood
Cliffs, NJ: Prentice Hall, 1983).
26. A good case in point is Bell Atlantic’s “client-service’’ groups. During a recent reorien-
tation process, many of its support functions have had to operate as line groups-en-
listing the support of other units in order to continue their operations. In this case, the
line units can either buy their required services from the support groups or from out-
side vendors; the only constraint is that they can not develop their own support func-
tions. Reciprocally, thc support units can also offer their services to outside clients,
within certain competitive guidelines. In this way, the old staff entities have been trans-
formed into line operations-with their own clients, revenues, profits, and direct ac-
The Ernergirzg Flexible Organization: Perspectives from Silicon Valley 73

countability to the user. If the activity can not be supported, it is disbanded, in direct
response to market feedback. See Kanter, op. cit. The point was also discussed by Ray
Smith, Bell Atlantic’s Chairman and CEO, in a keynote speech given at a U.C. Berkeley
Executive Program, Sturbridge, MA, October 1990.
27. For a detailed study of how the human resource function is being transformed along
some of these dimensions, see The Conference Board, “The Changing Human Re-
source Function,” Report number 950, New York, 1990.
28. This was the finding of a study conducted by the author; H. Bahrami, “Strategic Plan-
ning in Emerging and Established Firms: A Comparison,” paper presented at the fifth
Strategic Management Society Conference, Barcelona, Spain, 1985.
29. See R. Karlgaard, “Sculley Looks Ahead,” Upside (October 1991), p. 101.
30. For more detail on how strategic decisions in high technology companies entail con-
tinuous re-calibration see H. Bahrami and s. Evans “Strategy Making in High-Tech-
nology Firms,” California Management Review (Winter 1989); for a study on how
effective executives in the microcomputer industry relied on real-time information to
make decision, see K. Eisenhardt, “Speed and Strategic Choice,” California Manage-
ment Review (Spring 1990).
31. Internal presentation, Apple Computer, September 23, 1988.
32. Spindler, M . Speech given to Apple’s worldwide sales force, March 1987.
33. Internal document on Apple’s New Enterprise Project, 1988.
34. Sun Jose Business Journal, Special Report (March 1992), p. 14.
35. Personal Communication with executives at Apple Computer, 1990.
36. Indeed the phrase “people are our key asset” is often used not just in high-tech compa-
nies, but increasingly in traditional corporations. This is partly reflected in the recent
focus on “human resources” as a critical function. See London’s Financial Times,
January 28, 1991, p. 14. Extensive reference to the importance of “networking” is an-
other manifestation of this trend. In a recent article, Lorenz discusses the confusion
surrounding the concept of the “networked” organization, especially concerning “of-
ficial” and “unofficial” networking: “Official networks are certainly a praiseworthy
advance over the bureaucracies they replace. But they are often little more than
streamlined, well-run, and physically-dispersed committees . . . the real breakthrough
is official blessing for all sorts of unofficial, informal networks. . . it is they, much
more than the official variety, which will really help create the open and flexible learn-
ing organizations which most forward looking companies aspire to become.” “Net-
work Organizations,” Financial Times, Management Page, April 3, 1991,
37. Personal Communication, W. New, M.D., August 1989.
38. Evans, op. cit.
39. It is commonplace to find employees assigned to new roles or seconded to other
groups on a frequent and, at times, unplanned basis. For example, a key member of
the engineering staff in a surveyed company was unexpectedly seconded for nine
months to the manufacturing group because he was very knowledgeable about an out-
sourcing issue-a critical priority for the manufacturing group at the time.
40. See Powell, op. cit.; Evans, op. cit.; Kanter, op. cit.; B. Borys and D.B. Jemison, “Hy-
brid Organizations as Strategic Alliances,” Working Paper 951r, Graduate School of
Business, Stanford University, 1987; D. Mowery, International Collaborative Ventures
in Manufacturing (Cambridge, MA: Ballinger, 1988); W. Ouchi and M. Bolton, “The
74 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

Logic of Joint Research and Developmerit,” California Management Review (Fall

1988);Y.L. DOZ,“Technology Partnerships between Larger and Smaller Firms,” Inter-
national Studies of Management and Organization, 1714 (1987).
41. See A. Besher, “Asian Investor Feast,” Upside, (November/December 1989).
42. In order to differentiate between the information given to “outsiders” versus “employ-
ees,” Apple has its “Hotlinks” electronic bulletin board which is only accessible by
company employees.
43. See N. Rutter, “Kubota-San,” Upside, (JanuaryIFebruary 1990).
44. See Bahrami and Evans (1989),op. cit.
45. Interview with R. Maxfield, Executive Vice President and co-founder, ROLM Corpo-
ration, April 1985.
46. Well-known examples include ROLM and Ken Oshman, Digital Equipment and Ken
Olson, Intel and Andrew Grove, Sun Microsystem and Scott McNeilly, Cypress Semi-
conductor and T.J. Rogers, Microsoft and Bill Gates, Software Publishing and Fred
Gibbons, NeXT and Steve Jobs, ASK and Sandra Kurtzig, Electronic Arts and Trip
Hawkins, Oracle and Larry Ellison, among others.
47. This contrasts with the practices observed in many traditional firms. As recent studies
have shown, new activities in established entities are, by and large, initiated at the pe-
riphery of the firm [see Beer et al., “Why Change Programs Don’t Produce Change,”
Harvard Business Review (November/December 1990)] or by lower level champions
in an autonomous manner [see R.A. Burgelman and L.R. Sayles, Inside Corporate In-
novation (New York, NY: Free Press, 1986)l. This makes sense since in established
firms those o n the periphery have much to gain and little to lose by embarking on pre-
sumably risky, new initiatives. By contrast, their senior colleagues would have typi-
cally advanced in the organization by playing according to the prevalent rules, without
disturbing the status quo. In the absence of a major crisis, many are thus likely to sup-
port a major change either under crisis conditions, or when it has proved to be feasible
and entails minimal risk to the organization.
48. New technologies have provided the means to transcend the centralizatioddecentrali-
zation spectrum, see J. Child, “New Technology and Developments in Management
Organization,” Omega, 12/3 (1984);and J. Child, “Information Technology and Re-
sponse to Strategic Challenges,” California Management Review (Fall 1987); G.
Huber, “The Nature and Design of Post Industrial Organization,” Management Sci-
ence, 30 (1984);M.L. Markus and D. Robey, “Information Technology and Organiza-
tional Change,” Management Science, 34 (1988); S . Zuboff, In the Age o f t h e Smart
Machine (New York, NY: Basic Books, 1988).
49. For further discussion, see Bahrami and Evans (1989),op.cit.
50. Personal Communication, Executive Vice President of a software company, prior to a
major re-organization, March 1989.
51. For an assessment of the critical role of corporate culture in business organizations,
see C. O’Reilly, “Corporations, Culture, and Commitment,” California Management
Review (Summer 1989).
52. “Managing at Conner,” Internal Employee Orientation Handbook, Conner Peripher-
als, 1992.
The Emerging Flexible Organizatiorz: Perspectives from Silicon Valley 75

The author would like to express sincere thanks to Glenn Carroll of the Haas School of
Business, U.C. Berkeley, John Rollwagen of Cray Research, and Kevin Sullivan of Apple
Computer for their helpful comments on earlier drafts of this article. The field research
was made possible, in part, by the participation of many high-tech companies and their
executives. Special thanks are due to Bob Maxfield, Ken Oshman, and Dennis Paboo-
jian, formerly of ROLM Corporation, for sharing their insights and experiences.
This page intentionally left blank
 6
The Organization ~

of Innovation
Tom Burns & G.M. Stalker

TECHNICAL PROGRESS AND SOCIAL CHANGE

In their most general form, the findings of this research can be put into two
statements: Technical progress and organizational development are aspects of one
and the same trend in human affairs; and the persons who work to make these
processes actual are also their victims.
The connexion between progress in material technology and the emergence
of new forms of social organization is familiar enough. But it has become sub-
merged in the century-old controversy about the correct causal sequence of tech-
nical progress and social evolution. According to one view, widely held by
American social scientists (see e.g4J),but stated in its most uncompromising form
by Marx, technical progress underlies every kind of change in the social order:
“Assume a particular state of development in the productive faculties of man and
you will get a corresponding form of commerce and consumption. Assume par-
ticular degrees of development of production, commerce, and consumption, and
you will have a corresponding form of social constitution, a corresponding or-
ganization of the family, of orders or of classes, in a word, a corresponding civil
society.”6 The argument has an oddly up-to-date ring, not so much in terms of the
interpretation of history as of the actualities of international politics. So con-
vinced have we become of the dependence of the total social, political, and
economic order on technical development that national output of scientific dis-
coveries and rate of technological advance have begun to appear as an ultimate
criterion of culture, and different political and social systems are compared as fa-
cilitators of this kind of achievement.
According to the other view, technical progress is the outcome of changes in
the institutions of society, even simply, as Durkheim’ argued, of population
growth, which produces not only new needs but the improvements and expan-
sions of knowledge and equipment necessary to satisfy them.

Copyright 0 Tom Burns and G.M. Stalker 1961. Reprinted from The Maiiagement of hinovation by
Tom Burns and G.M. Stalker (1961; revised edition 1994) by permission of Oxford University Press.
77
78 KN 0W L ED G E MAN A G EM ENT AN I> 0 K G A N IZ AT10N A L. DESIGN

Like many another instance of the chicken-and-egg conundrum, the ques-

tion which comes first conceals a false antithesis. A social technology, as witnessed
in the growth of modern institutions, has been developed alongside technology in
the material sense.
The clue lies in Tonnies’ perception of the development of modern society as
itself a technological process. The relationships and institutions he regarded as
characteristic of the modern world are those which enable persons to manipulate
others, individually and eu masse, in the pursuit of their own ends. Thus other
people take on more and more of the quality of a natural environment which man
looks on as resources for him to consume and manipulate. In many important re-
spects we treat “human beings like inanimate objects and tools.”x It could almost
be said that Tonnies regarded social change as one aspect of technical develop-
ment; organizational techniques and devices for manipulating others were con-
stantly invading the social order.
Progress in power technology, in agriculture, in engineering, in chemicals,
and the rest have proceeded-quite inevitably and necessarily-alongside devel-
opments in working organizations and in communications, and alongside the
elaboration of social and political controls, financial and other economic mecha-
nisms.:’ Developments on each side have often been by way of adaptation to
changes on the other side. Yet there remains the discrepancy, the obvious and aw-
ful gap between technical achievement and the constraint and fears which bear
upon people in their day-to-day lives. Consciousness of the gap has sometimes, in
recent years, found express on in opaque generalizations about man’s control of
nature out-running his social abilities.
The trouble about such statements is that they mask the realities to which
they refer. What is true is that developments in the one have been forced through
with just as little regard for ultimate consequences to human welfare as in the
other. The advantages looked for, and won, by the progress of technology, both
material and social, have been immediate or short-run. Very often, these rewards
are of the kind which can benefit only individuals or interested minorities. The
cost, in more destructive wars, in dust-bowls, in road accidents, on the one hand,
and in slums and industrial servitude, emotional deprivation, loss of intellectual
and economic independence on the other, only becomes apparent in the long run.
It then manifests itself often enough as a price to be paid by others. Unforeseen
long-term advantages have of course accrued too. The point is that technological
development has typically occurred as a consequence of decisions made in the
light of short-term views of the balance of advantage and cost to people in con-
trolling positions.
All novelty involves some degree of risk. The vast majority of biological mu-
tations are said to be harmful. When, as in human affairs, enormous numbers of

“Professor Jewkes has suggested that “Whilst no one would wish to deny that technology and sci-
ence (in that order) have contributed much to the raising of standards of living in the last two cen-
turies, there is a disposition in these days to exaggerate the contribution they have made and to
underestimate that made by new social organisations and institutions.”’
The Organization of Innovation 79

random possibilities are eliminated by rational choice, the chances of harm rather
than good resulting are reduced, not eliminated. The harm consists in both cases
in making the individual or organization less fit to survive in its environment than
was its predecessor. Very often, the environment of the person or organization is
itself changing, so that even to maintain the same degree of fitness for survival,
people and institutions may have to change their ways. So the risks attendant
upon change may have to be weighed against other risks arising from maintaining
the same state of affairs.
This condition of ordinary human existence is made explicit and articulate
in the institutions and procedures of industry. And in those sectors of industry in
which the creation of innovations is a constant and important part of the total en-
terprise, the processes of change become visible in an obvious and dramatic way.
Here too, the mutual, procreative impact of developments in material technology
and social organization finds its clearest expression.
In one very important sense, the link between the two trends is a necessary
interdependence. Invention, even more than science, is a social phenomenon; in
quite matter-of-fact ways, it is a human activity which can only be fulfilled when
certain social conditions obtain, when the inventor inhabits a milieu which
prompts him to devote himself to a specific line of work with the promise of re-
wards-in money, power, or even a secure livelihood, in fame, or even self-es-
teem-and which will thereafter support him economically and intellectually. The
notion of the hermit genius, spinning inventions out of his intellectual and psychic
innards, is a nineteenth-century myth, useful then, as myths may always be, but
dangerous, as myths always are, once its period of usefulness is past.
If, as Whitehead said, the greatest invention of the nineteenth century was
the invention of the method of invention,'O the task of the succeeding century has
been to organize inventiveness. The difference is not in the nature of invention or
of inventors, but in the manner in which the context of social institutions is organ-
ized for their support.

THE SOCIAL CONTEXT OF INVENTION

The review of the past institutional context of industrial innovation which
follows is designed, therefore, to underline the importance of that context and to
point out the ways in which it has significantly changed. It is not offered as a his-
tory, even in a very abridged form, of the relationship between science and indus-
try during the last two hundred years, but rather as a sketch of the phases of
change in institutions of some importance to society. It forms the background to
the succeeding account of the attempt by industrial concerns to digest the thing
they have swallowed.
During the middle years of the last century the electrical industry was estab-
lished on the basis, largely, of supplying telegraph services. Within a few years the
development of electric motors for tramways and stationary machinery led to very
considerable expansion. As new applications multiplied, the need for heavier and
more efficient generating plant and distribution equipment accelerated the proc-
80 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

ess. By 1880 there was a flourishing, keenly competitive, electrical industry not
only in Britain and in the United States, but also in Germany, France, and other
European countries. It was an industry, moreover, in which the technological base
was very recent-middle-aged men in the industry would be well aware of the
first commercial applications-and in which new applications and design im-
provements followed each other extremely rapidly. Yet the two major innovations
during the last twenty years of the century, incandescent electric lighting” and ra-
dio,12 were the work of newcomers, of inventors and enterprises unconnected
with the existing industry. No spectacular “discovery” lighted upon by an individ-
ual genius was really responsible; electric lamps and wireless transmission were
“in the air” many years before the first commercial companies were floated.
Swan, a chemist, made experimental incandescent lamps in 1860 which em-
ployed the same high-resistance conductor, carbonized paper, as was used in the
first commercial lamps marketed twenty years later. There were, by 1880, large
industrial concerns manufacturing lighting and other electrical equipment; yet in
the event it was Edison who, two years after becoming interested in the possibility,
first developed the lamp and formed an independent concern to manufacture it.
Lodge, following upon Hertz’s earlier experimental work, demonstrated
wireless reception before the British Association in 1894, and two years earlier a
physicist had written in the Fortnightly Review of the “possibility of telegraphy
without wires, posts, cables, or any of our present costly appliances,” adding “this
is no mere dream of a visionary philosopher. All the requirements needed to bring
it to within grasp of daily life are well within the possibilities of discovery, and are
so reasonable and so clearly in the path of researches which are now being ac-
tively prosecuted in every capital of Europe that we may any day expect to hear
that they have emerged from the realms of speculation to those of sober fact”
(quotedt2).Yet the development of this obviously profitable venture interested no
commercial concerns for ten years.*
In the case of radio, it was the twenty-year-old Marconi who, on the basis of
Hertz’s work as described in an Italian journal, constructed home-made equip-
ment which was sufficiently advanced after three years’ work to communicate
messages over eight miles and to bring the Marconi Company into being.
Anyone who has read accounts of technological advances, of inventions,
during the nineteenth century will perceive this pattern of development as in many
ways entirely typical. It is typical not only of the way in which invention then
“happened,” but, even more, of the way people thought of invention as happen-
ing a t any time. Invention was seen as the product of genius, wayward, uncon-
trollable, often amateurish; or if not of genius, then of accident and sudden

*In America the pattern was oddly repetitive, for in the 1870’s Graham Bell had tried to interest
telegraph companies in the new, and rival, method of communication by telephone which he had
invented; after unavailing efforts he founded the American Telephone and Telegraph Company. By
the 1890’s this company was the “most research-minded concern” in the industry. Yet it felt uncon-
cerned about radio (apart from one brief and unsuccessful episode in 1906) until 1911, when the
threat from wireless telegraphy was too strong to be any longer ignored.
The Organization of Innovation 81

inspiration, As such, it could not be planned for, organized as a part of the field of
existing industry, the idea was intrinsically absurd. In nineteenth-century Britain
the archetypal formula for the process of innovation was enshrined in the fantasy
of Watt and the kettle.
The fitting of this latter myth to the key episode of the earlier technical revo-
lution was itself characteristic. Of course, the myth of accident and inspiration did
go some way towards accounting for the nineteenth-centtrry facts. And the out-
standing fact was the random distribution of scientific and technical information
through the new journals, popular lectures, and societies. These diffusers, and the
continued exploitation of major inventions by craftsmen, made it seem possible
for any individual innovation to be produced by almost anybody, almost any-
where. Again, the disciplined attack on one difficulty after another, which is how
the gap between the scientific idea and the ultimate product is bridged, was still
intrinsic to the achievement, but the process was an individual, usually personal,
enterprise. Often, as in the case of the electric lamp and radio, many individuals at
great removes from each other were involved over a period of years in the devel-
opment of a single invention.
Images and myths about the past had to fit these contemporary facts. So the
boy Watt sat dreaming in front of a boiling kettle and later invented the steam en-
gine. The essential condition of membership of a closely linked group of “applied
scientists,” as they would now be called, in the Universities of Glasgow and Edin-
burgh, the especial circumstance of friendship with Joseph Black, whose discovery
of latent heat lay at the bottom of Watt’s improvements to the Newcomen engine,
the inclusion of the industrialist Roebuck in the circle of personal acquain-
tanceship-these, the really significant factors, were simply left out of popular ac-
count. They were social circumstances which were no longer appropriate to the
progress of technology.

Coteries and Clubs

In the latter half of the eighteenth century the Scottish universities were cen-
trally involved not only in the primal discoveries of the industrial revolution in
chemistry and engineering but in the technical applications and commercial ven-
tures which exploited them. The rapidity of technological development in so
many fields which were being explored simultaneously in the laboratories of
Edinburgh and Glasgow was the direct outcome of close personal association be-
tween persons with different expertises and different resources. But the associa-
tion between people like Watt, Black, and Roebuck was founded not so much on
their membership of a common profession or organization as on membership of a
small, closely integrated ~0ciety.l~ In the Scotland of the eighteenth century, for
such men to be acquainted with each other was virtually inevitable.
Such circles of personal acquaintanceship served as a social medium for a
further decade or so. By the beginning of the nineteenth century, fellow-students
and friends sought to institutionalize their informal acquaintanceships. Clubs
82 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

rather than learned societies, as the Lunar Society and the Royal Society of Edin-
burgh were, they and their off-spring and kindred in Manchester and Newcastle,
and the archetype in London, included the persons responsible for scientific ad-
vance, technical invention and, to a large extent, industrial innovation.
“Towards the close of the last century,” says Smiles in his life of Boulton and
Watt,I4 “there were many little clubs or coteries of scientific and literary men es-
tablished in the provinces, the like of which d o not now exist-probably because
the communication with the metropolis is so much easier, and because London
more than ever absorbs the active intelligence of England, especially in the higher
departments of science, art, and literature. The provincial coteries of which we
speak were usually centres of the best and most intelligent society of their neigh-
bourhoods and were for the most part distinguished by an active and liberal spirit
of inquiry. Leading minds attracted others of like tastes and pursuits and social
circles were formed which proved, in many instances, the source of great intellec-
tual activity, as well as enjoyment. At Liverpool, Roscoe and Currie were the cen-
tres of one such group; a t Warrington Aiken, Enfield, and Priestley of another; a t
Bristol Dr Beddoes and Humphrey Davy of a third; and at Norwich the Taylors
and the Martineaus of a fourth. But perhaps the most distinguished of these pro-
vincial societies was that at Birmingham, of which Boulton and Watt were among
the most prominent members.
“ T h e object of the proposed Society was to be at the same time friendly and
scientific. The members were to exchange views with each other on topics relating
to literature, arts, and science; each contributing his quota of entertainment and
instruction. ” (our italics.)
But the rate of expansion of science and technology was too rapid to be ac-
commodated by adapting and multiplying the institutions of sociable intercourse,
vigorous as they were in middle-class society a t that time. The founding, in 1831,
of the British Association, a self-conscious attempt to institute personal links
between all scientists and technologists, may be regarded as marking the end of
the period when a network of personal relationships on the necessary scale was
feasible.

The Diffusion of Technical Information

What took the place of the circle of people who were at once friends, fellow
scientists, and business partners, or the coterie whose common interests were a t
the same time scientific, technological, and financial? Instances of the kind quoted
earlier, and the myth, still surviving today, of the lonely inventive genius, suggest
that in the period roughly from 1825 to 1875-succeeding the great days of the
provincial Societies-information about scientific discoveries became available to
a wide variety of people. Personal communication was replaced by mass commu-
nication.
The change from speech and letters, involving small-scale contact of a pecu-
liarly intimate and undisturbed kind, to print, by which leisurely and undisturbed
The Organization of Innovation 83

communication is effected impersonally, randomly, and with large numbers of in-

dividuals, is an institutional change of a particularly potent kind; we are familiar
with the part played by printing in accelerating the diffusion of information and
ideas in the Renaissance and Reformation, and thus acting as a multiplier of their
social impact. A similar function as accelerator and diffuser may be ascribed to
the appearance of scientific journalism in the nineteenth century.
There are three phases of this latter-day diaspora, according to the kind of
evidence which may be found from the catalogues of journals published by the
British Museum. In 1800 there were less than ten scientific journals published in
the British Isles. By 1900 there were over a hundred and thirty. The first quarter-
century saw the appearance of new general scientific journals, but their numbers
are fairly constant (about fifteen) from 1830 onwards. From 1825, the main
growth is in specialized and technical journals; there were three in 1825 and over
forty by 1860. From mid-century onwards the published transactions and jour-
nals of learned societies begin to grow rapidly in numbers, from a dozen in 1850
to about seventy in 1900.’5
Not only did books and journals appear at an accelerating rate, but clubs
and institutes spread the new learning to the utmost limits of literacy in industrial
Britain. Attendance at lectures by eminent scientists became as obligatory in the
manufacturing towns as was attendance at church and chapel in the country; in-
terest in science, even to the point of patronizing individual scientific workers or
building a private laboratory, became gentlemanly. In short, the institutional proc-
ess which we know as technology-the linking of ( a )knowledge of the laboratory
demonstrations which established scientific hypotheses with ( b ) knowledge of
manufacturing operations, and of these two with (c) knowledge of existing or pre-
sumptive demand for goods and services-this process was spread at random
among a very large proportion of the literate population. The fact that it was so
spread meant that innovations might appear almost anywhere, might be lighted
upon by almost anyone.
The incoherence of the social institutions of technology at the time was re-
flected in the rudimentary social forms by which innovations were socialized.
Technological changes occurred largely through the birth and death of organiza-
tions, the simplest form of institutional change. Capital was plentiful, liquid, and
diffused, and was readily available for the exploitation of a new device or prod-
uct. But the institutional build-up around the invention was normally rigid and
was identified closely with the line of application and development originally con-
ceived. New concerns had a fairly restricted expectation of life, even if they sur-
vived the highly lethal period of early infancy; “clogs to clogs in three
generations” was a piece of proverbial wisdom current in the oldest factory area
in the world. The new devices which arose to render the old ones obsolete were
generally exploited by new concerns. As Elton Mayo has remarked, the small
scale of business enterprises allowed this change to take place without too much
dislocation of the social and economic order.l6
Invention could, and did, make big fortunes in astonishingly short times.
The supply of risk capital was relatively enormous. Technical limitations to large
84 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

lot production were such that the build-up of production had to be slow, and the
manufacture of a single device, progressively improved, could absorb all the ener-
gies and resources of its designer and backers over a period of years. In this milieu
the economic, social, and psychological pressures were all against the organiza-
tion of research as an industrial resource, and against instituting invention in a
professional salaried occupation. The anecdote of Ferranti’s weaning from the
post he entered at Siemens when he left school reflects the ethos of technical inno-
vation at that time. Ferranti, at the age of seventeen, had invented his first alterna-
tor. A meeting with another engineer, Alfred Thompson, led to an introduction to
a London barrister:

“And you mean to tell me you’re content to be at Siemens’,” he said, “earn-

ing € 1 a week! Good God!”
Lawyers see life on the seamy side; small wonder then that they be-
come suspicious of all men.
“Ferranti,” he said, “if you continue at a job like that, I’ll tell you
what will happen. As soon as they discover you’ve got an inventive ability
they’ll offer you f.5 a week and proceed to rob your brains. You’ll do the in-
venting and they’ll collect the cash.”

This was rather bewildering, but it chimed in with certain thoughts

that had arisen in Ferranti’s own mind.
“Perhaps I’d better ask for a rise,” he suggested.
“For God’s sake, don’t do anything of the sort,” Francis lnce advised.
“Justclear out. That’s no place for you. You might stay there till your teeth
fall out and never get a dog’s chance to doing anything. There’s only one
thing for you to do. You must start right away on your own. ”

Ferranti objected that he had no capital.

“Leave that to me, ’’ said his new friend.”

But even before Ferranti had his conversation with Ince, the situation was
changing. The distribution of scientific information was rapidly becoming organ-
ized. The English provincial universities were founded in the second half of the
nineteenth century; the major scientific and professional societies were created
during the same period.

Professional Scientists and Technologists

By the end of the century, science was the province of groups of specialists
working in and supported by universities or quasi-academic institutions. The
unity of natural philosophy became separated into departments of chemistry,
physics, geology, and later derivatives and hybrids. Information was organized in
the form of textbooks and courses; traditions as to what was relevant and irrele-
vant were created under the authority of qualifying examinations. The intellectual
segregation of scientific specialists was promoted by the way in which the new
and reformed universities organized studies and teaching. Exchanges of the kind
The Organization of Innovation 85

which had been characteristic of the earlier social milieux tended, outside the de-
partmental enclave, to become attenuated and formalized in the meetings and
journals of learned societies, where geologists produced papers for other geolo-
gists, physicists communed with physicists, and so on. By 1900 scientists were
salaried professional men.
On the other hand, the situation of industrial technology was itself chang-
ing. When, before the middle of the century, the major scientific discoveries had
been and could be the work of gifted amateurs and a few academic scientists, tech-
nically competent craftsmen like Maudslay, Nasmyth, and Whitworth had created
the machine-tool industry. The engines and machines that were the showpieces of
the 1851 Exhibition were largely the work of skilled mechanics and master men
who had matched the opportunities presented all around them with the basic
training of their apprenticeship, self-acquired mathematics, and a clear grasp of
the principles of the new engineering. Yet even then, the development by improve-
ment and new application was becoming a task beyond the capacity of men
trained according to traditional craft methods. The outclassing of British products
by European competitors at the 1867 Paris Exhibition made this quite explicit.
The Royal Commission appointed thereafter to survey technical progress in a
number of countries confirmed the impression that Britain had lost, or was losing,
the technical lead established in the previous hundred years.
In Britain the answer to the problem was sought in improving and expand-
ing the educational system. It is sought there now; it always is. One may remark
at this point the different course followed in Germany. With social distinctions in
many social, political, and economic fields more rigid and often more crippling
(given the course then set for Western societies) than those prevailing in Britain,
yet in one generation Germany overhauled and at many points outdistanced the
technical advance of British industry. It was very puzzling.‘ Perhaps the clue to
this sudden acceleration lies in the alliance between the new ethos of nationalism
with science and technology, as the other presumptive heirs to the future; the cult
of Reason in revolutionary France had set the fashion. The alliance was the ortho-
dox basis of progressive ideas all over Europe, but the arrest of political liberalism
in Germany, and its later asphyxiation, may have channelled aspirations and ef-
fort much more powerfully in the direction of scientific and technical achieve-
ment. Whatever the reason, there is little doubt that the rise of German industry
was the consequence of the energy and enthusiasm with which academic scientists
like Liebig and the members of the Berlin Physical Society preached their technical
gospel and, in the case of the Siemens brothers, themselves created industrial
empires.I9Given this kind of liaison, the appropriate educational system followed.

*It still is. Sir Charles Snow, in his 1959 Rede Lecture, remarked “The curious thing was that in
Germany, in the 1830’s and 1840’s, long before serious industrialization had started there, it was
possible to get a good university education in applied science, better than anything England or the
U.S. could offer for a couple of generations. I don’t begin to understand this: it doesn’t make social
sense: but it was SO.’”’
86 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

Without it, as in England, the educational system which was devised-in imita-
tion, as it was thought-widened the breach between science and industry.
As the numbers of scientists rose with the foundation of the provincial uni-
versities and university colleges, another educational system was devised “to meet
the needs of industry for technical training.” Graduate scientists went for the most
partZoto teach in the schools and universities. For industry, there were the poly-
technics and technical colleges, trade schools and evening institutes,
So with the founding of the provincial universities and university colleges, a
parallel network of polytechnics, technical colleges, and evening institutes was
created. A central examining body for technical subjects was provided in the City
and Guilds Institute. By 1902, when local government authorities became largely
responsible for all education below university level, the main structure of a sepa-
rate educational system “to meet the needs of industry for technical training” was
established and lasted for the first half of the present century.
The whole context of industrial innovation had changed. Before 1850 the
worlds of science and industry, though separate, had not been distinct; the very
existence, on such a large scale, of amateur scientific and technical enquiry dem-
onstrates the ease of access to the world of science enjoyed by anyone with inter-
ests which might be satisfied by scientific information. By 1900 science and
industry were distinct social systems, entered by different routes, and with very
few institutional relationships by which people or information could pass between
them. And by 1900, says Cardwell, “The new applied science industries had left
this country, or else had never been started here. In the natural sciences it could
hardly be doubted that the lead was Germany’s, while in technology the enormous
possibilities of the internal combustion engine, for example, were being developed
by the French, the Germans, and the Americans. . . . Lockyer, writing in 1901,
compared our position at the beginning of the new century with what it had been
in 1801, a t the outset of the railway age-now, the chief London electric railway
was American.”20(p. 147).

The New Technological System

Eventually, with a continuing need for the gap to be bridged, new social in-
stitutions have been developed. The gap became itself a new territory, explored,
mapped, and eventually controlled by new specialists, the professional technolo-
gists, going by the name of applied scientists or industrial scientists.
Leaving out of account the prior development in Germany of a liaison initi-
ated and purposefully maintained by the scientists themselves, the first successful
institution set up to exploit this new territory was not only outside Great Britain,
where the worst effects of the separation were experienced, but independent of
both industry and established scientific institutions. Edison’s Menlo Park
Laboratory, employing a hundred workers, was established in 1870. But the en-
trepreneurial method followed contemporary practice: the concerns to manufac-
ture the new devices were set up and financed as separate ventures.
The Orgariization of Innovation 87

This earliest model was not followed until the founding of the Department
of Scientific and Industrial Research by the British Government in 1917. And until
the years immediately before the first World War, very little had been done by in-
dustry, apart from chemicals,” to provide the link itself.
Ever since 1918 the development of industrial research in Britain has de-
pended on Government action much more than in the United States. This may be
attributed, as it usually is, to the unenterprising character of British industry in the
fields of technical development, to the unwillingness of entrepreneurs to divert re-
sources to development work as being too risky. Yet there were exceptions be-
tween 1900 and 1938, notably in chemicals, the industry which had learned from
German methods and technological organizations, and enjoyed most stability;
and the case is now altered.
It is in the situation of industry as it was in the first decades of the century
rather than in such hazy ineluctables as national character that the explanation
lies; for such over-caution, such reluctance to take the profit-making opportunities
latent in new scientific discoveries, can be explained only by the ignorance of the
run of industrialists about the utility of contemporary scientific activity, by the
lack of effective means of communication between the two worlds.
By the end of the first World War the need for such communication was
publicly acknowledged. Since industry itself was not supplying the intermediary
technologists, the Government set up, in 1917, the Department of Scientific and
Industrial Research. Between the wars also, the Government supply of intermedi-
ary resources increased very considerably with the need to assure the translation
of new inventions with military applications into manufactured weapons. It was
from these sources that most of the industrial research and development effort in
contemporary Britain has grown.
In 1938 Bernal put the amount spent on industrial research, apart from
Government expenditure, at L2,000,000, a figure which possibly includes expen-
diture on routine testing. From a survey carried o u t by D.S.I.R. in 1955, it was es-
timated that British firms spent L183 million on research and development during
that year.2’
The Report on Scientific and Engineering Manpower in Great Britain, 1956,
put the total numbers of qualified scientists and engineers employed in industry
on research and development as 22,000. There is no comparable pre-war figure,
but if the figures of scientific staff employed by Defence Ministries and by the De-

“The chemical industry had long before this incorporated scientific laboratory work as part of the
normal organization of the business concern, but apart from the notable association of Lawes and
Gilbert in fertilizer production, the function of the laboratory seems to have been, what it still is in
the smaller chemical concerns, to test the product, and control and refine the processes. As in other
branches of industry in the nineteenth century, discovery was normally the starting-point of new
concerns which exploited it, but firms did not set aside resources of capital and technically qualified
people to search for further innovations. In Britain the change came at the end of the century with
Brunner and Mond, with the United Alkali Company, and with Nobel, i.e., with the stabilization of
the industrial concern.
88 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

partment of Scientific and Industrial Research are taken as a guide, the story is
plain enough. Scientific staff were first employed in peacetime by the Army, Navy,
and Air Force in the early 1920’s. In 1935 the total number employed was less
than 500. By 1939 there were 2,000, and by 1951, 15,000. D.S.I.R., which em-
ployed less than 300 salaried staff in its research establishments in 1920, had
about 1,000 in 1938, & almost 3,500 by 1950.**
The work of producing innovations is now largely in the hands of salaried
professionals employed in industrial firms, government establishments, or in insti-
tutions directly dependent on industry or government for funds.
Technologists (industrial scientists) are now normally people who have
graduated in science at a university or a technical college and have thereafter
served a period as junior members of a development or design team in an indus-
trial, governmental, or other laboratory. The essential factor is not who employs
them, but membership of the appropriate system of communication-electronics,
biochemicals, fibres, nucleonics, metallurgy, aeronautics-through which flows
information which may contribute to the development of any individual innova-
tion. The comparative independence of technology from industry is reflected in
the comparative independence of the technologist. The status of the technologist
is a professional one, and there is fairly precise equivalence between ranks as well
as salaries in the types of organizations employing technologists. They move
freely, and would like to move much more freely, between posts in governmental,
university, and industrial establishments. The career is not enclosed by the pres-
sure either of “loyalty to the firm” or of “best prospects” within the individual
firm.
For the individual firm the technologist is an alien element; he does not fit
into the factory system in the same way as other functional specialists, since these
are no more than bits of the general management-entrepreneurial function. The
actual information held by the technologist, as well as his training and skill, has
value outside the firm. This lies at the bottom of the differences in manners, be-
haviour, dress, and language which so clearly distinguish him from the other
members of the firm which employs him.

THE ROLE OF INDUSTRIAL RESEARCH AND

DEVELOPMENT
Yet technological progress has become of vital concern for the individual
firm in many industries, and the increasing pace of innovation makes it inevitable
that the firm provide more and more support for research and development as a
condition of its own survival. This is not only because other sectors of industry
have become “infected” by the work of government establishments, or even be-
cause industrialists have experienced the profitability of such work, and have
overcome their inhibitions about scientific work. There is also a fundamental
change in the institutional character of the industrial firm.
The Organization of Innovation 89

The most familiar aspect of this change is in scale, which is a function of al-
terations in the balance of production and of consumption within the economy.
Mass markets have created, and in turn been created by, techniques of mass pro-
duction; the use of such techniques has made possible certain economies by mere
increase in the size of plant.
Secondly, concurrently with increase in scale, there has developed a separa-
tion between ownership and control, between the holding of shares and the con-
trol of the policy and activities of a company by management itself or by holders
of a minority of sharesSz3This tendency is held to be as inherent in the structure of
capitalist enterprise as is the tendency towards monopoly in the economy, arising
as it does out of the division between the ownership and the use of property.24(p.
244)25(p. 15).During earlier periods of capitalism economic power resided with
the owners of the property, i.e., the shareholders-although in law such owners
merely possess documents which give them certain claims against the company,
which formally has full ownership. During the present century, however, power
has been passing more and more into the hands of the management, of the direc-
tors of enterprise. Shares are commonly dispersed among multitudes of small
shareholders whose joint influence does not compare with that of a single com-
pact minority interest. The technical and administrative complexities of modern
large-scale enterprise have transformed the relationship between the shareholding
owner and the manager of productive capacity.
Both these developments have affected the character of the industrial con-
cern. Their influence on the internal organization has been considerable. Increased
size has made necessary the division of the general task of management into a
multiplicity of individual tasks, each of which has become the province of special-
ists-salesmen, cost accountants, works managers, designers, planners, secretaries
trained in company law, personnel managers, production engineers. Greater ad-
ministrative complexity, bigger size, and the development of the specialist skills
called for have both aided and been promoted by the shift of control from owner
to manager.
A significant fraction of resources in Britain have become concentrated and
comparatively inelastic. Too much capital, and, more important, too many social
commitments are involved in industrial concerns for change to occur through the
elementary birth and death cycle usual a hundred years ago. Firms employing
many thousands of people cannot close down without wrecking large areas of so-
cial organization. Such concerns must keep alive, and in order to keep alive they
must become adaptive; change must occur within the organization and not
through its extinction and replacement, if it is to occur at all.
Survival of the individual firm becomes a more significant criterion of eco-
nomic activity the closer the approximation to monopolistic conditions. Keirstead
introduces a lengthy exposition of actual pricing policies employed by a “giant
multi-product corporation’’ as follows: When we define time concretely in terms
of the processes of which it is constituted, we are obliged to ask whether the firm
aims at a maximum temporal rate of profit. . . or whether it aims to obtain the
largest estimated profit over some (indefinite)period of time. I am convinced that
 KNOWLEDGE IMANAGEMENT AND ORGANIZATIONAL DESIGN

the latter notion is . , , more in accord with real facts of actual situations. . . . The
maximization of profit at any moment may result in the appearance of competi-
tors whose supplies would reduce price to the point where total net profits over a
sufficient period would be reduced below what they might have been had a lower
rate been accepted and the potential competitors kept out of the market.”26
Directly one introduces time as a function of the profit-maximizing assump-
tion, it is obvious that almost every consideration tends to become subordinate to
survival. Directly, that is, the realities of industrial enterprise are organized in
terms of the individual firm rather than of the individual entrepreneur, then al-
most any profit terms upon which the firm can survive become preferable to
grosser profits on which it might possibly not survive. There is, in fact, no change
in the logical basis, but merely in the way in which it works: ( a )through individ-
ual mortals, ( b )through corporations which are relatively potentially longer-lived.
For an individual entrepreneur, profit-taking can be maximized for any period of
time however short, since the rewards will certainly be a substantial help towards
his own survival. Moreover, for the individual the random sector of circumstances
affecting his strategies increases enormously with time. And he makes hay, there-
fore, when the sun shines, and a bird in the hand is worth two in the bush. For the
corporation, randomness does not increase at anything like the same rate. And
survival means only survival of the firm. The birds in the bush, which are tomor-
row’s or the next ten years’ production, are just as important as that in the hand.
To sum up: two major changes have occurred in the social circumstances
affecting the production of innovations. First, industrial concerns have increased
in size: ever greater administrative complexity has brought a wide range of bu-
reaucratic positions and careers into being; control has moved from owners to
management. Their survival is therefore a matter of much more intense and wide-
spread concern to themselves and to society; the chances of survival are improved
if the technical innovations which might render its processes or products obsolete
are developed within it and not by newcomers.
The other change has occurred in the form of institutional relationships and
roles within which invention has been possible. The familiar and sociable relation-
ships typical of the eighteenth century provided the ease of communication neces-
sary for the major syntheses of ideas and requirements which introduced the early
revolutionary inventions. The scale of scientific and industrial activity rapidly out-
grew the social institutions within which the Industrial Revolution was generated;
the syntheses which produced inventions and innovations tended to be random or
opportunistic. Later in the nineteenth century, new institutional forms introduced
barriers between science and industry, and between “pure” and “applied” science,
as well as between departments of science. In the twentieth century the new and
elaborate organization of professional scientists has been eventually matched by
one of technical innovators into groups overlapping teaching and research institu-
tions, Government departments and agencies, and industry.
Neither change is complete. Neither set of contrasts is clear. Few sectors of
industry, outside chemicals, have fully accepted the changed situation. It was still
possible, in the years between the wars, for a major innovation like the gas turbine
The Organization of l?inovatioii 91

to be developed in ways reminiscent of the classic days of nineteenth-century

back-parlour invention. The jet engine’s invention depended on an individual’s
persistence and enterprise, although the new massive organizations of government
and industry were also involved.27On the other hand, the career of the most pub-
licized inventor of the nineteenth century, Edison, reflects both the previous ep-
och, in the almost conscious exploitation of sociable contact with scientists and
technologists, and the later, in the maintenance of development groups and the
opening of professional careers in invention. Yet the process of change is now far
enough advanced for the shape of the forms characteristic of the present system to
be discernible.

REFERENCES
4. Ogburn, W. E Social Chnnge. New York: Viking Press, 1922.
5. Boulding, K. E. The Organizational Revolution. New York: Harper, 1953.
6. Marx, K. “Letter to P, V. Ennenkov,” 1846. In: Karl Marx and Frederick Engels, Se-
lected Works, Vol. 11. London: Lawrence and Wishart, pp. 401-2.
7. Durkheim, E. De la Division du travail social. 1893. Trans.: On the Division of La-
bour in Society, by G. Simpson. Glencoe, Ill.: Free Press, 1954.
8. Tonnies, E: Gemeirischaft und Gesellschaft, Leipzig, 1887. Trans.: Fundamental Con-
cepts ofSociology, by C. I? Loomis. New York: American Book Company.
9. Jewkes, J. “How much Science.” Presidential Address to British Association, Eco-
nomic Section, 1959. Ecoizomic]ournal, No. 277. March 1960, 70, p. 12.
10. Whitehead, A. N. Science arid the Modern World. London: Cambridge Univ. Press
1926. (7th Impression, 1933, p. 120.)
1 1 . Bright, A. A. The Electric Lamp Industry: Technological Change and Economic De-
velopment from 1880 to 1947. London: Macmillan, 1949..
12. Maclaurin, W. R. Invention and Innovation in the Radio Industry. New York:
Macmillan, 1949.
13. Clow, A., and Clow, N. The Chemical Revolution. London: Batchworth, 1952 (pp.
593-4 ) *
14. Smiles, S. Life of Boulton and Watt. London: Murray, 1865 (p. 367).
15. British Museum Catalogue of Printed Books, Vol. 36 (Periodicals: Enlarged Edn.),
1899.
16. Mayo, E. The Social Problems of an lndustrial Civilization. London: Routledge, 1949
(P. 32).
17. Ferranti, G . Z. De, & Ince, R. The Life and Letters of Sebastian Ziani de Ferranti.
London: Williams and Norgate, 1934 (pp. 51-2).
1 8 . Snow, C. P. The Two Cultures and the Scientific Revolution. London: Cambridge Uni-
versity Press, 1959.
19. Bernal, J. D. Science and Industry in the Nineteenth Century. London: Routledge,
1954 (pp. 63-4).
92 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

20. Cardwell, D. S. L. T h e Organization of Science in England. London: Heinemann,

1957.
21. Department of Scientific and Industrial Research: Estimates of Resources devoted t o
Scientific and Engineering Research and Development in British Manufacturing In-
dustry, 1955. H.M.S.O., 1958.
22. Treasury: Civil Estimates, 1918-51.
23. Berle, A. A., Jc, and Means, G. C. T h e Modern Corporation and Private Property
London: Macmillan, 1932.
24. Friedmann, W. Legal Theory. London: Stevens, 1949 (p. 244).
25. Friedmann, W. L a w and Social Change in Contemporary Britain. London: Stevens,
1951 (p. 15).
26. Keirstead, B. S. T h e Theory of Economic Change. London: Macmillan, 1948 (p. 254).
27. Whittle, H. l e t . London: Muller, 1954.
 7
When a Thousand Flowers
Bloom: Structural,
Collective, and Social
Conditions for Innovation
in Organizations
Rosabeth Moss Kanter

“Let a thousand flowers bloom.” This slogan, designed to awaken an entire na-
tion to new ideas, offers an apt metaphor for innovation. Innovations, like flow-
ers, start from tiny seeds and have to be nurtured carefully until they blossom;
then their essence has to be carried elsewhere for the flowers to spread. And some
conditions-soil, climate, fertilizer, the layout of the garden-produce larger and
more abundant flowers.
Innovations can grow wild, springing up weed-like despite unfavorable cir-
cumstances, but they can also be cultivated, blossoming in greater abundance un-
der favorable conditions. If we understand what makes innovations grow-the
microprocess by which they unfold-we can see why some macro-conditions are
better for their cultivation.
It is increasingly common among writers to emphasize the nonlinear, slightly
chaotic, usually sloppy, sometimes random, and often up-and-down nature of in-
novation (Quinn, 1985). Taken to an extreme though, as some popular writers
have done, it might be tempting to conclude that it is impossible to plan for inno-
vation, manage it, or design an organization structure to support it. This extreme
viewpoint holds individual variables like creativity and leadership to be more im-
portant than structural variables and, indeed, tends to see organizations in general
as negative forces, with innovations generally occurring despite the organization,
through accidents, lucky breaks, and bootlegged funds.
My own conclusion, after systematic comparative research (Kanter, 1983),
in depth fieldwork, and literature review, is more moderate. Organizational con-

From Research in Organizational Behavior, Vol 10: 169-211. Copyright 0 1988 JAI Press Inc. Re-
printed with permission.

93
94 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

ditions-structure and social arrangements-can actively stimulate and produce

innovation, as long as those conditions take into account the “organic,” “natu-
ral,” and even the “wild” side of innovation. Innovation is the creation and ex-
ploitation of new ideas. At its very root, the entrepreneurial process of innovation
and change is at odds with the administrative process of ensuring repetitions of
the past. The development of innovation requires a different set of practices and
different modes of organization than the management of ongoing, established op-
erations where the desire for or expectation of change is minimal. Stevenson and
Gumpert (1985) have cast this management difference in terms of the contrast be-
tween the “promoter” type stance of the entrepreneur, driven by perception of op-
portunity, and the “trustee”-like stance of the administrator, driven to conserve
resources already controlled (see also Hanan, 1976). Structures and practices that
may work well for the perpetuation of the known tend to be a t odds with innova-
tion.
Innovation-whether technological or administrative, whether in products
or processes or systems-tends to have four distinctive characteristics (Kanter,
1985).

1. The innovation process is uncertain. The source of innovation or the oc-

currence of opportunity to innovate may be unpredictable. The innova-
tion goal may involve little or no precedent or experience base to use to
make forecasts about results. Hoped-for timetables may prove unrealis-
tic, and schedules may not match the true pace of progress. “Progress on
a new innovation,” Quinn (1979) wrote, “comes in spurts among un-
foreseen delays and setbacks. . . in the essential chaos of development.”
Furthermore, anticipated costs may be overrun and ultimate results are
highly uncertain. Indeed, analysts have variously estimated that it takes
an average of 10 to 12 years before the return on investment of new ven-
tures equals that of mature businesses (Biggadike, 1979); 7 to 15 years
from invention to financial success (Quinn, 1979); and 3 to 25 years be-
tween invention and commercial production (Quinn, 1985).
2 . The innovation process is knowledge-intensive. The innovation process
generates new knowledge intensively, relying on individual human intelli-
gence and creativity and involving “interactive learning” (Quinn, 1985).
New experiences are accumulated a t a fast pace; the learning curve is
steep. The knowledge that resides in the participants in the innovation ef-
fort is not yet codified or codifiable for transfer to others. Efforts are very
vulnerable to turnover because of the loss of this knowledge and experi-
ence. There need to be close linkages and fast communication between all
those involved, at every point in the process, or the knowledge erodes.
3 . The innovation process is controversial. Innovations always involve
competition with alternative course of action. The pursuit of the air-
cooled engine at Honda Motor, for example, drew time and resources
away from improving the water-cooled engine. Furthermore, sometimes
the very existence of a potential innovation poses a threat to vested inter-
ests-whether the interest is that of a salesperson receiving high commis-
When a Thousand Flowers Bloom 95

sions on current products, or of the advocates of a competing direction.

(Fast, 1979, for example, argues that “political” problems are the pri-
mary cause for the failure of corporate New Venture Departments.)
4. The innovation process crosses boundaries. An innovation process is
rarely if ever contained solely within one unit. First, there is evidence that
many of the best ideas are interdisciplinary or interfunctional in origin-
as connoted by the root meaning of entrepreneurship as the development
of “new combinations”-or they benefit from broader perspective and
information from outside of the area primarily responsible for the inno-
vation. Second, regardless of the origin of innovations, they inevitably
send out ripples and reverberations to other organization units, whose
behavior may be required to change in light of the needs of innovations,
or whose cooperation is necessary if an innovation is to be fully devel-
oped or exploited. Or there may be the need to generate unexpected in-
novations in another domain in order to support the primary product,
like the need to design a new motor to make the first Apple computer
viable.

If innovation is uncertain, fragile, political, and imperialistic (reaching out

to embrace other territories), then it is most likely to flourish where conditions al-
low flexibility, quick action and intensive care, coalition formation, and connect-
edness. It is most likely to grow in organizations that have integrative structures
and cultures emphasizing diversity, multiple structural linkages both inside and
outside the organization, intersecting territories, collective pride and faith in peo-
ple’s talents, collaboration, and teamwork. The organizations producing more in-
novation have more complex structures that link people in multiple ways and
encourage them to “do what needs to be done” within strategically guided limits,
rather than confining themselves to the letter of their job. Such organizations are
also better connected with key external resources and operate in a favorable insti-
tutional environment.
Not all kinds of innovation appear everywhere in equal proportions of
course. Product innovations are more likely in new entrant organizations and
process innovations in established ones. Product innovations are more common in
earlier stages of a product’s history; process innovations in later stages (Abernathy
& Utterback, 1978).Technological innovations are more frequent when resources
are abundant; administrative innovations when resources are scarce (Kimberly,
1981). Evolutionary innovations (modest, incremental changes) are more likely in
organizations that are more formalized and “centralized”; more revolutionary in-
novations in organizations that are more complex and “decentralized” (Cohn &
Turyn, 1984). But in general, the overall rate of innovation across types should be
associated with the circumstances I have outlined.
Some of these structural and social conditions are more important at some
points in the innovation process than at others. Like the flowers whose cultivation
requires knowledge of its growth pattern, so does the understanding of innovation
benefit from examining structural and social facilitators as they wax and wane
with the innovation development process. This requires a dynamic model, a com-
96 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

bination of a “variance” model of the factors influencing innovation and a “proc-

ess” model showing how innovation unfolds (Mohr, 1978).
Recent research examining sets of innovations as they unfold over time
(Schroeder, Van de Ven, Scudder, & Polley 1986; Van de Ven, 1986) has discred-
ited the usual process models of innovation that posit discrete stages through
which an innovation idea progresses. I agree that stage models d o not always ade-
quately capture the give-and-take of innovation, and they risk artificially segment-
ing the process. But I propose that the structural and social conditions for
innovation can be understood best if the innovation process is divided into its ma-
jor tasks.
There are four major innovation tasks, which correspond roughly (but no-
where near exactly) to the logic of the innovation process as it unfolds over time
and to empirical data about the history of specific innovations. These tasks are:
( a ) idea generation and activation of the drivers of the innovation (the “entre-
preneurs” or “innovators”); (b) coalition building and acquisition of the power
necessary to move the idea into reality; (c) idea realization and innovation produc-
tion, turning the idea into a model-a product or plan or prototype that can be
used; (d) transfer or diffusion, the spreading of the model-the commercialization
of the product, the adoption of the idea.
While sometimes occurring in sequence, these tasks also overlap. But by un-
derstanding the nature of each task, we can see more easily why certain properties
of organizations are related to the success of innovation. This, in turn, contributes
to our knowledge of the relationship between structure and behavior, between
macro-context and micro-process.

IDEA GENERATION AND INNOVATION ACTIVATION

Innovation begins with the activation of some person or persons to sense or

seize a new opportunity. Variously called “corporate entrepreneurs” (Kanter,
1983), “intrapreneurs,” “idea generators,” or “idea champions” (Galbraith,
1982), such individuals are able to initiate a process of departing from the organi-
zation’s established routines or systems.
Innovation is triggered by recognition of a new opportunity. Once the op-
portunity is “appreciated,” as Van de Ven (1986) put it, someone needs to supply
the energy necessary to raise the idea over the threshold of consciousness, much as
Schon (1971) described the emergence of new public policies as a result of being
pushed into awareness. The first key problem in the management of innovation,
then, is how to get people to pay attention-how to trigger the action thresholds
of individuals to appreciate and pay attention to new ideas, needs, and opportuni-
ties.
Drucker (1985) has argued that the opportunities that give rise to innova-
tion lie in incongruities and discontinuities-things that d o not fit expected pat-
terns or that provide indications that trends may be changing. But unless we are to
When a Thousand Flowers Bloom 97

assume these are purely individual cognitive abilities, it is important to look at the
structural conditions that facilitate the ability to see new opportunities.

Close Connection with Need Sources

Opportunity exists because need exists, so it is not surprising that close cus-
tomer or user contact is an important innovation activator. An often cited na-
tional study found that over three-fourths of a set of 500 important industrial
innovations owed their origins to user suggestions and even user invention; only
one-fifth originated in technical ideas looking for a home (Marquis & Myers,
1969). Users had originated 81% of the innovations in scientific instruments in
another study, and 60% of those in process machinery (von Hippel, 1981).
Effective innovation thus derives from active awareness of changing user
needs and sometime from direct user demands or solutions. Therefore, structural
arrangements and social patterns that facilitate contact across boundaries, be-
tween potential innovators and their “market,” help produce more innovation.
Potential innovators benefit from being linked directly to the market, to gain a
fuller personal appreciation for what users need, as well as from being connected
with those functions inside the organization that manage the interface with the
outside. Quinn (1985) found that high innovation companies in the United States,
Japan, and Europe were characterized by a strong market orientation a t the top of
the company and mechanisms to ensure interaction between technical and mar-
keting people at lower levels. At Sony, for example, new technical hires were as-
signed to weeks of retail selling as part of their orientation. In the prosperous
years for People Express Airlines, the incentive system was designed to ensure that
all executives spend at least some time each year flying as crew on their planes.
Van de Ven (1986) hypothesized that direct personal confrontations with
problem sources are needed to reach the threshold of concern and appreciation re-
quired to motivate people to act. Perhaps this is why it has been observed that
well-managed companies search out and focus on their most demanding custom-
ers, not the ones who are easily satisfied. Similarly, successful examples of innova-
tions offered by managers in high technology firms tended to involve radical
redefinition of the product or service as a result of encounters with the “real
world” of customers or users-direct, first-hand experience of their need (Delbecq
& Mills, 1985).
Raytheon’s New Products Center demonstrates this principle in action. The
center services a series of consumer products divisions, and also it has two levels
of “users” and need sources: its internal divisional customers and the ultimate ex-
ternal consumer. Center practices involve frequent visits and tours to all of these
need sources. Technical staff routinely attend trade shows, tour manufacturing fa-
cilities, and browse at retail outlets, striking up conversations with consumers.
Extra-organizational ties with users can be formalized, to ensure continuing
close connection. Many computer and software companies have formed user
groups, which allow them to gather ideas for new products and product improve-
98 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

ments. Some manufacturing application laboratories solicit proposals from their

customers for things they might work on (von Hippel, 1981).
These principles apply to internal administrative or organizational innova-
tions as well as technological or product innovations. I propose, based on field
observations, that those staff groups successful a t creating innovations are the
ones with the closest connections with the needs in the field; Honeywell’s corpo-
rate human resources staff, for example, has created “councils” of key executives
to ensure the continuing relevance of its offerings to the changing needs of users.
In general, then, innovation activation benefits from structural or social con-
nections between those with the technical base (potential innovators) and those
with the need (potential users). Indeed, one research group found that a higher
proportion of new products failed for “commercial” reasons (misreading the
need) rather than technical ones, indicating a poor interface between developers
and users (Mansfield, Rapoport, Schnee, Wagner, & Hamburger, 1981).

“Kaleidoscopic Thinking”: Cross Fertilization

Awareness of need is one element; ability to construct new ways to address

the need is a second. I have come to refer to the creativity involved in activating
innovaticn as “kaleidoscopic thinking” (Kanter, 1986).
The kaleidoscope is an apt metaphor for the creative process, because the
kaleidoscope allows people to shake reality into a new pattern. In a kaleidoscope
a set of fragments form a pattern. But the pattern is not locked into place. If the
kaleidoscope is shaken or twisted, or the angle of perspective is changed, the same
fragments form an entirely new pattern. Often, creativity consists of rearranging
already existing pieces to create a new possibility. For example, Malcolm McLean
did this about 30 years ago when he developed the concept for Sea-Land, the first
company to offer containerized shipping. Before Sea-Land, shipping was a tedious
matter of packing and unpacking crates in order to move objects from one form
of transportation to another. McLean’s innovation was simple: move the whole
container.
Contact with those who see the world differently is a logical prerequisite to
seeing it differently ourselves. “Cosmopolitan” rather than ‘‘local’’ orientations-
seeing more of the world-has been identified by many researchers as a factor in
high rates of innovation (Rogers & Shoemaker, 1971). So the more innovative or-
ganizational units who face outward, as well as inward, taking in more of the
world around them, and taking better advantage of “boundary spanners” to
bring them intelligence about the world beyond (Robertson & Wind, 1983; Tush-
man, 1977). High-performing research and development (R&D) project groups
have far greater communication with organizational colleagues outside the group
than low-performing teams (Allen, 1984); sometimes this communication occurs
in two steps, mediated by certain communication “stars” who then transmit it to
the rest of the group (Tushman, 1979).
One classic set of studies of research scientists found that the most produc-
tive and creative ones were those who had more contacts outside their fields, who
When n Thousand Flowers IIloorn 99

spent more time with others who did not share their values or beliefs (Pelz & An-
drews, 1966). At the same time, the dangers of closing otf were also clear. I t took
only 3 years for a heterogeneous group of interdisciplinary scientists who worked
together every day to become homogeneous in perspective and approach to prob-
lems. Sociologists have used the terms “occupational psychosis” and “trained in-
capacity” to describe the tendency for those who concentrate on only one area
and interact only with those who are similar in outlook to become less able over
time to learn new things.
The “twists” on reality causing creativity may derive from uncomfortable
situations where basic beliefs are challenged and alternatives suggested. It is not
surprising, then, that the patterns in most large, established bureaucracies inhibit
rather than activate innovation. Once people enter a field, they spend most of
their time (especially their discretionary time) with other people just like them
who share their beliefs and assumptions. At the top, leaders are increasingly insu-
lated from jarring experiences or unpleasant occurrences that cause them to con-
front their assumptions about the world, and they spend a n increasing portion of
their time with people exactly like themselves. And if corporate culture encour-
ages an orthodoxy of beliefs and a nonconfrontational stance, then idea genera-
tion is further discouraged.
Cross-fertilization of ideas instead comes from cross-disciplinary contact.
Creativity often springs up at the boundaries of specialties and disciplines, rather
than squarely in the middle. I t is often a matter of combining two formerly sepa-
rate ideas-wafers and ice cream making the world’s first ice cream cone. A large
oil company considers one of its greatest innovations the development of a new,
highly useful chemical compound that was created because researchers from two
distinct fields collaborated. Ocean Spray staged a comeback for cranberry juice
because a marketing executive spent time learning about packaging; the company
was the first in its industry to put juice in paper bottles. Some organizations ac-
tively facilitate cross-disciplinary exchange through product fairs or cross-division
“show and tell” meetings or cross-functional teams that visit customers together
(Tushman & Nadler, 1986).
But when departments of specialties are segmented and prevented from con-
tact, when career paths confine people to one function or discipline for long peri-
ods of time, and when communication between fields is difficult or excessively
formal, creativity is stifled. Huge buildings consisting of all those in one field,
physically separated from people in another field, make contact impossible.
Under that kind of circumstance, outsiders may be better able to see the big
picture and take a new angle on the pattern, because they are not yet aware of all
the details the “experts” see that inevitably confirm the view that no change is
possible. People too close to a situation often become hopeless about change,
blind to the possibilities.
Thus, a great deal of important industrial innovation comes from what
Schon (1967) called “innovation by invasion”: a new player enters the game,
bringing a new method or technique. For example, half of all major innovations
in pharmaceuticals from 1935 to 1962 were based on discoveries made outside
the firm that later exploited them (Mansfield et al., 1981). It was Apple that first
100 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

successfully commercialized the personal computer; IBM was a latecomer-and it

is hard to imagine more outsiderlike amateurs than Steve Jobs and Steve Wozniak.
Similarly, in my study of leading companies, newly appointed managers who
came from a different field by an unusual career route-in a word, outsiders-
were somewhat more likely to innovate than those who rose by orthodox means
(Kanter, 1983).
In general, then, contact with those who take new angles on problems facili-
tates innovation.

Structural Integration: Intersecting Territories

Activation of innovation is encouraged by structural integration across

fields-by intersecting territories. Researchers have long observed that “communi-
cation integration” (closer interpersonal contact or connectedness via interpersonal
communication channels in an organization) is positively related to the innovation
rate (Rogers & Shoemaker, 1971; Tushman & Nadler, 1986). Isolation of individu-
als and units tends to reduce innovation at the idea generation stage by limiting
awareness of opportunity, alternative approaches, and the perspective of those
functions who need to contribute other “parts” to make the innovation add up to
a “whole.” (Van de Ven, 1986, considered the management of part-whole relation-
ships one of the four critical innovation tasks.) These who are isolated, in short, are
less attuned to alternatives than those who are well-connected.
“Matrix” organization structures (Davis & Lawrence, 1977)are highly inte-
grative, and it is not accidental that they were first developed to aid technological
innovation-the large-scale development projects in the aerospace industry-and
are found more frequently in rapidly changing, highly innovating organizations
(Kanter, 1983). Matrix organizations, in which mid-level employees report to
both a project boss and a functional boss, force integration and cross-area com-
munication by requiring managers from two or more functions to collaborate in
reaching a decision or taking some action. This is frequently characterized as a
“dotted line” relationship for those in one department to another department, sig-
nifying a working relationship but not always direct authority.
By requiring extensive cross-functional consultation, the matrix diffuses
authority among a group of managers. In many instances, this opportunity can be
used in a positive manner by particularly entrepreneurial managers who are able
to envision alternatives and assume responsibility for pursuing them-alternatives
that cut across territories.
In general, measures of complexity and diversity in an organization are posi-
tively related to initial development of innovations (though they are sometimes
negatively related to eventual acceptance of the same innovation by the rest of the
organization). Diversity gives the individual more latitude for discovery, but may
make it difficult later to get agreement on which many proposals or demonstra-
tion projects should be implemented on a wider scale. Similarly, innovation is
aided by low formalization at the initiation stage, when freedom to pursue untried
possibilities is required.
When a Thousand Flowers Bloom 101

Therefore, to produce innovation, more complexity is essential: more rela-

tionships, more sources of information, more angles on the problem, more ways
to pull in human and material resources, more freedom to walk around and across
the organization (also see Burns & Stalker, 1968; Mintzberg, 1981). One does not
need a formal matrix structure to d o this. Indeed, it is the general characteristics
of an integrative structure that make a difference in terms of encouraging innova-
tion: looser boundaries, crosscutting access, flexible assignments, open communi-
cation, and use of multidisciplinary project teams. So specifying multiple links
between managers in a formal sense (through showing more than one solid-line or
dotted-line reporting relationship on an organization chart) is merely a way of ac-
knowledging the interdependencies that complex products and innovative pro-
jects require.
Dividing the organization into smaller units based on a common end use
goal but not around function or specialty also aids activation of innovation by
producing structural integration at micro-level. When it comes to innovation,
“small is beautiful,” and flexible is even better (see Quinn, 1985). Or at least
small is beautiful as long as the small unit includes all functions or disciplines and
forces contact across them. Cross-fertilization across disciplines and a focus on us-
ers is built into the structure.
The idea of dividing into smaller but complete business units has been ap-
pealing to organizations seeking continual innovation. In smaller business units
it is possible to maintain much closer working relationships across functions than
in larger o n e s - o n e of the reasons for Hewlett-Packard’s classic growth strat-
egy of dividing divisions into 2 when they reached more than 2,000 people or
$100,000,000 in sales. Even where economies of scale push for larger units, the
cross-functional project or product team within a single facility (captured in such
ideas as the factory-within-a-factory) helps keep the communication and the con-
nection alive.

Broad Jobs

Idea generation is also aided when jobs are defined broadly rather than nar-
rowly, when people have a range of skills to use and tasks to perform to give them
a view of the whole organization, and when assignments focus on results to be
achieved rather than rules or procedures to be followed. This, in turn, gives people
the mandate to solve problems, to respond creatively to new conditions, to note
changed requirements around them, or to improve practices, rather than mind-
lessly following procedures derived from the past.
Furthermore, when broader definitions of jobs permit task domains to over-
lap rather than divide cleanly, people are encouraged to gain the perspective of
others with whom they must now interact and therefore to take more responsibil-
ity for the total task rather than simply their own small piece of it. This leads to
the broader perspectives that help stimulate innovation.
In areas that benefit from more enterprise and problem solving on the part
of job holders, broader jobs seem to work better. This is the principle behind work
102 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

systems that give employees responsibility for a major piece of a production proc-
ess and allow them to make decisions about how and when to divide up the tasks.
Pay-for-skill systems similarly encourage broader perspectives by rewarding peo-
ple for learning more jobs (Tosi & Tosi, 1986).
Does this argument conflict with the numerous findings that adoption of in-
novation is more likely in organizations with more specialists and professionals?
(E.g., Hage and Aiken’s [1967] conclusion that the rate of innovation is higher
when there are occupational specialties, each with a greater degree of profession-
alism.) No, because while specialized knowledge is an asset, confinement to a lim-
ited area and minimal contact with other professionals inhibits the ability for
experts to use their knowledge in the service of change.
Potential innovators can become interested in a particular issue that devel-
ops into an innovation for several reasons. The initial impetus for innovation ac-
tivation can stem from ( a ) an obligation of his or her position (March & Olsen,
1979); ( b ) a direct order; (c) a stimulus from the environment or “galvanizing
event” (Child, 1972; Kanter, 1983); ( d ) self-motivated, entrepreneurial behavior;
(e) organizational rewards and payoffs; or ( f ) accidental conditions (Perrow,
1981).
While much of the literature emphasizes the random, spontaneous, or devi-
ant aspects of idea generating, some research has found that the nature of job as-
signments can be an activating force-either directly, because the assignment
requires a new solution, or indirectly by allowing a scanning process to occur be-
yond what is programmed into the position. Job assignments (new ones or simply
those understood as part of the job) stimulated a high proportion (51%)of the in-
novations in one study (Kanter, 1983). Managers did not necessarily have to think
up projects by themselves to begin acting as organizational entrepreneurs; their
enterprise came from accepting the responsibility and finding a way to build
something new while carrying out an assigned task.
What is important is not whether there is an assignment, but its nature:
broad in scope, involving change, and leaving the means unspecified, up to the
doer. In my study, a manager’s formal job description often bore only a vague or
general relationship to the kinds of innovative things the manager accomplished
(Kanter, 1983). Indeed, the more jobs are “formalized,” with duties finely speci-
fied and “codified,” the less innovation is produced in the organization. An em-
phasis on the “numbers” (a quantitative versus a qualitative thrust in jobs) and on
efficiency also depresses the amount of innovation. “LOWformalization,” on the
other hand, is associated with more innovativeness (Hage & Aiken, 1967).
Broad assignments are generally characteristics of staff managers in problem
solving or bridging positions who have a general change mandate to “invent
something” or “improve something.” The innovation-producing companies are
often marked by a large proportion of problem solvers in operating departments
who float freely without a “home” in the hierarchy and thus must argue for a
budget of find a constituency to please. The incentive to enterprise is the luck of
defined tasks (Kanter, 1983). Thus, organizational slack (Galbraith, 1982) and
stack in assignments enables the activation of innovation.
When a Thousand Flowers Bloom 103

The more routinized and rules-bound a job is, the more it is likely to focus
its performers on a few already-known variables and to inhibit attention to new
factors. Starbuck (1983) argued that highly programmed jobs are like supersti-
tious learning, recreating actions that may have little to d o with previous success
or future success. Overly elaborate and finely detailed structures and systems
make organizational participants unable to notice shifts in their environment and
the need for innovation, especially if they are required to send “exceptions” some-
where else for processing.
Where jobs are narrowly and rigidly defined, people have little incentive to
engage in either “spontaneous” innovation (self-generated, problem-solving at-
tempts with those in neighboring tasks) or to join together across job categories
for larger top-directed innovation efforts-especially if differences in job classifi-
cation also confer differential status or privilege. Companies even lose basic effi-
ciency as some tasks remain undone while waiting for the person with the “right”
job classification to become available-even though others in another classifica-
tion may have the skills and the time. And people tend to actively avoid doing any
more work than the minimum, falling back on the familiar excuse. “That’s not my
job”-a refrain whose frequent repetition is a good sign of a troubled company.

Organizational Expectations for Innovation

Even if people are able to generate new ideas in the innovation activation
stage, they must also feel confident that their attempts a t innovation will be well
received. The signals they receive about the expectations for innovation play a
role in activating or inhibiting innovation.
One way organizations signal an expectation for innovation is by allocating
funds specifically for it. In one study comparing innovation successes with fail-
ures, it was found that the failures were handicapped by a lack of resources any-
where other than in already committed operating budgets, while the successes
benefitted from the existence of special innovation funds (Delbecq & Mills,
1985). Despite all the heroic glamour of associating innovation with “bootleg-
ging’’ funds spent on the sly, it is clearly easier to innovate when funds exist for
this purpose.
Since innovations generally require resources beyond those identified in op-
erating budgets (Kanter, 1983) for reasons that are logical-the exact nature and
timing of innovation is often unpredictable-the existence of multiple sources of
loosely committed funds at local levels makes it easier for potential innovators to
find the money, the staff, the materials, or the space to proceed with an en-
trepreneurial idea. Because no one area has a monopoly on resources, there is little
incentive to hoard them as a weapon; instead, a resource holder can have more in-
fluence by being one of those to fund an innovative accomplishment than by being
a nay sayer. Thus, managers at one computer company could go “tin-cupping” to
the heads of the various product lines in their facility who had big budgets, col-
lecting a promise of a little bit of funding from many people (Kanter, 1983). This
104 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

process reduced the risk on the part of all “donors” a t the same time that it helped
maintain the “donee’s” independence.
Sheer availability of resources helps, of course. Research shows that richer
and more successful organizations innovate more than poorer and less successful
ones, especially in technology areas (Kimberly & Evanisko, 1979; Kimberly,
1981; Zaltman, Duncan, & Holbek, 1973).
There are a variety of ways that high innovation companies make resources
accessible locally or give middle-level people alternatives to tap when seeking money
or materials for projects. One is to have formal mechanisms for distributing funds
outside the hierarchy. 3M has put in place “innovation banks” to make “venture
capital” available internally for development projects. Honeywell divisions have top-
management steering committees guiding their organizational-change activities. The
original steering committee solicited proposals quarterly from any employee for the
formation of a problem-solving task team; the teams may receive a small working
budget as needed. Also, “decentralization” keeps operating units small and ensures
that they have the resources with which to act, and thus makes it more likely that
managers can find the extra they need for an innovation locally.
Of course, some innovations, particularly organization ones, can be handled
without money at all. Instead, the most common resource requirement in one
study was staff time (Kanter, 1983). This was also decentralized in the form of
“slack” and local control: people locally available with uncommitted time, or
with time that they could decide to withdraw from other endeavors to be attached
to an appealing project. Because mid-level personnel, professionals, and staff ex-
perts had more control over the use of their time in the more frequently innovat-
ing companies, it was easier to find people to assist in a project, or to mobilize
subordinates for a particular activity without needing constant clearances from
higher-level, nonlocal bosses.
A second general source of expectations for innovation lies in whether the
organization’s culture pushes “tradition” or “change.” Innovators and innovative
organizations generally come from the most modern, “up-to-date” areas rather
than traditional ones with preservationist tendencies, and they are generally the
higher-prestige “opinion leaders” that others seek to emulate (Rogers & Shoe-
maker, 1971; Hage & Dewar, 1973). But opinion leaders are innovative only if
their organizations’ norms favor change; this is why the values of the leaders are
so important. Most people seek to be culturally appropriate, even the people lead-
ing the pack. There is thus more impetus to seek change when this is considered
desirable by the company.
Pride in company, coupled with knowing that innovation is mainstream
rather than countercultural, helps to stimulate innovation (and occurs as a result
of innovation as well). A feeling that people inside the company are competent
leaders, that the company has been successful because of its people, supports this.
For instance, of the companies in one study, Polaroid Corporation knew that it is
the technological leader in its field; Hewlett-Packard prided itself on its people-
centered corporate philosophy, the H-P way, as well as on its reputation for qual-
ity, important in its retention of customers (Kanter, 1983).
When a Thousand Flowers Bloom 105

Such cultures of pride stand out in sharp distinction to the cultures of inferi-
ority that lead less innovating companies to rely on outsiders for all the new ideas,
rather than on their own people.
Success breeds success. Where there is a “culture of pride,” based on high
performance in the past, people’s feeling of confidence in themselves and others
goes up. They are more likely to take risks and also to get positive responses when
they request cooperation from others. Mutual respect makes teamwork easier.
High performance may Cause group cohesion and liking for workmates as well as
result from it (Staw, 1975);pride in the capacity and ability of others makes team-
work possible. In an extension of the “Pygmalion Effect” to the corporation,
supervisors who hold high expectations of subordinate’s abilities (based on inde-
pendent evaluations) may enhance that person’s productivity (Wortman & Lin-
senmeier, 1977).
Thus, organizations with “cultures of pride” in the company’s achievements
and in the achievements and abilities of individuals will find themselves more
innovative. This is why formal awards and public recognition make a difference-
sometimes less for the person receiving them (who has, after all, finished an
achievement) than for the observers in the same company, who see that the things
they might contribute will be noticed, applauded, and remembered.
It is a self-reinforcing upward cycle-performance stimulating pride stimu-
lating performance-and is especially important for innovation. Change requires
a leap of faith, and faith is so much more plausible o n a foundation of successful
prior experiences.
Finally, feeling valued and secure helps people relax enough to be creative, as
Amabile’s (1983)experiments on the conditions facilitating creative problem solv-
ing indicate. Groups were asked to solve problems in one of two conditions, and
the creativity of their solutions was rated. In condition I, they were paid for their
participation before they began to work. In condition 11, which tended to resemble
the corporate norm, they were paid o n a contingency basis, depending on how
well their group performed. In which condition were groups more creative? The
first, the one that can be called a high securitylhigh value condition. Knowing that
they were already paid, members could relax, and they could assume that they
were with a set of talented people. Without the tension that worry about paycheck
might have caused, they could free themselves to be much more creative. Further-
more, they “rose to the occasion”: because expectations for innovation were set
by advance pay, they innovated.

Integration versus Isolation

Overall, I argue that the generation of new ideas that activates innovation is
facilitated by organizational complexity: diversity and breadth of experience, in-
cluding experts who have a great deal of contact with experts in other fields; links
to users; and outsiders, openness to the environment; and integration across fields
via intersecting territories, multiple communication links, and smaller interdisci-
106 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

plinary business units. Conversely, isolation, or what can be termed “segmental-

ism” (Kanter, 1983), inhibits this critical first phase of innovation.
It is important to explain an apparent contradiction in the literature here.
Some analysts appear to argue that innovation does indeed require isolation, a
special organization separated from the rest and dedicated to innovation. For ex-
ample, Galbraith (1982)argued that innovation requires an organization specifi-
cally designed for that purpose, with a structure, processes, rewards, and people
combined in a special way; he also made clear that his focus was on “good ideas
that d o not quite fit into the organization’s current mold.” But note that the
“good idea” already exists, and the special organization is designed to focus on
developing and elaborating it without distraction once it has been identified.
Thus, isolation of the innovator group appears appropriate later in the process,
when project ideas have been formulated.
To generate ideas in the first place, a great deal of diverse outreach is in-
volved. R&D units that remain isolated are less creative than those that maintain
close integration in the search of exploration stage. Recall the example of the
Raytheon new products department, a unit with an unusually strong track record
of creative outputs. It is indeed physically isolated from the rest of the organiza-
tion to allow it to work on projects undistracted. But to generate ideas and acti-
vate innovation in the first place, department members immerse themselves in the
world outside the lab, wandering around the organization, seeking problems to
work on from their dense network of ties in other units, attending professional
conferences in scientific fields other than their own, going to trade shows to view
the exhibits, etc.

COALITION BUILDING
Once a specific project idea has taken shape, it must be sold-a necessity
even when the innovator was initially been handed the area as an assignment. It
must be sold because the initial assignment, though bearing some legitimacy, may
contain no promises about the availability of resources or support required to do
something of greater magnitude than routine activities (Kanter, 1982; 1983).
Thus, the second task of the innovation process involves coalition building, ac-
quiring power by selling the project to potential allies.
Overwhelmingly, studies of innovation show the importance of backers and
supporters, sponsors and friends in high places, to the success of innovation
(Quinn, 1979; Maidique, 1980). Galbraith (1982) distinguished the roles of
“sponsor”-those who discover and fund the increasingly disruptive and expen-
sive development and testing efforts that shape an innovation-and “orchestra-
tor”-managers of the politics surrounding a new idea. Observing that sponsors
were usually middle managers and orchestrators were higher level executives, he
argued that these informal roles could be formalized, with sponsors given re-
sources earmarked for innovation, and orchestrators allocating time to protecting
innovations-in-progress.
When a Thotisand Flowers Bloom 107

While most studies emphasize single roles (the “champion,” the “sponsor”),
detailed accounts of the history of innovations reveals the importance of a whole
coalition, embryonic and informal or assembled and formal (Summers, 1986).
Van de Ven (1986), in a similar vein, focuses not on a single sponsor but on the
importance of transactions or “deals” in the innovation process, and he sees the
management of the innovation process as managing increasing bundles of transac-
tions over time, Indeed, he and his colleagues found, in a comparative study of
seven very different large scale innovations in different sectors, that “much more
than sponsorship” was involved; higher management, one or two levels removed
from the innovation was directly involved in making major decisions about the
project and often “ran interference” for it as well as securing necessary resources
(Schroeder et al., 1986). Furthermore, a comparison of over 115 innovations
found in the successful ones a set of allies, often peers from other areas as well
as more senior managers, behind successful innovations, ranging from the
“stakeholders” who would be affected if the project was implemented to the
“power sources” who contributed the tools to ensure that implementation (Kan-
ter, 1983).
Thus, it is more appropriate to conceptualize the second major innovation
task as coalition building, a broader notion that ties in more of the organization,
rather than as seeking sponsorship, a narrower concept. In general, the success of
an innovation is highly dependent on the amount and kind of power behind it. In
contrast, innovation failures are characterized by ambivalent support; inadequate
resources during the initial fragile stages of development; constant efforts to “sell”
and “justify”; and personalized infighting over resources (Delbecq & Mills,
1985).
Thus, the effectiveness of the political activity the innovation entrepreneur
engages in, coupled with structural conditions conducive to power acquisition
and coalition building, may largely account for whether an idea ever moves into
the later phase of innovation production. Social and political factors, such as the
quality of the coalition building, may account for as much or more than technical
factors, such as the quality of the idea, in determining the fate of innovation.
Research shows that there are some kinds of ideas that are inherently better
able to attract support. The most salable projects are likely to be trialable (can be
demonstrated on a pilot basis-see especially Delbecq & Mills, 1985); reversible
(allowing the organization to go back to pre-project status if they d o not work);
divisible (can be done in steps of phases); consistent with sunk costs (build on
prior resource commitments); concrete (tangible, discrete); familiar or compatible
(consistent with a successful past experience and compatible to existing practices);
congruent (fit the organization’s direction); and have publicity value (visibility po-
tential if they work) (Kimberly, 1981; Zaltman et al., 1973). When these features
are not present, as they are unlikely to be in more “radical” innovations, then pro-
jects are likely to move ahead if they are either marginaf (appear off-to-the-side-
lines so they can slip in unnoticed) or idiosyncratic (can be accepted by a few
people with power without requiring much additional support) (Zaltman et al.,
1973; Kanter, 1983).
108 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

The features of successful ideas have more to do with the likelihood of gath-
ering political support than with the likelihood of the idea to produce results. In
general, the relative economic advantage of a new idea, as perceived by members
of an organization, is only weakly related to its rate of adoption (Rogers & Shoe-
maker, 1971). Instead, “political” variables may play a larger role, especially the
acquisition of “power tools” to move the idea forward.

Power Tools

Organizational power tools consist of supplies of three “basic commodities”

that can be invested in action: information (data, technical knowledge, political
intelligence, expertise); resources (funds, materials, space, time); and support (en-
dorsement, backing, approval, legitimacy) (Kanter, 1983).
To use an economic strategy, it is as though there were three kinds of “mar-
kets” in which the people initiating innovation must compete: a “knowledge mar-
ket” or “marketplace of ideas” for information; an “economic market” for
resources; and a “political market” for support or legitimacy. Each of the “mar-
kets” is shaped in different ways by conditions in the environment (e.g., critical
contingencies, resource scarcity; Pennings & Goodman, 1977; Pfeffer & Salancik,
1977), and by organizational structure and rules (e.g., how openly information is
exchanged, how freely executives render support). And each gives the person a
different kind of “capital” to invest in a “new venture” (also see Pfeffer & Salan-
cik, 1977).
We can hardly speak of “market” at all, of course, where the formal hierar-
chy fully defines the allocation of all three commodities, for example, when
money and staff time are availably only through a predetermined budget and
specified assignments, when information flows only through identified communi-
cation channels, and when legitimacy is availably only through the formal author-
ity vested in specific areas with no support available for stepping beyond official
mandates. In organizations where there is really no market for exchanging or re-
arranging resources and data, for acquiring support to do something outside the
formal structure, because it is tightly controlled either by the hierarchy or by a few
people with “monopoly” power, then little innovative behavior is likely. Indeed,
when people feel “powerless” through structural locations that limit them access
to the tools, they become more controlling and conservative (Kanter, 1977; Kan-
ter & Stein, 1979).
While some portion of the power innovators need may be already attached
to their positions and available for investing in an innovation, the rest must be
sought through allies. Thus, the organization’s structure determines the amount
and availability of power via both the distribution of power tools and the ease
with which coalitions can be formed. Access to external and internal sources of
power increases an innovation entrepreneur’s chances of successfully creating an
innovation.
When a Thoitsand Flowers Bloom 109

Coalition Structure

Which parties are potential coalition members? Principally those on whom

the innovator may be dependent-where there is interdependency affecting the
fate of the idea. The concept of organizational interdependency has both a techno-
logical (Thompson, 1967) and a political (Pfeffer, 1981) component.
First, people often form interdependent relationships because of mutual task
dependence. For example, a manager in a finance department may require finan-
cial information on operations costs from a production manager, who in turn re-
ceives back the financial information in some evaluated or analyzed form and uses
it to assess production efficiency. The timeliness and quality of the information
provided by each manager affects the other’s work.
Second, interdependencies may be political in nature, since organizations are
tools for “multiple stakeholders” (Kanter, 1980); managers identify and seek out
others with complementary and sometimes competing interests for the purpose of
trading resources, demands, etc. (March, 1962; Cyert & March, 1963). Networks
of interdependent members also form where people are joined by a variety of links
through which goods, services, information, affect and influence flow (Tichy &
Fombrun, 1979; also see Kaplan & Mazique, 1983).
In short, there are many types of interdependent relationships: hierarchical
(Weber, 1978; Schilit & Locke, 1982); lateral (Thompson, 1967; Burns & Stalker,
1968); oblique (Kaplan & Maidique, 1983). In addition, people also work in the
midst of multiple constituencies that are defined by common political or organiza-
tional interests and include persons outside the formal boundaries of the organiza-
tion (Pennings & Goodman, 1977; Connolly, Conlon, & Deutsch, 1980).
Constituencies may form around task, issues, attempts to create change or block
change, or salient values.
The size of the coalition is affected by how many territories the innovation
crosses. The broader the ramifications of the issues involved in the proposed inno-
vation and the greater the attendant uncertainties, the larger the coalition of sup-
porters needs to be if the idea for innovation is to result in product action
(Thompson, 1967).
Mobilizing a few potential members into an active, visible coalition also
mirrors a classic dilemma in organization theory, that of finding the appropriate
mix of inducements to obtain the desired contributions and work behavior from
employees (Barnard, 1938). The inducements an innovator can offer to partici-
pate in a coalition include a variety of payments, such as financial incentives, re-
sources, information, policy promises, learning experience, personal development,
or emotional satisfaction (March, 1962; Riker, 1962; Gamson, 1968). The ex-
change of inducements for coalition participation can also extend across both ver-
tical and lateral levels of an organization (e.g., Dalton, 1959; Blau, 1963).
Mobilizing coalition members through exchange assumes that “commodi-
ties” are available for trade, and that the organizer had some control over their
distribution (managers we interviewed often referred to this process as one of
110 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

“horse trading”). Such commodities used to mobilize coalition members can also
serve as the basis of organizational power; e.g., resources, slack, information, and
political support (Mechanic, 1962; Kanter, 1977).
Access to these commodities depends to a large degree on their distribution
within the firm; their munificence increases the ability to draw people into coali-
tion that can work on an innovation.
Because corporate entrepreneurs often have to pull in what they need for
their innovation from other departments or areas, from peers over whom they
have no authority and who have the choice about whether or not to ante up their
knowledge, support, or resources, to invest in and help the innovator, their work
is facilitated by integrative devises that aid network formation and collaboration
across areas; open communication; frequent mobility, including lateral career
moves; extensive use of formal team mechanisms; and complex ties permitting
crosscutting access.

Communication Density

Innovation flourishes where “communication integration” is high (Rogers

& Shoemaker, 1971). Open communication patterns make it easier to identify
and contact potential coalition members and to tap their expertise.
Examples of “open communication” systems from innovating companies
stress access across segments. “Open door” policies mean that all levels can, theo-
retically, have access to anyone to ask questions, even to criticize. At several high
innovation companies examined in one study, there were policies barring closed
meetings. In others, the emphasis was on immediate face-to-face verbal (not writ-
ten) communications (Kanter, 1983), unlike “mechanistic,” low innovation or-
ganizations where written communications prevail (Burns & Stalker, 1968).Such
open communication norms acknowledge the extent of interdependence-that
people in all areas need information from each other.
“Openness” a t such organizations is reflected in physical arrangements as
well. There may be a few “private” offices, and those that do exist are not very
private. One manager had a “real” office enclosed by chest-high panels with
opaque glass, but people dropped by casually, hung over the walls, talked about
anything, and looked over his desk when he was not there. In general, people
walk around freely and talk to each other; meetings and other work are easily in-
terrupted, and it is hard to define “private” space. They often go to the library or
conference room to “hide” to get things done, especially on “sensitive” matters
like budgets (Kanter, 1983).
Open communication serves a very important function for the potential in-
novator. Information and ideas flow freely and were accessible; technical data and
alternative points of view can be gathered with greater case than in companies
without these norms and systems. And thus both the “creative” and the “politi-
cal” sides of innovation are facilitated.
When a Thousand Flowers Bloom 111

Network Density

Coalition formation in the interest of innovation is also aided by conditions

that facilitate dense ties through networks. Circulation of people is a first net-
work-facilitating condition. Mobility across jobs means that people rather than
formal mechanisms are the principle carriers of information, the principal integra-
tive links between parts of the system. Communication networks are facilitated
(see Thurman, 1979-1980), and people can draw on first-hand knowledge of
each other in seeking support. Knowledge about the operations of neighboring
functions is often conveyed through the movements of people into and out of the
jobs in those functions. As a set of managers or professionals disperse, they take
with them to different parts of the organization their “intelligence,” as well as the
potential for the members to draw on each other for support in a variety of new
roles. In just a few moves, a group that has worked together is spread around, and
each member now has a close colleague in any part of the organization to call on
for information or backing.
A second network-forming device is more explicit: the frequent use of inte-
grative team mechanisms at middle and upper levels. These both encourage the
immediate exchange of support and information and create contacts to be drawn
on in the future. The organizational chart with its hierarchy of reporting relation-
ships and accountabilities reflects only one reality in innovating organizations; the
“other structure,” not generally shown on the charts, is an overlay of flexible, ad
hoc problem-solving teams, task forces, joint planning groups, and information-
spreading councils.
It is common a t innovating, entrepreneurial companies to make the assign-
ments with the most critical change implications to teams across areas rather than
to individuals or segmented units for example, at one company a team of mixed
functional managers created a five-year production and marketing plan for a new
product. This was a model of the method that top management endorsed for car-
rying o u t major tasks and projects. At a computer company, the establishment of
formal interdepartmental or cross-functional committees was a common way
managers sought to improve the performance of their own unit (Kanter, 1983).
The legitimacy of crosscutting access promotes the circulation of all three of
the power tools: resources, information, and support. This allows innovators to
go across formal lines and levels in the organization to find what they needed-
vertically, horizontally, or diagonally-without feeling that they are violating pro-
tocol. They can skip a level or two without penalty. This is essential if there is to
be hands-on involvement of managers up several levels, as Schroeder et al. (1986)
found characteristic of large-scale innovations.
Matrix designs, though not essential for crosscutting access, can be helpful
in legitimizing it, for the organization chart shows a number of links from each
position to others. There is no “one boss” to be angered if a subordinate manager
goes over his head or around to another area; it is taken for granted that people
move across the organization in many directions; and there are alternative sources
112 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

of power. Similarly, formal cross-area and cross-hierarchy teams may provide the
occasion and the legitimacy for reaching across the organization chart for direct
access (Kanter, 1983).

IDEA REALIZATION AND INNOVATION PRODUCTION

The third task of the innovation process involves assembling a working
team to “complete” the idea by turning it into a concrete and tangible object
(physical or intellectual) that can be transferred to others. The idea becomes a re-
ality; a prototype or model of the innovation is produced that can be touched or
experienced, that can now be diffused, mass-produced, turned to productive use,
or institutionalized.
There are a number of critical organizational issues related to the ability to
move a innovation through this phase. These issues join with social psychological
(intragroup) variables to account for the performance of the group responsible for
producing the innovation model.

Physical Separation

While structural isolation is a liability for idea generation or innovation ac-

tivation, it is an asset for idea completion or innovative production.
Differentiated innovation units, separated from ongoing operations in both
a physical and an organizational sense, are not necessary to stimulate or activate
innovation (a task for which isolation is counter-productive), but they d o appear
helpful for ensuring that the working out of the innovation, the production of the
initial model, actually occurs. Lockheed’s term, “skunkworks,” (taken from a
Peanuts cartoon) has been used to refer to the special setting where innovation
teams can create new things without distractions.
Galbraith (1982) has argued for the importance of “reservations”-organ-
izational units, such as R&D groups, totally devoted to creating new ideas for fu-
ture businesses-havens for “safe learning” managed by a full-time sponsor.
Reservations can be internal or external, permanent or temporary. Galbraith
found that some innovations, including the new electronics product he studied,
were perfected at a remote site before being discovered by management; thus “the
odds [for innovation] are better if early efforts to perfect and test new ‘crazy’ ideas
are differentiated-that is, separated-from the function of the operating organi-
zation” (Galbraith, 1982).
High innovation companies in the United States, Europe, and Japan have
flatter organizations, smaller operating divisions, and smaller project teams
(Quinn, 1985). Small teams of engineers, technicians, designers, and model mak-
ers are placed together in “skunkworks,” with no intervening organizational or
physical barriers to developing the idea to prototype stage. Even in Japanese or-
ganizations supposedly known for elaborate (and slow) consensus-building proc-
When a Thousand Flowers Bloom 113

esses, innovation projects are given autonomy, and top managers often work di-
rectly on projects with young engineers, including the founder of Honda himself.
This approach eliminates bureaucracy, allows fast and unfettered communication,
enables rapid turnaround time for experiments, and instills a high level of group
loyalty and identity by maximizing communication and commitment among team
members.

Boundary Management

If small, separate units aid idea model production, then boundary manage-
ment is a particular problem. The team must continue to procure information and
resources and return output to the rest the organization (Gladstein & Caldwell,
1984), but without becoming so outwardly focused that ability to d o the job is
jeopardized.
Success in building the innovation may be a function of how well external
relations are handled as much as the technical feasibility of the idea. On the one
hand, those who are prone to interfere must be kept from distracting the focus of
the working team; on the other hand, the stakeholders, coalition members, and
others whose support will be required at the transfer phase must be communi-
cated with and involved, to ensure their support. The group must both buffer it-
self against too much input from its environment (Thompson, 1967) as well as
manage the demand for what it is producing so that it has an appropriate level-of
exchange with the world around it-not too much, and not too little.
While many analysts have argued that “gatekeeping” is an important func-
tion in the management of innovation. Gladstein and Caldwell (1984) have gone
further by identifying four boundary management roles in the new product teams
they studied, roles that can all be played by one person or distributed throughout
the group:

Scouts, bringing in information or resources needed by the groups;

Ambassadors, carrying out items that the group wants to transmit to oth-
ers;
Sentries, controlling the transactions that occur at the boundaries, decid-
ing how much can come in;
Guards, controlling how much goes out of the group.

Whereas scouts and ambassadors keep extra group relationships smooth

and get the group its needed supplies, sentries and guards buffer the group from
outside interference. But note that all of these roles may be played by one person
or just a few people, allowing the rest of the group to work on tasks without pay-
ing any attention to the world outside the project team. In the much publicized
case of the building of a new computer at Data General, the project manager and
his two aides handled all of the boundary tasks, allowing the team members to fo-
cus on completing the project in what proved to be record time (Kidder, 1981).
114 KNOWLEDGE h4ANAGEh4ENT AND ORGANIZATIONAL DESIGN

Boundary management is important not merely to get the working group

what it needs and save it from unnecessary interference but also to handle any
subtle threats to the continued existence of the innovation project. In one study, it
was striking how little overt opposition is encountered by entrepreneurial manag-
ers-perhaps because their success at coalition building determines whether a pro-
ject starts at all. Opposition or resistance seemed to take a more passive form
instead: criticism of specific details of the plan, foot-dragging, low response to re-
quests, unavailability, or arguments for preferential allocation of scarce time and
resources to other pet projects. Early opposition was likely to take the form of
skepticism and therefore reluctance to commit time or resources. Later opposition
was likely to take the form of direct challenge to specific details of the plan that is
unfolding (Kanter, 1983).
The nature of the opposition becomes clearer at the idea production points
in the innovation process for several reasons. First, the very act of contacting oth-
ers in the course of realizing an idea may mobilize what would otherwise have
been latent or unorganized opposition. Most people will not spend their fund of
political capital by overtly opposing a new idea right away, especially if it has the
support of someone who is powerful, because it may never “get off the ground.”
Political capital would have been depleted unnecessarily. It is when it looks as
though the project might actually happen that the critics begin to surface, gener-
ally arguing at the project’s most vulnerable point that it has had enough time to
prove itself; time to move on to something else (usually the critic’ own pet project)
(Kanter, 1983).
At the same time, many new ventures or innovation projects tend to be rela-
tively invisible in the beginning, occurring in hidden corners of the organization or
not significant enough to warrant their rivals’ or competitors’ attention. But as the
effort gets closer and closer to results, it becomes more of a threat, and rivals be-
gin to take action to crush it. At Apple, for example, a start up company by an
employee was passively tolerated by chairman Jobs, but when it looked like that
group might actually have a rival technology, he threatened suit, saying that it had
been developed on Apple time and Apple owned it (Moritz, 1984).
My research identified a number of tactics that innovators used to disarm
opponents: waiting it out (when the entrepreneur has no tools with which to di-
rectly counter the opposition); wearing them down (continuing to repeat the same
arguments and not giving ground); appealing to larger principles (tying the inno-
vation to an unassailable value or person); inviting them in (finding a way that
opponents could share the “spoils” of the innovation); sending emissaries to
smooth the way and plead the case (picking diplomats on the project team to pe-
riodically visit critics and present them with information); displaying support
(asking sponsors for a visible demonstration of backing); reducing the stakes
(deescalating the number of losses or changes implied by the innovation); and
warning the critics (letting them know they would be challenged a t an important
meeting-with top management, for example). Note that many of these are more
likely to succeed when the innovation group has a strong coalition backing it. The
effectiveness of interpersonal processes depends on structural conditions.
When a Thoirsand Flowers Bloom 115

Because of the controversy that surrounds many innovations, it is important

for the working team to continue to send information outward. For example,
when the project nears completion and there are things to see, they may begin to
bring important people in to view the activities. Successful innovators have been
observed to “manage the press,” working to create favorable and up-to-date im-
pressions in the minds of peers and key supporters (Kanter, 1983). Similarly,
Friedlander and Scott (1981) found that activities of change teams were given
more legitimation and were more likely to be implemented when there was a great
deal of communication with top management, including two-way dialogue about
particular project ideas.

Continuity

Structural and social conditions within the innovation team also make a dif-
ference in success. Because “interactive learning” (Quinn, 1985) is so critical to in-
novation, innovation projects are particularly vulnerable to turnover. Continuity
of personnel, up to some limits (Katz, 1982), is an innovation-supporting condi-
tion.
There are sometimes good reasons, from the project’s standpoint, for people
to leave: inadequate performance, interpersonal tensions, the wrong skills. But
every loss-and-replacement can jeopardize the success of the innovation process,
in three different ways:

1. Each person leaving removes knowledge from the pool, that has not yet
been routinized or systematized. In a sense, everyone leaving an innova-
tion project does indeed take “secrets” with them-private knowledge
they may have gained that has not yet been shared with the rest of the
team because of the intensity with which everyone is gathering knowl-
edge.
2. Each person entering deflects the energies and attention of the others
from knowledge development to education-to try to duplicate the expe-
rience base of current staff and avoid reinventing the wheel. But telling
about it is not only time consuming; it is indeed no substitute for having
been there.
3. Each person entering in a key position may wish to change course in or-
der to exercise his or her own power, thereby failing to take advantage of
accumulated knowledge. So every new boss is indeed a new beginning.

Turnover in key positions outside the project team can also create problems,
though not necessarily as severe: The division is reorganized, for example, and the
new management does not “understand” the venture. The coalition is disrupted
and needs to be rebuilt. An organization can easily undermine an innovation with-
out “officially” stopping it simply by reorganizing and changing its reporting rela-
tionships.
116 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

In one case, the problems of turnover are illustrated. A senior executive of a

major instruments manufacturing company was recalling one of the company’s
venture failures-a new product start-up in one of the divisions. He knew this
project well, because he had been the venture manager for the first 6 months. “I
think about this often,” he said, “because if I had stayed I think I could have made
it work.” Six months into the project, he was offered a promotion up several lev-
els, from managing 15 people in the start-up to managing 6,000 in an established
division. The career implications were clear: take it now or lose his place in line.
The rewards were also clear: “The corporation was set up to reward the person
running a stable $200 million business more than someone growing a business
from zero to $10 million to $200 million, which is much, much harder.” Even so,
he remembered, “I wanted a week to think about it; I felt torn.” Eventually he
took the promotion. The start-up team understood the corporate career message,
but they still felt abandoned. And the new manager sent in to replace him simply
did not have “the feel” for what it would take to get his business going. Even
more than loss of leadership, it was loss of experience that hurt this project.
Ironically, creating change requires stability-continuity of people especially
during the information-rich, knowledge-intensive development stage. But estab-
lished corporations often exacerbate the vulnerabilities of their new ventures and
innovation efforts by the instability they encourage in and around them. Lock-
step career systems that tie rewards to promotions, thus requiring job changes in
order to “advance,” or that put more value on the “safer” jobs in already-estab-
lished businesses, encourage people to abandon development efforts before their
knowledge has been “captured.” Thus, organizational structures and cultures that
allow continuity on innovation teams by facilitating unusual or “off-line” career
paths, allocating human resources on a project basis rather than a time basis, and
rewarding completion are helpful ingredients for successful innovation produc-
tion.
Continuity is also supported where strong commitment is generated, so that
people want to stay and want to contribute. Three kinds of commitment mecha-
nisms are relevant to innovation efforts:

0 Conditions encouraging a rational calculation of the benefits of continu-

ing participation;
0 Those encouraging strong social and emotional ties with the group;
0 And those encouraging a strong belief in the fundamental values or pur-
poses of the efforts (Kanter, 1972).

Structural and social facilitators of commitment to innovation teams would thus

include these kinds of things, among others: A sense of “investment” might
be produced by a financial stake in outcomes which grows with time spent, as
AT&T’s new venture teams have. A sense of “communion” might come from
clear group identity and sense of specialness through team names, rituals, and
celebrations like those in Data General’s new computer development group (Kid-
der, 1981). A sense of strong values might come from reminders of the connection
When a Thoirsand Flowers llloorn 117

to user needs. A t the same time, where there is also physical isolation of the team
and very long working hours, energies have to focus inward, and the lure of com-
peting ties is diminished. (See the discussion of “renunciation” in Kanter, 1972.)
It is important to note, however, that if too much time goes by before inno-
vation completion, then team loyalty and stability can become a liability instead
of an asset. Katz (1982) found that the “ideal” longevity of R&D teams is be-
tween 2 and 5 years. It takes 2 years to begin to work well together, but after 5
years the group becomes stale.

Flexibility
Flexibility is another requirement for idea realization. It is quite common for
innovations to fail to proceed as planned but instead to encounter unexpected
roadblocks or obstacles that require replanning and redirection if the innovation
is ever to be produced. Cost overruns and missed deadlines are common, due to
the inherent high uncertainty of the development process. For example, in one
pharmaceutical company the ratio of actual to expected cost of new products was
2.11; the ratio of actual to expected time was 2.95 (Mansfield et al., 1981).
Numerous cases in numerous fields illustrate the unpredictable nature of in-
novation, and therefore the need for flexibility in order to persist with a project.
For example:

GTE’s Telemessenger would not show returns fast enough because, like
most innovation, the product employed technology so unknown in the
marketplace that prospective customers were not receptive to it, and sev-
eral rounds of replanning were necessary to get the right configuration.
Even assumptions about the scope of the test market had to be changed in
the light of experience. What the team had originally imagined was a
local test had to be rethought when the product was reconceived (success-
fully) as an aid to communication across the time zones, thus necessitat-
ing a national test. This change in tactics paid off. Though only 6 units
were sold after a local mailing of 60,000 letters, 200 were sold a t one
crack to a multi-national company immediately after the test went na-
tional (Powell, 1985).
The historic town of Alexandria, Virginia, now has an important factory
redevelopment project on its waterfront, a project that seemed simple and
straightforward when it was first voted on 10 years earlier but required
several changes of direction midstream. But the city-owned Alexandria
Torpedo Factory and Art Center almost didn’t happen. Among a number
of unexpected obstacles that nearly killed the project and required addi-
tional entrepreneurial effort to resolve was the fact that it threatened a
small building used by a public school rowing program. Without the
flexibility to make changes in order to persist with the project, the city
118 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

would never have seen the results: a rise in the value of its property from
$4 million to $31 million.

Thus, as Quinn (1985) found across three countries, multiple approaches,

flexibility, and quickness are required for innovation because of the advance of
new ideas through random and often highly intuitive insights and because of the
discovery of unanticipated problems. Project teams need to work unencumbered
by formal plans, committees, board approvals, and other “bureaucratic delays”
that might act as constraint against the change of direction.
Furthermore, innovations often engender secondary innovations, a number
of other changes made in order to support the central change (Kanter, 1983). As
necessary, new arrangements might be introduced in conjunction with the core
tasks. Methods and structure might be reviewed and when it seems that a project
is bogging down because everything possible has been done and no more results
are on the horizon, then a change of structure or approach, or a subsidiary project
to remove roadblocks, can result in a redoubling of efforts and a renewed attack
on the problem. This is why Van de Ven (1986),among others, argued for the lack
of utility of distinctions between technical and organizational innovations; in
practice, one often entails the other. Indeed, restructuring of the organization
often occurs during the innovation process, including joint ventures, changes in
organizational responsibilities, use of new teams, and altered control systems
(Schroeder et al., 1986).
Flexibility is an organizational rather than a purely individual variable.
Those organizations that permit replanning, give the working team sufficient op-
erating autonomy, and measure success or allocate rewards for results rather than
adherence to plan are likely to have higher rates of innovation production. Be-
cause of the inherent uncertainty of innovation, advance forecasts about time or
resource requirements are likely to be inaccurate; it is difficult to budget or to
forecast when lacking an experience base by definition, in the case of a new idea.
The GTE Telemessenger was almost aborted when the project manager’s first
market test failed, because he had not brought in the results he promised, and he
went through several rounds of argument to get an original “15 days to fix it” ex-
tended to 2 months (Powell, 1985). Requiring commitment to a predetermined
course of action interferes with the flexibility needed for innovation.

Balancing Autonomy and Accountability

Some analysts argue that innovation production occurs better when the
working team is left completely alone, freed from all bureaucratic procedural de-
mands and allowed total concentration, total focus on its work. But there is a
middle ground between the extreme of so many reporting requirements that the
team spends more of its time preparing reports than doing the work, and the other
extreme of no controls or measures until the end.
When a Thousand Flowers Bloom 119

If some innovation projects fail because they are overly constrained by the
need to follow bureaucratic rules and seek constant approvals, others may equally
fail because they are overfunded and undermanaged by top leaders, which can re-
move the incentive to produce results efficiently, Indeed, Bailyn (1985) learned
from her studies of R&D labs that many engineers were subject to overly con-
straining operational controls while permitted too much “strategic autonomy” to
set their own research goals-just the opposite of the combination needed for suc-
cess.
This can be a particular problem in large new ventures. In one case in a lead-
ing corporation, top management generously funded a new project development
effort and then left it alone, assuming that they had done the right thing by pro-
viding abundant resources. Because they were so rich, the team wasted money on
dead-ends and intriguing but unnecessary flourishes and failed to replan when
early results were disappointing. The team did not need to justify their actions to
anyone, and the project eventually failed. This is one reason why Stevenson and
Gumpert ( 1985) argued that successful entrepreneurship involves mtrfti-stage
commitments-smaller amounts of money at more frequent intervals.
The ideal structural context surrounding an innovation project, then, should
offer procedural autonomy coupled with multiple milestones that must be reached
in order for the project to continue. These milestone points represent the major in-
terface with organizational decision makers and perhaps coalition members. They
also help maintain team members’ own commitment by giving them targets to
shoot for and occasions to celebrate.

TRANSFER AND DIFFUSION

The culmination of innovation production is transfer to those who will ex-
ploit the innovation or embed it in ongoing organizational practice. Transfer
needs to be handled effectively, if new products are to be successfully commercial-
ized or new organizational practices or techniques to be successfully diffused. Iso-
lated in its development, the innovation must again be connected with the actors
and activities that will allow it to be actually used.
Social arrangements, from organization structures to patterns of practice,
again make the principal difference, even more than the technical virtues of the in-
novation (Rogers & Shoemaker, 1971).

Strategic Alignment and Structural Linkages

Whereas creation and development-production of the innovation model-

can occur with few resources, little visibility, modest coalitions, and the isolated
activity of relatively small teams, use of the innovation is a different matter. If
creation is an intensive process; diffusion is an extensive process. Use requires
120 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

many other people, activities, patterns and structures to change to incorporate the
innovation.
Thus, a first condition for effective transfer is minimal new change require-
ments because the innovation is aligned with strategy or direction and linked to
the other parts of the structure, so that adjustments and changes have already
been made in anticipation of the innovation.
It is not surprising that innovations are more successfully transferred, com-
mercialized, or diffused where the organization or market is already receptive to
the idea and prepared for its use. This is almost tautological. Where there is
stronger organizational commitment in the development process, signified by
funding, visibility, coalition support, and so forth, there are more “side bets”
placed on the idea (that is, staking of reputations in the outcome) as well as
greater “sunk costs.” Thus, there will be more pressures to use the innovation in
more ways and make it more central to the organization’s strategy. Organizational
arrangements will already have begun to bend in anticipation of the successful de-
velopment, often through the negotiations among departments, the “logical incre-
mentalism” through which new strategies are adopted (Quinn, 1980).
On the other hand, those innovations that begin life as random deviance, or
unofficial bootlegging in a hidden corner of the organization, or the idiosyncratic
dream of a tolerated-but-marginal actor, have a harder time getting adopted re-
gardless of their virtues. Other actors, other departments have already made their
plans without taking the possible availability of an innovation into account.
Therefore, structures and practices have already been established that would have
to be rearranged. These structural constraints to diffusion or transfer may be
matched by political constraints: controversy over the innovation or refusal to use
it by those uninvolved in its development. The latter is the common NIH (not in-
vented here) problem: this problem particularly plagues organizational innova-
tions (Kanter, 1983; Walton, 1975).
It has long been a cliche in the innovation literature (primarily because most
scholars cite the same handful of studies) that diffusion or adoption of an innova-
tion, once developed, is aided by formalization and centralization in the organiza-
tion, by a concentration of power and a set of employees accustomed to following
orders, The opposite structural features, then, from those that are conducive to a
free flow of many new ideas are held to be necessary for ensuring the rapid accep-
tance of any one.
Recent evidence, however, makes this a much more contingent proposition
(Kimberly, 1980). Cohn and Turyn (1984), in a quantitative comparison of inno-
vations in the domestic footwear industry, found that formalization and centrali-
zation were associated with adoption of evolutionary innovations but not with
revolutionary ones.
A concentrated source of power is needed to impose the innovation on the
organization or move it quickly through preexisting formal channels whenever
the innovation has not already been appropriately linked to the units to which it
will be transferred. Indeed, strong central authority can be argued to be just a
functional alternative to strong direct links between an innovation project and
those to whom its product is handed-off.
When a Thousand Flowers Bloom 121

If an innovation development project is structurally well-integrated as it

comes to completion, rather than segmented and isolated from the rest of the or-
ganization (Kanter, 1983), then it does not require the power of centralized
authority to ensure its effective transfer. Other units have readied themselves to re-
ceive the innovation. Indeed, the hand-off or diffusion process is more difficult in
organizations where interdepartmental rivalries and lack of integration cause fric-
tion when anything comes from a sister unit; then only “orders” from central
authority are attended to. Perhaps this is why evidence indicates that successful
new ventures in large corporations are more likely to be the ones sponsored by op-
erating line executives rather than by corporate executives (Hobson & Morrison,
1983); the line-sponsored ventures are already closely connected with implemen-
tors.
Of course, effective transfer also requires a strategic decision that this in-
novation should get resources allocated to it, resources necessary to exploit its
potential. For product and technical process innovations, and even for some or-
ganizational innovations, the greatest financial requirements begin after the model
has been developed. Thus, the nature of the strategic decision process and how
top management is linked to the innovation project is another critical structural
element in an innovation’s success or failure (Burgelman, 1984).
At the transfer point, when resources to exploit the innovation are allocated,
visible and well-connected projects already aligned with the organization’s strate-
gic objectives are likely to fare better. In turn, the degree of investment the project
gets, as it is moved into commercialization, routine production, or institutionali-
zation affects its prospects for success as an ongoing product or practice. “Think-
ing small” and not providing adequate investment is often identified as a reason
for new venture failures (Drucker, 1985). Research on the first 4 years of opera-
tion of 117 corporate ventures in established markets in manufacturing found
that the businesses above the median in success began with capacity that could
meet twice the current total market demands, whereas those below the median be-
gan with a capacity that could meet only 6% of the current total market demands.
Furthermore, the “winning” ventures initially set higher market share objectives,
had R&D spending levels twice those of the other ventures in the first 2 years and
marketing expenditures about 1.5 those of the other ventures in the same period
(Hobson & Morrison, 1983).

Interface Structures: Active Agents and Communication

Channels

The transfer or diffusion issue should be conceptualized as a continuum. At

one extreme there is perfect identity between the developers and the ultimate us-
ers, so that the innovators are essentially producing the innovation for themselves,
to their own specifications, with foreknowledge that they will be using whatever it
is that they make. Organizations can come close to replicating this condition in
customized development work for specific clients already internally committed to
use, in which client representatives actually sit on the development team. In this
122 KNOWLEDGE MANAGEiMEN’T AND ORGANIZATIONAL DESIGN

case, transfer or diffusion is nonproblematic; it is a n inevitable part of successful

development.
At the other extreme, there is little or no connection between developers and
those to whom the innovation could potentially be transferred, nor is there an es-
tablished transfer process. There is high uncertainty (an information issue) and
controversy (a political issue) about what the next step is to get anyone to use the
innovation, who should take it, and whether there are identifiable customers for
the idea, whether anyone does or should want the innovation.
A variety of interface or bridging structures can reduce both the uncertainty
and the controversy, thus making it more likely that successful transfer will occur,
One method for diffusing new ideas is to establish a group whose formal re-
sponsibility is to move new ideas into active use (Engel, Kollate, & Blackwell,
1981). Members serve as active agents of diffusion, managing the process by
which the realized idea is transferred to those who can use it. Part of their man-
date is to gather the information to make systematic the process of getting the in-
novation to users.
Inside organizations, such bridging structures might take the form of prod-
uct managers, whose job is to manage the successful entry of a new product into
the marketplace, drawing on every function in the organization that might con-
tribute, from continuing work on the design to the manufacturing process to the
sales effort. Or, in the case of organizational or work innovations, the bridging
structure might be a transition team or “parallel organization” (Stein & Kanter,
1980) that concentrates on the change process as a management task in and of it-
self.
Agents of diffusion may also exist outside the organization. Indeed, it can be
argued that external agents are even more important in diffusion than champions
inside the organization, for they add real or imagined legitimacy to the idea, why
Rogers and Shoemaker (1971) found contact with consultants such an important
part of the diffusion of innovation. What is important is not only the cloak of re-
spectability in which the external party clothes the innovation, but also the com-
munication service provided. Thus, Walton (1987) found that the diffusion of
work innovations in shipping in eight countries was aided by formal organiza-
tions set up to study and write about those innovations. They served as a neces-
sary communication channel to transfer innovations to other users.
H o w well organized the environment is for the transfer of ideas can account
for how rapidly a particular innovation is diffused. By “organized” I mean the
case with which those with common interests can find each other, and therefore
how easily connections can be made between innovations and users. Thus, the ex-
istence of conferences, meetings, and special interest associations should all be
valuable in diffusing innovations, even product innovations, which have to be
brought to the attention of specific groups. Again, this can occur within as well as
outside a particular organization. 3 M and Honeywell both organize a large num-
ber of internal conferences and “idea fairs” to connect ideas with those who can
use it or help take it the next step.
Trade associations, professionals and societies, and specialist consulting or-
ganizations are among those serving this purpose more broadly. The Food Mar-
When u Thorisaitd Flowers 1 3 1 0 0 ~ i 123

keting Institute, trade association for grocers and supermarkets, w a s largely re-
sponsible for facilitating the spread of universal price codes on packages from
manufacturers a n d hence the spread of scanners in stores.

The Institutional Environment

The last issue in transfer and diffusion is a receptive and social and legal en-
vironment. The institutional environment, I propose, is so often taken for granted
in the study of innovation that it tends to be visible mostly when it impedes. But
the institutional environment is one of the most important factors distinguishing
between eight nations in their overall record of diffusion of work innovations in
shipping (Walton, 1987). Among the specific elements making a difference are
patterns of labor organization and government policy and regulations. In the
United States, for example, where innovation diffusion has been low, a series of
fragmented labor unions bargaining independently with shipowners, with n o ve-
hicles for industry-wide collaboration by either party, accounted in part for the
low diffusion rate.
The role of government in influencing innovation transfer can be a strong
one. Hollomon and colleagues (1981) identified specific ways in which govern-
ment policies a n d programs directly affect innovation adoption patterns:

Assessment of new and existing specific technologies

Direct regulation of the research o r development of new products and
processes
Direct regulation of the production, marketing, and use of new or exist-
ing products
Programs to encourage the development and utilization of technology i n
and for the private goods and services sector
Government support of technology for public services for consumers
Policies to affect industry structure that may affect the development and
use of innovation
Policies affecting supply and demand of human resources having an im-
pact o n technological change
Economic policies with unintended or indirect effect on technological in-
novation
Policies affecting international trade and investment
Policies intended to create shifts in consumer demand
Policies responding to worker demand having impact on technological
change.

Whether innovations are ultimately spread and used, then, may be a matter
of societal as well as industry organization. This level of analysis is not common
in the innovation literature, but it demands more attention, particularly with re-
spect to innovations that themselves have organizational consequences. Unfortu-
124 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

nately, much of the literature is shortsighted in still looking for determinants of

adoption of innovations in individual attitudes or intraorganizational structures.
But as organizations themselves bump up against the institutional limits to
innovation diffusion, then the issues become clearer. For example, if the use of
technological innovations has implications for job security, then the institutional
patterns of labor relations in the industry may be among the most important de-
terminants of an organization’s ability to use such innovations. Several major
companies are now attempting to reshape the broader institutional context in or-
der to create conditions for more rapid diffusion of innovation within their bor-
ders. General Motors, is a notable example, planning the new Saturn subsidiary
jointly with the United Auto Workers Union, using a series of joint committees.
Pacific Telesis is also reshaping relations with its principal unions through local
common interest forums of company officers and union presidents that define
many workplace policies together. But even if the unions concur, the current labor
law framework may be a significant impediment; Pacific Telesis has already faced
one legal challenge to institutional restructuring.
Innovation, and the spread of innovation, is also a function of industry con-
ditions and the support an organization can draw from its larger community, as
research by Ruttan and Hayami (1984) and Trist (1981) shows. The more de-
pendent an organization is on others (Pfeffer & Salancik, 1977), the more likely
that it will be shaped or constrained in its internal innovation by those portions of
the environment which dominate it. But the opposite also holds. Some environ-
ments represent “fertile fields” that provide more of the surrounding conditions
conducive to innovation.
“Fertile fields’’ include these kinds of features, associated with entrepreneur-
ship in the form of start-ups as well as innovation in established organizations:

Close proximity and ample communication between innovators and us-

ers
A more highly skilled, professionalized, cosmopolitan workforce
A flow of new technical ideas from R&D centers
A more complex, heterogeneous environment that encourages innovation
as an uncertainty-reduction strategy (Kimberly, 1981)
Channels of communication for exchange of innovation ideas
Competition from entrepreneurial new companies, in turn benefiting
from the availability of venture capital
More interorganization interdependence and integration (Pierce and Del-
becq, 1977)
Public encouragement of new ideas as social goods.

This brings us full circle, for many of these same conditions help activate the
innovation process as well as diffuse the models later.
The ultimate set of social structural factors supporting innovation, then,
comes from the nature of the environment in which an organization operates as
When a Thoiisand Flowers Bloom 125

well as its connections to various key units in that environment. Although an in-
novation model may be produced in one organization independently and in isola-
tion, it takes the actions of many for the innovation to diffuse.
It is appropriate to look beyond the borders of one organization for the de-
terminants of innovation. Indeed, some innovations can start life as the joint
product of more than one organization, through joint ventures, cooperative re-
search efforts, and strategic alliances. The reputed Japanese “edge” in technology
diffusion is said to come precisely from an institutional context allowing and en-
couraging such interorganizational cooperation in the same industry-a strategy
still largely limited by U.S. antitrust laws. Furthermore, sometimes organizations
unwittingly cooperate in innovation. For example, the failure of innovation in one
organization can be the trigger for the creation of a new organization designed
solely to develop that same innovation, the entrepreneurial process that has led to
spinoffs from larger companies that reject innovations developed and exploited
successfully by start-up companies. And the contribution of some organizations to
innovation is to generate new organizations (e.g., Wiewel and Hunter, 1985).

CONCLUSION
I have tried to connect the major tasks in the innovation process to those
structural arrangements and social patterns that facilitate each. Innovation con-
sists of a set of processes carried out at the micro-level, by individuals and groups
of individuals; and these micro-processes are in turn stimulated, facilitated, and
enhanced-or the opposite-by a set of macro-structural conditions. Overall, the
common organizational threads behind innovation are breadth of reach, flexibil-
ity of action, and above all, integration between those with pieces to contribute,
whether inside or outside a single organization.
Undeniably, innovation stems from individual talent and creativity. But
whether or not individual skills are activated, exercised, supported, and chan-
nelled into the production of a new model that can be used, is a function of the
organizational and interorganizational context. Throughout, I have marshalled
evidence to show the importance of integration to the innovation process, close
structural connections between potential innovators and iisers, between functions
and departments, between the innovation project and the units or organizations
that will move the model into production and use. I have also shown that the in-
tegrative organizational model helpful for innovation extends beyond the borders
of a single organization. Innovation benefits from interorganizational ties and or-
ganization-environment linkages as well as from internal integration.
Making a thousand flowers bloom is not a fully random or accidental proc-
ess, unless we are satisfied with spindly, fragile wildflowers. Instead, the flowers of
innovation can be cultivated and encouraged to multiply in the gardens of organi-
zations designed on the integrative model, organizations where the growth
rhythm of innovation is well understood.
126 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

ACKNOWLEDGMENT

T h e highly competent assistance of David V. S u m m e r s and Paul M y e r s is

gratefully acknowledged.
Funding w a s provided by t h e Division of Research of t h e H a r v a r d Business
School, w h o s e generosity is appreciated.

REFERENCES
Abernathy, W.J., & Utterback, J.M. (1978). Patterns of industrial innovation. Technology
Review, 80 (June/July) 41-47.
Allen, T.J. (1984).Managing the flow of technology: Technology transfer and the dissemi-
nation of technological information within the R 6 D organization. Cambridge, MA:
MIT Press.
Amabile, T.M. (1983).The social psychology of creativity. New York: Springer-Verlag.
Axelrod, R. (1970). Conflict of interest. Chicago: Marham.
Bacharach, S.B., & Lawler, E.J. (1960). Power and politics in organizations. San Francisco:
Jossey-Bass.
Bailyn, L. (1985). Autonomy in the industrial R&D lab. Human Resource Management,
24 129-146.
Baldridge, J.V., & Burnham, R.A. (1975).Organizational innovation: individual, organiza-
tional, and environmental impacts. Administrative Science Quarterly, 20 165-1 76.
Barnard, C.I. (1938).The functions of the executive. Cambridge, MA: Harvard University
Press.
Bendix, R. (1956). Work and authority in industry. Berkeley, CA: The University of Cali-
fornia Press.
Biggadike, R. (1979).The risky business of diversification. Harvard Business Review, 57
(May-June), 103-111.
Blau, P.M. (1963). The dynamics of bureaucracy (2nd ed.). Chicago: The University of
Chicago Press.
Block, Z. (1982). Can corporate venturing succeed? The Journal of Business Strategy, 3
(Fall), 21-33.
Block, Z. (1983).Some major issues in internal corporate venturing. In J.A. Hornaday, J.A.
Timmons, & Karl H. Vesper (Eds.). Frontiers of entrepreneurial research (pp. 382-389).
Wellesley, MA: Babson College Center for Entrepreneurial Studies.
Boissevain, J. (1971).Second thoughts on quasi-groups, categories, and coalitions. Man. 6,
468472.
Brittain, J.W., & Freeman, J.H. (1980). Organizational proliferation and entity develop-
ment in selection. In J.R. Kimberly & R.H. Miles (Eds.). The organizational life cycle
(pp. 291-341). San Francisco: Jossey-Bass.
Burgelman, R.A. (1984).Managing the internal corporate venturing process. Sloan Man-
agement Review, 25 (Winter), 33-48.
Burns, T., & Stalker, G.M. (1968). The management of innovation (2nd ed.). London:
Tavistock.
When a Thousand Flowers Bloom 127

Carroll, J. (1967). A note on departmental autonomy and innovation in medical school.

Journal of Business, 40 531-534.
Child, J. (1972). Organizational structure, environment and performance: The role of stra-
tegic choice. Sociology, 6, 1-22.
Cohn, S.F., & Turyn, R.M. (1984). Organizational structure, decision-making procedures,
and the adoption of innovations. I E E E Transactions on Engineering Management,
EM32 (November), 154-161.
Connolly, T., Conlon, E., & Deutsch, S. (1980). Organizational effectiveness: A multiple-
constituency approach. Academy of Management Review, 5,211-21 7.
Cyert, R.M., & March, J.G. (1963). A behavioral theory of the firm. Englewood Cliffs, NJ:
Prentice-Hall.
Dalton, M. (1959). Men who manage. New York: John Wiley & Sons.
Davis, S.M., & Lawrence, PK. (1977). Matrix, Reading, MA: Addison-Wesley.
Dean, R.C. (1974). The temporal mismatch-innovation’s pace versus management’s time
horizons. Research Management, 1 9 (May), 12-15.
Delbecq, A.L., & Mills, P.K. (1985). Managerial practices that enhance innovation. Organ-
izational Dynamics, 14 (Summer), 24-34.
Drucker, P.F. (1985). Innovation and entrepreneurship-Practice and principles. New
York: Harper & Row.
Engel, J.F., Kollat, D.T., & Blackwell, R.D. (1981). Diffusion of innovations. In R.R. Roth-
berg (Ed,). Corporate strategy and product innovation (pp. 472-481). New York: Free
Press.
Fast, N.D. (1979). The future of industrial new venture departments. Industrial Marketing
Management, 8 (November), 264-273.
Friedlander, E, & Scott, B. (1981).The use of task groups and task forces in organizational
change. In C. Cooper & R. Payne (Eds.), Groups at work (pp. 191-217). New York:
John Wiley & Sons.
Galbraith, J. (1982). Designing the innovating organization. Organizational Dynamics, 10
(Summer). 5-25.
Gamson, W.A. (1961). A theory of coalition formation. American Sociological Review, 26,
373-382.
Gamson, W.A. (1968). Coalition formation. International Encyclopedia of the Social Sci-
ences (pp. 529-534). New York: The Macmillan Co. and The Free Press.
George, R., & MacMillan, I.C. (1984). Corporate venturing-ventue management chal-
lenges. Working Paper, New York University.
Gladstein, D., & Caldwell, D. (1985). Boundary management in new product teams. In
Academy of Management Proceedings (pp. 161-165).
Hage, J., & Aiken, M. (1967). Program change and organizational properties: A compara-
tive analysis. American Journal of Sociology, 72, 503-579.
Hage, J., & Dewar, R. (1973). Elite values vs. organizational structure in predicting inno-
vation. Administrative Science Quarterly, 18 279-290.
Hanan, M. (1976). Venturing corporations: Think small to stay strong. Harvard Business
Review, 54 (May-June), 139-148.
Hobson, E.L., &Morrison, R.M. (1983). How d o corporate start-up ventures fare? In J.A.
Hornaday, J.A. Timmons, & Karl, H. Vesper (Eds.), Frontiers of entrepreneurial re-
128 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

search (pp. 3 9 0 4 1 0 ) . Wellesley, MA: Babson College Center for Entrepreneurial Stud-
ies.
Hollomon, J.H. (1981). Government and the innovation process. In R.R. Rothberg (Ed.).
Corporate strategy and product innovation (pp. 46-59). New York: Free Press.
Kanter, R.M. (1972). Commitment and community. Cambridge, MA: Harvard University
Press.
Kanter, R.M. (1977).Men and women in the corporation. New York: Basic Books.
Kanter, R.M. (1980).Power and change in and by organizations: Setting intellectual direc-
tions for organizatioml analysis. Paper presented at American Sociological Association
Annual Meeting.
Kanter, R.M. (1982). The middle manager as innovator. Harvard Business Review, 61
(July-August). 95-1 05.
Kanter, R.M. (1983). The change masters. New York: Simon & Schuster.
Kanter, R.M. (1984). Innovation: Our only hope for times ahead? Sloan Management Re-
view, 25 (Summer), 51-55.
Kanter, R.M. (1985). Supporting innovation and venture development in established cor-
porations. ]oumal of Business Venturing, 1 (Winter), 47-60.
Kanter, R.M. (1986). Creating the creative environment. Management Review, 75 (Febru-
ary), 11-12.
Kanter, R.M., & Stein, B.A. (Eds.) (1979). Life in organizations. New York: Basic Books.
Kaplan, R.E., & Mazque, M. (1983). Trade routes: The manager’s network o f relution-
ships (Technical Report No. 22). Greensboro, NC: Center for Creative Leadership.
Katz, R. (1982).Project communication and performance: An investigation into the effects
of group longevity. Administrative Science Quarterly, 27, 81-104.
Kidder, T. (1981). The soul of a new machine. Boston: Atlantic Little, Brown.
Kimberly, J.R. (1980). Initiation, innovation, and institutionalization in the creation proc-
ess. In J.R. Kimberly & R.H. Miles (Eds.). The organizational life cycle ( 1 8 4 3 ) . San
Francisco: Jossey-Bass.
Kimberly, J.R. (1981). Managerial innovations. In W.H. Starbuck (Ed.). Handbook of OT-
ganizatioml design (Vol. 1. pp. 84-104). New York: Oxford University Press.
Kimberly, J.R., & Evanisko, M.J. (1979). Organizational innovation. Working paper A27.
School of Organization and Management, Yale University.
Kohn, M. (1971).Bureaucratic man: A portrait and an interpretation. American Sociologi-
cal Review, 38 4 6 1 4 7 4 .
Lawrence, P.R., & Lorsch, J.W. (1967). Organization and environment. Boston: Graduate
School of Business Administration, Harvard University.
MacMillan, I.C., Block, Z., & Subba Narsimha, P.N. (1986). Corporate venturing: Alter-
natives, obstacles encountered, and experience effects. Journal of Business Venturing, 1
(Spring), 177-1 91.
Maidique, M.A. (1980). Entrepreneurs, champions, and technological innovation. Sloan
Management Review, 21 (Winter), 59-76.
Mansfield, E., Rapoport, J., Schnee, J., Wagner, S., &Hamburger, N. (1981).Research and
innovation in the modern corporation. In R.R. Rothberg (Ed.), Corporate strategy and
product innovation (pp. 4 1 6 4 2 7 ) . New York: Free Press.
When a Thousand Flowers Bloom 129

March, J.G. (1962). The business firm as a political coalition. Thelournal of Politics, 24,
662-678.
March, J.G., & Olsen, J.P (1976). Ambiguity and choice in organizations. Bergen, Nor-
way: Universitetsforlanget.
Marquis, D.G., & Myers, S. (1969). Szrccessfirl indrrstrial innovations. Washington, DC:
National Science Foundation.
Martilla, J.A. (1971). Word-of-mouth communication in the industrial adoption process.
Journal of Marketing Research, 8, 173-178.
Mechanic, D. (1962). Sources of power of lower participants in complex organizations.
Administrative Science Quarterly, 7, 347-364.
Mintzberg, H. (1981). Organizational design: Fashion or fit? Hrirvard Business Review, 59
(Jan.-Feb.). 103-1 16.
Mintzberg, H. (1983). Power in and around organizations. Englewood Cliffs, NJ: Prentice-
Hall.
Mohr, L.B. (1978). Explaining organizational behavior. San Francisco: Jossey-Bass.
Moritz, M . (1984). The little kingdom: The private story ofApple Computer. New York:
Morrow.
Murnighan, J.K. (1978). Models of coalition behavior: Game theoretic, social psychologi-
cal, and political perspectives. Psycho/ogical Bulletin, 85, 1130-1 153.
Nelson, R.R. (Ed.) (1962). The rate and direction of inventive activity. Princeton, NJ:
Princeton University Press.
Normann, R. (1971). Organizational innovativeness: Product variation and reorientation.
Administrative Science Quarterly, 16, 203-21 5.
O’Toole, J. (1983). Declining innovation: The failure of success. Human Resource Plan-
ning, 6, 125-141.
Pelz, D., & Andrews, E (1966). Scientists in organizations. New York: John Wiley & Sons.
Pennings, J.M., & Goodman, PS. (1977). Toward a workable framework. In PS. Good-
man, J.M. Pennings & Associates. N e w perspectives in organizational effectiveness (pp.
146-1 84). San Francisco: Jossey-Bass.
Perrow, C. (1981). Normal accidents. New York: Basic Books.
Pfeffer, J. (1981). Power in organizations. Boston: Pitman.
Pfeffer, J,, & Salancik, G.R. (1977). Organization design: The case for a coalition model of
organizations. Organizational Dynamics, 6 (Autumn), 15-29.
Pfeffer, J., & Salancik, G.R. (1978). The external control of organizations. New York: Har-
per & Row.
Pierce, J.L., & Delbecq, A.L. (1977). Organization structure, individual attitudes and inno-
vation. Academy of Management Review, 2 , 27-37.
Pondy, L.R. (1980). Organizational conflict: Concepts and models. In H.J. Leavitt, et al.
(Eds.). Readings in Managerial Psychology (pp. 473-492). Chicago: The University of
Chicago Press.
Powell, J. (1985). Bootstrap entrepreneurs at GTE TeleMessenger. GTE Together. Winter.
Quinn, J.B. (1979). Technological innovation, entrepreneurship and strategy. Sloan Man-
agement Review, 20 (Spring), 19-30.
Quinn, J.B. (1980). Strategies for change. Logical incrementalism. Homewood. IL: Irwin.
130 KNOWLEDGE hlANAGEMENT AND ORGANIZATIONAL DESIGN

Quinn, J.B. (1985). Managing innovation: Controlled chaos. Harvard Business Review, 63
(May-June), 73-84.
Riker, W.H. (1962). The theory of political coalitions. New Haven: Yale University Press.
Roberts, E.B. (1980). New ventures for corporate growth. Harvard Business Review, 58
(July-August), 134-142.
Robertson, T., & Wind, Y.(1983). Organizational cosmopolitanism and innovation. Acad-
emy of Management Journal, 26, 332-338.
Rogers, E.M., & Shoemaker, F.E (1971). Communication of innovations: A cross-cultural
approach (2nd ed.). New York: The Free Press.
Rothberg, R.R. (Ed.) (1981). Corporate strategy and product innovation. New York: Free
Press.
Ruttan, V.W., & Hayami, Y. (1984). Toward a theory of induced institutional innovation.
Journal of Development Studies, 20, 203-223.
Salancik, G.R., & Pfeffer, J. (1977) Who gets power-and how they hold on to it. Organ-
izational Dynamics, 5 (Winter), 2-21.
Sapolsky, H. (1967). Organizational structure and innovation. Journal of Business, 40,
497-5 10.
Schilit, W.K., & Locke, E.A. (1982). A study of upward influence in organizations. Admin-
istrative Science Quarterly, 2 7, 304-3 16.
Schon, D. (1967). Technology and change. New York: Delacorte.
Schon, D. (1971). Beyond the stable state. New York: Norton.
Schroeder, R., Van de Ven, A. Scudder, G., & Polley, D. (1986). Observations leading to a
process model of innovation. Discussion Paper No. 48, Strategic Management Research
Center, University of Minnesota.
Shubik, M . (1982). Game theory in the social sciences: Concepts and solutions. Cam-
bridge, MA: The MIT Press.
Starbuck, W.H. (1983). Organizations as action generators. American Sociological Review,
48,91-115.
Staw, B.M. (1975). Attribution of the ‘causes’ of performance: A general alternative inter-
pretation of cross-sectional research on organizations. Organizational Behavior and Hu-
man Performance, 13, 414-432.
Stein, B.A., & Kanter, R.M. (1980). Building the parallel organization: Toward mecha-
nisms for permanent quality of work life. Journal of Applied Behavioral Science, 16,
371-3 8 8,
Stevenson, H., & Gumpert, D. (1985). The heart of entrepreneurship. Harvard Business
Review, 64 (March-April), 84-94.
Stumpe, W.R. (1979). What the research manager should know about new product psy-
chology. Research Management, 22 (March), 13-1 7.
Summers, D.V. (1986). Organizing in middle management: A politico-structural model of
coalition formation in complex organizations. Ph.D. dissertation. Department of Sociol-
ogy, Yale University.
Thompson, J.P (1967). Organizations in action. New York: McGraw-Hill.
Thurman, B. (1979-1980). In the office: Networks and coalitions. Social Networks, 2,47-
63.
When a Thousand Flowers Bloom 131

Tichy, N., & Fombrun, C. (1979). Network analysis in organizational settings. Human Re-
lations, 32, 923-965.
Tosi, H., & Tosi, L. (1986). What managers need to know about knowledge-based pay.
Organizational Dynamics, 24 (Winter), 52-64.
Trist, E. (1981). The evolution of sociotechnical systems as a conceptual framework and as
an action research program. In A. Van de Ven & W. Joyce (Eds.).Perspective on organi-
zation design and behavior (pp. 19-75). New York: John Wilcy & Sons.
Tushman, M. (1977). Special boundary roles in the innovation process. Administrative Sci-
ence Quarterly, 22, 587-605.
Tushman, M. (1979). Work characteristics and subunit communication structure: A con-
tingency analysis. Administrative Science Quarterly 24, 82-98.
Tushman, M., & Moore, W.L. (Eds.) (1982). Readings in the management of innovation.
Marshfield, MA: Pitman.
Tushman, M., & Nadler, D. (1986). Organizing for innovation. California Management
Review, 28 (Spring), 74-92.
Van de Ven, A.H. (1986). Central problems in the management of innovation. Manage-
ment Science, 32, 590-607.
Vesper, K.H. (1984). Three faces of corporate entrepreneurship: A pilot study. Working Pa-
per, University of Washington.
Von Hippel, E. (1977). Successful and failing internal corporate ventures: An empirical
analysis. Industrial Marketing Management, 6, 163-1 74.
Von Hippel, E. (1982). Get new products from customers. Harvard Business Review, 60
(March-April), 117-122.
Von Hippel, E. (1981). Users as innovators. In R.R. Rothberg (Ed.), Corporate strategy
and product innovation (pp. 239-251). New York: Free Press.
Walton, R. (1975). The diffusion of New York structures: Explaining why success didn’t
take. Organizational Dynamics, 4 (Winter), 3-21.
Walton, R. (1987). Innovating to compete. San Francisco: Jossey-Bass.
Weber, M. (1978). Guenther Ross and Claus Wittich, (Trans.) Economy and society (Vol.
2). Berkeley, CA: The University of California Press.
Weick, K.E. (1976). Educational organizations as loosely coupled systems. Administrative
Science Quarterly, 21, 1-19.
Weick, K.E. (1979). The social psychology oforganizing (2nd ed.). Reading, MA: Addison-
Wesley.
Wiewel, W., & Hunter, A. (1985). The interorganizational network as a resource: A com-
parative case study on organizational genesis. Administrative Science Quarterly, 30,
482-497.
Wortman, C.B., & Linsenmeier, J.A.W. (1977). Interpersonal attraction and techniques of
ingratiation in organizational settings. In B.M. Staw & G. Salancik (Eds.), New Direc-
tions in Organizational Behavior (pp. 133-178) Chicago: St. Clair Press.
Yin, R.K. (1977). Production efficiency versus bureaucratic self-interest: Two innovative
processes? Policy Sciences, 8, 381-399.
Zagare, F.C. (1984). Game theory concepts and applications. Beverly Hills, CA: Sage.
Zaltman, G., Duncan, R., & Holbeck, J. (1973). Innovations and organization. New York:
John Wiley & Sons.
This page intentionally left blank
 Joseph Badaracco

Managers are playing with fire when their company does not own and control its
crucial resources, core capabilities, and key technologies. But what about the
knowledge, resources, and skills that play supporting roles? As embedded knowl-
edge and specialized capabilities are created in a growing number of companies
and other organizations around the world, firms often find it too costly and cum-
bersome to develop, on their own, all the knowledge and capabilities they need or
want to have available. In the words of IBM president, Jack Kuehler, “It’s a dan-
gerous thing to think we know everything.”’
In response, many firms are creating knowledge links-alliances that give
them access to the skills and capabilities of other organizations and sometimes en-
able them to work with other organizations to create new capabilities. Knowledge
links can be tactical or strategic. A single knowledge link can help a company
build new skills in a limited area of its operations. This is a tactical effort. In con-
trast, when a company creates a multitude of knowledge links with customers,
suppliers, labor organizations, universities, and other organizations, and when
these alliances strengthen each other and support the company’s long-term objec-
tives, then knowledge links are genuinely strategic.
This is how G M and IBM have sought to use many of their knowledge links
during the 1980s. Their new knowledge-intensive relationships have played im-
portant supporting roles in each company’s effort to renew and reshape its core
capabilities and to change the competitive rules of the game in its favor. Such rna-
jor strategic efforts are usually accompanied by changes in organizational struc-
ture, and this has been the case at both IBM and GM. They are now structured
more like city-states and less like citadels, because of their growing reliance on
knowledge-rich alliances. Like city-states, each firm has at its core a dense net-
work of relationships defined by ownership, control, and social bonds. It is no
longer easy to define what is inside and outside the two companies. Instead, each
firm is linked to other organizations through a multitude of arrangements in
which control and ownership are shared, social bonds blurred, classical contract-
ing compromised, and embedded knowledge is transferred, renewed, and created.

Reprinted by permission of Harvard Business School Press from Chapter 5 of The Knowledge Link:
How Firms Compete Through Strategic Alliances. Boston: 1991, pp. 107-128. Copyright 0 1991 by
the President and Fellows of Harvard College.
133
134 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

G M and IBM, like many other American firms, would be creating far fewer
knowledge links today if it were not for the economic threats posed by competi-
tors in Japan and elsewhere. By the 1980s, many of these foreign competitors had
combined knowledge they had secured from the United States with their indige-
nous capabilities and were vying for the lead in worldwide technology and prod-
uct contests. Sometimes, they held the lead. In the mid-l980s, for example,
Toyota and other Japanese auto companies were pioneers in introducing powerful
multivalve engines. At the same time, Japanese and U.S. car makers were racing
each other to develop small, fuel-efficient, low-cost two-stroke engines and “ac-
tive suspension systems” that electronically sensed road conditions and adjusted a
car’s ride. In computers, Fujitsu sold mainframes that ran faster than comparably
priced IBM machines. (In the 1970s, a single American firm, Cray Research, had
dominated the supercomputer field.) By the late 1980s, breakthroughs were tak-
ing place at Fujitsu, Hitachi, and NEC, as well as at small, startup firms in the
United States such as Sequent and Thinking Machines. These products were the
progeny of highly refined and specialized capabilities. Companies and countries
were competing to develop, not just the latest products, but the capabilities to de-
velop, refine, and sometimes revolutionize these products year after year.
American firms often create product links in response to such developments,
thereby quickly securing products that other companies already have. Knowledge
links d o more. They are, in effect, a higher step in an evolutionary chain of alli-
ances, Like product links, their forebears, they usually produce products or serv-
ices, and they can also help to reduce risks, cut costs, increase speed to market,
and so forth. But knowledge links also help the partners learn and sometimes cre-
ate new capabilities. In fact, this is often a prerequisite for the alliance’s success:
without acquiring new knowledge, many partnerships could not produce the
products or services that its parent organizations want.

CHARACTERISTICS OF KNOWLEDGE LINKS

The first distinguishing trait of knowledge links is that learning and creating
knowledge is a central objective of the alliance. Knowledge links can help one
company learn specialized capabilities from another; they can help a company
combine its special capabilities with those of another organization to create new
embedded knowledge; and they can enable one company to help another organi-
zation build up its skills and capabilities in ways that will benefit both companies
later on.
Second, knowledge links are more intimate than product links. In order for
two organizations to learn, create, or strengthen specialized capabilities, person-
nel from each must work together closely. This would not be the case if the com-
panies were trying to transfer migratory knowledge: then they could simply
exchange cash for a book of blueprints or a set of formulas. When companies seek
to learn embedded knowledge from each other, their relationship resembles that of
a master and an apprentice, which Michael Polanyi describes in this way:
Knowledge Links 135

You follow your master because you trust his manner of doing things even
when you cannot analyze and account in detail for its effectiveness. B y
watching the master and emulating his efforts in the presence of his exam-
ple, the apprentice learns unconsciously, picks up the rules of the art, includ-
ing those which are not explicitly known to the master himself. These
hidden rules can be assimilated only by a person who surrenders himself un-
critically to the imitation of another.=

The third distinctive feature of knowledge links is the extraordinarily wide

range of partners with which these links can be formed. Product links are usually
formed with competitors or potential competitors. GM, as we have seen, turned
to four Asian car companies to help it fill the small car gap in its product line.
Knowledge links, in contrast, can be formed with virtually any other organiza-
tion-as long as it has a specialized capability to contribute to the partnership.
Through knowledge links, buyers and suppliers can share expertise on manufac-
turing processes and work together to improve both the buyer’s product and the
components the supplier provides. Through knowledge links, university laborato-
ries and companies share and create knowledge. Knowledge links can also include
participatory, cooperative relations between companies and their workers and la-
bor unions. Through these, managers learn from workers how to make higher-
quality products and how to d o so more cheaply and efficiently. At the same time,
company-union alliances often involve extensive training programs, so that work-
ers become “multiskilled”: instead of performing simple, repetitive tasks, workers
develop, as individuals and as teams, the broader range of capabilities that a com-
pany needs.
Finally, knowledge links differ from product links because of their greater
strategic potential. Product links help one company catch up, buy time, defend it-
self, or recapture its investment in fixed costs by selling the product quickly and in
high volume through partners around the world. Knowledge links can help a firm
extend or modify one of its basic capabilities, and a constellation of knowledge
links can contribute to a larger strategic effort to renew core capabilities or create
new ones.
Knowledge links and product links differ more sharply in theory than in
practice. Both are members of the same organizational species, corporate alli-
ances, and both blur traditional firm boundaries through shared ownership and
control, linked social systems, and departures from classical, arm’s-length con-
tracting. The two kinds of alliances often differ from each other in degree, not in
kind. Just as the evolution of life has proceeded through trial-and-error adapta-
tion to varied circumstances, so alliances have produced a spectrum of hybrids in
response to the needs of companies. At one end are nearly pure cases of product
links, where learning is much less important than access to a product or wider dis-
tribution for an existing product. At the other end, the parties seek to learn or cre-
ate new capabilities as well as to develop a new product. Many alliances fall into
the middle ranges of this spectrum.
The GM-Suzuki alliance and IBM’s early PC alliances fall nearer the end of
the spectrum defined by pure product links. In these alliances, learning played a
136 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

small role. Neither company sought to learn new capabilities or to achieve a prod-
uct breakthrough. To be sure, the parties did need to learn enough about each
other to coordinate activities, and Suzuki needed to learn from G M about its
product specifications and US. regulations. Suzuki also received some engineering
assistance from GM. But the main aim of the alliance was not for G M to learn
from Suzuki, nor for Suzuki to learn from GM. G M needed a product quickly,
and Suzuki wanted financing for a subcompact it had designed and access to dis-
tribution in the United States without the cost and risk of creating its own dealer
network. This alliance, in its early phases, was a nearly pure case of a product
link.
The GM-Daewoo alliance involved more learning than the GM-Suzuki deal,
but mainly for Daewoo, which learned about many aspects of automobile manu-
facture. While G M learned more about the Korean market, its principal aim was
to secure another source of low-cost small cars, and it was not creating a new
product. G M provided Daewoo with existing G M technology and a vehicle from
its Adam Ope1 subsidiary that G M was already making and selling in West Ger-
many. In doing so, G M did not develop new knowledge or capabilities. The Dae-
woo alliance, for GM, was a product link; for Daewoo, a knowledge link.
NUMMI was a more complex hybrid. It was, in part, a product link, pro-
viding G M with access to a large number of very high-quality small cars. The
product was not new (it was a car Toyota was already making and selling in Ja-
pan), but NUMMI helped both companies learn and create new capabilities. It
helped Toyota learn about managing U.S. workers, suppliers, and trucking firms,
and about dealing with the UAW and state and local governments. G M gained the
opportunity to learn firsthand about the Toyota production system-its collabo-
rative approach to worker and supplier relations, its just-in-time inventory man-
agement, and its highly efficient plant management. Most important, both
companies were jointly creating new knowledge and capabilities, each experi-
menting with a United States-Japanese way of managing an auto plant.
Further along the spectrum was the alliance announced in early 1990 by
IBM and Siemens, the West German electronics giant. In this case, the partners
planned to pool existing skills and to acquire new ones in the design, manufac-
ture, and testing of computer chips, in order to develop chips two generations
more sophisticated than any on the market. To be sure, IBM brought greater tech-
nological prowess to the alliance, and, like a product link, the partnership would
manufacture a product and reduce the partners’ financial risks (designing an ad-
vanced microchip and building a plant were expected to cost over $1 billion). But
the creation of new capabilities played a vital role in the partnership and in this
way it differed dramatically from the GM-Suzuki or IBM PC alliances.

EXTENDING CAPABILITIES THROUGH ALLIANCES

When go-it-alone strategies, classic market transactions, or mergers and ac-
quisitions seem unable to meet a company’s needs, knowledge links can help a
company gain access to the capabilities of other organizations or work with them
Knowledge Links 137

to create new capabilities. One of GM’s knowledge links, the GMFanuc Robotics
Corporation (GMF),demonstrates how a knowledge link can help two companies
turn the challenges of embedded knowledge into opportunities for each firm to
extend and broaden its capabilitie~.~
In 1982, G M and Fanuc, the Japanese controls and robotics company, each
invested $5 million to create GMF. Its charter was to design, market, service, and
develop applications for factory automation robots. Technology would flow to
and from GMF’s parents, without royalties or licenses. GMF expected to move far
beyond “duck-drinking-water” robots that perform the same simple task at the
same spot on the same product time and time again. Future robots would have
sensory functions; they would use television cameras and laser beams or extend
sensitive probes to locate objects, reducing the need for manufacturers to develop
ways of aligning objects for robot processing. Clearly, GMF was a knowledge link
between G M and Fanuc.*
Fanuc was the personal handiwork of Dr. Sieuemon Inaba, its founder and
chief executive. In 1955, Fujitsu, the Japanese electronics and computer company,
placed him in charge of a team of 500 engineers whose mission was to develop a
factory automation business. Under Inaba’s leadership, Fanuc became the world’s
leader in computerized numerical controls-electronic boxes that control the
movement of machine tools such as lathes and milling machines. Fanuc was an ul-
tra-workaholic company: 14-hour workdays were ordinary for managers and re-
searchers. Inaba ran Fanuc with military precision. In fact, Fanuc and G M formed
their venture within three months of their first contact. Although this pace of de-
cision making was almost unheard of a t GM, it was quite natural for Fanuc,
where Inaba had installed a clock in the product development lab that ran at 10
times normal speed.
G M executives gave several reasons for joining forces with Fanuc. First, as
the largest U.S. user of robots, G M bought about a third of all robots sold in the
United States. These, along with tens of thousands of computers and numerical
control units, were part of GM’s high-tech manufacturing strategy. Second, G M
was dissatisfied with some of its own robot vendors. Third, G M had developed an
expertise in robotics and wanted to find a way to convert this knowledge into
products and sell it. Fourth, GM was afraid it might lose some of its robotics per-
sonnel and technology to other robotics companies. Inaba joined with G M be-
cause he wanted to build Fanuc’s robotics business. In particular, he felt limited by
Japanese robot technology. He believed that U.S. and European firms were ahead
of the Japanese in developing intelligent robots with visual functions, robots capa-
ble of walking around factory floors and offices on their own feet, and robots
connected with CAD/CAM systems.

‘As is often the case, GMF served many of the purposes of a product link. The company, which
would set up its headquarters and manufacturing facilities in Michigan, would have the exclusive
right to sell robots made in Japan by Fanuc throughout North and South America, Australia, and
New Zealand. In this way, GMF would help Fanuc increase its volume of operations and thereby
make additional contributions to its fixed costs.
138 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

GMF development efforts were coordinated by meetings held four times a

year and involving senior executives from GMF and Fanuc. Teams of engineers
from Fanuc, GMF, and GM conducted individual projects. Some projects concen-
trated on hardware, aiming at developing smaller, more economical, and more
specialized units. Others focused on programming and communications language.
Perhaps the most extensive effort was the development of Karel, a programming
language that linked GMF robots, vision systems, and other devices to both GMF
and non-GMF products.
In creating GMF, its parents avoided the difficulties of trying to work to-
gether through a series of arm’s-length market relationships. They were able to
collaborate even though a merger of the two companies was impossible. Neither
company had to rely wholly on its own resources for further development in ro-
botics. They concentrated only on the particular projects that interested both par-
ties. Their radically different company cultures could be kept largely separate and
intact, while the project-driven interactions could be carefully monitored and
managed. And, above all, key researchers and engineers from both companies
could work together, day by day, in order to learn from each other, create adapta-
tions of the technology and expertise that each had developed alone, and in so do-
ing strengthen their capabilities and those of their partners.
The GMF example is instructive but somewhat limited. It does display two
of the principal characteristics of knowledge links. GMF helped both partners
broaden certain skills and capabilities, and it also created intimate working rela-
tionships among personnel from the parent organizations. GMF does not, how-
ever, display the two other characteristics of knowledge links. One is the way a
company can use a multitude of knowledge links as part of an effort to transform
its core capabilities. The other is the variety of knowledge partners from which a
firm can choose.

TRANSFORMING CAPABILITIES THROUGH ALLIANCES

To understand these last two characteristics, it is necessary to take a strategic

view of the many alliances G M and IBM created in the 1980s. Examining the
risks, rationales, and structures of individual alliances is not enough. Such an ap-
proach results in scrupulous analysis of trees but little understanding of forests. It
fails to answer a crucial question: how d o all of a company’s alliances relate to
each other and how do they support and renew a firm’s core capabilities? An an-
swer requires an analysis of a company’s strategy, its other alliances, changes in its
core operations, and trends in its industry.
Viewed cumulatively rather than one at a time, the knowledge links created
by G M and IBM in the 1980s were genuinely strategic. They were not simply
product links, aimed at filling gaps in a product line or helping to cover fixed
costs. Nor were they simply efforts to add a handful of new capabilities to the rep-
ertories of these two giants. Instead, IBM used a myriad of knowledge links as
part of a larger effort to transform its traditional core capability of making and
selling mainframe computers. G M began to use knowledge links with its suppli-
Knowledge Links 139

ers, its workers and their union, and its dealers, as part of a bold, perilous effort
to dramatically alter its capabilities for designing and manufacturing cars. IBM, in
short, was trying to transform its products; GM, its operations.
The strategies of both firms were also preemptive: they aimed to change the
terms of competition in their industries and tilt the playing field in their favor.4
The 1990s will indicate whether G M and IBM succeed in their daring efforts;
meanwhile, Niccolo Machiavelli’s observation is pertinent to both firms: “There is
nothing more difficult to plan, more doubtful of success, nor more dangerous to
manage than the creation of a new order of things.”’

IBM’s Knowledge Links

Since the mid-l950s, IBM’s core capability has been financing, designing,
manufacturing, and selling mainframe computers. For most of this time, main-
frames dominated the computer industry, and IBM dominated the mainframe
business. As a result, it became one of the most profitable companies in the history
of commerce. By the 1980s, however, the computer industry began to shift course
dramatically and IBM’s environment became less stable and more hostile.
Above all, IBM had to respond to the radical changes in the computer indus-
try, which were driven by trends toward ever-cheaper computer power and to-
ward larger, better-integrated networks. Kojii Kobayashi, the chairman of NEC, a
major Japanese computer firm, described these developments by distinguishing
between “point” and “space.” Point represents the mainframe-dominated era in
which all computerized data flowed to and from a single centralized machine.
Space represents a future era in which networks of powerful local machines dis-
tribute data-processing capability throughout a company, a country, or the world.
The shift from point to space gives a radically different answer to the question
“What is a computer?” The old answer was: a solitary central-processing unit.
The new answer: a computer is a network.
These developments gave rise to a surge of new entrants in the computer in-
dustry. They aggressively attacked the developing areas of the industry as well as
its traditional market segments, including IBM’s inner sanctum, the mainframe
business. Moreover, entrants were not the only aggressors. Established computer
companies, long accustomed to following in IBM’s wake, were reinvigorated and
emboldened by the prospect of competing against the giant on a new playing field
with a new set of rules. Hundreds of companies were developing new capabilities
that IBM might or would need to serve its customers. These capabilities took four
basic forms: designing high-powered, customized computers for special tasks; us-
ing intimate knowledge of a customer’s particular needs to write software for the
customer; bringing state-of-the-art technology to market quickly in a rapidly
changing and intensely competitive business; and using familiarity with customer
requirements to design customized computer networks.
The transition from point to space reinforced old threats to IBM and created
new ones. Computer buyers wanted to build networks using compatible hardware
and software from a variety of suppliers. In computer jargon, they preferred open
140 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

architecture to the proprietary systems of a single supplier. Almost every major

computer company was offering customers an open architecture based on Unix,
the operating system software that Bell Laboratories had developed in the early
1 9 7 0 ~By . ~ 1987, AT&T had licensed Unix to 225 computer firms, including the
two largest Japanese computer makers, Hitachi and Fujitsu, which both sold
mainframes that ran Unix. For small specialized computer manufacturers trying
to compete in niches against IBM or DEC, Unix offered a way to provide products
that were compatible with and connected to a wide range of different computer
systems. Some industry estimates suggested that nearly 25% of all computers in
the world, as measured by dollar value, would use Unix by the end of 1991.’
In essence, Unix was accelerating the trend toward building ever more pow-
erful minicomputers, workstations, and microcomputers and toward linking these
machines in a vast communications network. It threatened to create an alternative
computer standard to IBM’s proprietary operating systems. A specter haunting
IBM was the prospect of a loose, worldwide alliance of Unix-based manufactur-
ers, including IBM’s most daunting Japanese competitors as well as the growing
number of Unix customers.
In response to this threat and to the growth of new capabilities throughout
the industry, IBM created scores of knowledge links. Through them it hoped to se-
cure access to and build new skills and, ultimately, to transform its basic capabili-
ties. IBM fought the battle to survive and prosper on two principal fronts: in the
United States, its largest market, and in Japan, where it could counterattack its
Japanese adversaries in their home market.
In telecommunications, for example, IBM Japan created a series of collabo-
rative endeavors with NTT, the Japanese telecommunications monopoly, which
the government was gradually deregulating in the 1980s. The two companies
worked together to develop hardware, software, and organizational capabilities
in large-scale computer networks. IBM Japan also formed alliances with Mit-
subishi, Japan’s largest trading company, to create a joint satellite communica-
tions service in Japan and to design and sell equipment for information network
services. As a trading company, Mitsubishi wanted to combine IBM’s technologi-
cal know-how with its own expertise in managing a global network of trading op-
erations and customer relationships.*
Starting in the mid-1970s, IBM USA formed a series of alliances in the tele-
communications business, One of its partners was MCI communications. MCI
was AT&T’s major U.S. competitor in long-distance telephone services, and its
market share was expected to climb from 1 0 % in 1987 to nearly 14% by 1991.
Through the deal, MCI gained credibility, customers, and access to a communica-
tions network that IBM was developing, while IBM strengthened its position in
telecommunications by becoming a major shareholder in the industry’s most im-
portant independent company. IBM and MCI engineers and managers collabo-
rated on large communications projects for the government and for firms creating
private networks. In the late 1980s, MCI decided to buy back IBM’s 16% stake,
but industry analysts expected the two companies to continue working closely to-
gether because of their complementary skills in computer equipment and commu-
nications linesW9
Knowledge Links 141

In the 1980s, IBM also formed an alliance with Rolm, one of the world’s
leading manufacturers of private branch exchanges (PBXs), which are, in effect,
computers that switch telephone calls and data. IBM hoped that Rolm would pro-
vide it with capabilities to route data and voice communications within an infor-
mation network. Rolm, on the other hand, hoped to benefit from the cash the
IBM deal provided, from IBM’s corporate relationships, and from IBM’s capabili-
ties and overseas marketing. The IBM-Rolm alliance was ultimately unsuccessful;
IBM soon bought all of Rolm’s shares and made it a subsidiary, and later IBM
placed Rolm’s operations in an alliance with Siemens. Despite the turbulence and
frustrations, the alliances with Rolm and, later, with Siemens were a part of IBM’s
worldwide effort to build up its capabilities in telecommunications.
To develop applications software and other capabilities, IBM created scores
of cooperative alliances. In Japan, it formed Nissan Systems Development with
Nissan Motor to create applications software for basic research, product develop-
ment, and manufacturing, particularly in automotive electronics. The Mitsubishi
Bank and IBM Japan embarked on a joint venture to develop and sell software for
Japanese banks. With Nippon Kokan, the second largest Japanese steel maker,
IBM jointly developed an artificial intelligence system for planning steelmaking
schedules. IBM Japan also worked with many of its newly created dealerships,
such as liquor wholesalers and heating oil distributors, to develop specialized soft-
ware that could be sold to them and to other companies in their businesses.’”
In the United States, in the late 1980s, IBM USA created more than a dozen
alliances-usually through minority equity investments-with companies that
had expert capabilities in particular areas of software design. These included firms
specializing in molecular simulation, image processing, insurance company trans-
actions, management of large commercial projects, and software used to design
software. Other new partners, such as a producer of signal converters for fiber op-
tics communications, made hardware that would help IBM build customized net-
works. IBM’s alliances also included a joint venture with Stephen Chen, one of the
world’s leading supercomputer designers; the creation and partial financing of Se-
matech, a consortium of U.S. semiconductor manufacturers; and an R&D part-
nership with Motorola to improve semiconductor manufacturing. Through these
partnerships, IBM secured access to capabilities developed by firms that com-
peted, in effect, in niches within niches. The firms relied upon specialized areas of
expertise that IBiM lacked and did not intend to develop on its own, but that it
needed to provide to some of its customers. Through the Sematech, Motorola,
and supercomputer alliances, it aimed to strengthen its partners and work with
them to create new capabilities.
Through this constellation of knowledge links, IBM was slowly transform-
ing itself from a supplier of mainframe hardware into an international computer
and telecommunications firm that could provide global companies, as well as
smaller firms, with companywide networks for transmitting and processing voice,
data, and images.” In essence, IBM was changing itself to adapt to and capitalize
on the transition from point to space. Its scores of knowledge links were part of
its effort to become efficiently global and, simultaneously, intensively local and
responsive. IBM could offer a company with worldwide operations, such as
142 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

Nomura Securities or Ford Motor, customized hardware and software that linked
all their desks, offices, machines, and factories around the world. IBM’s national
subsidiaries, like IBM Japan, could also rely upon their partners to offer the local
offices of Nomura or Ford customized applications software and computer net-
works.

GM’s Knowledge Links

Viewed one at a time, many of GM’s alliances, like GMF Robotics, served to
extend particular capabilities. Cumulatively, however, they also contributed to
most dramatic strategic changes at GM in the past half-century. Like IBM, GM
has used a multitude of new alliances as part of a strategy of transformation. The
alliances have given GM access to capabilities and pools of expertise that it needs,
but that it chose not to invent or reverse engineer or develop on its own. GM’s
vast array of new alliances can appear confusing at first glance. Fortunately, a sin-
gle effort-the Saturn project-can introduce GM’s new boundary arrangements
and can provide a helicopter view of GM’s effort to transform itself. Saturn and
Saturn-like changes throughout GM also represent GM’s bid to recapture leader-
ship in the world automobile industry and to hold it well into the twenty-first cen-
tury. In particular, GM threw down the gauntlet to Toyota, challenging it to
venture forth from its redoubt in Toyota City and confront GM in North Amer-
ica, in a battle to be fought with capital and advanced technology.
The basic facts about the Saturn subsidiary, GM’s sixth car division, are
straightforward. In 1983, GM announced that it would spend $3.5 billion to cre-
ate a compact car called Saturn, the first new GM nameplate since the introduc-
tion of Chevrolet in 1918. The Saturn plant, a new facility to be located in Spring
Hill, Tennessee, would employ 6,000 workers and produce 500,000 cars a year,
beginning in 1989. Because it would manufacture all the major car parts, the
plant would include extensive foundry, machining, engine, and transmission as-
sembly facilities, as well as metal stamping and final assembly operations. Parts
and components imported from Japan would account for less than 1 % of the cost
of the Saturn, a dramatic shift from GM’s reliance on car and parts suppliers in its
portfolio of Asian alliances. In 1986, GM scaled down plans for Saturn because of
capital constraints, halving its budget to about $1.7 billion. Its initial capacity
would be 240,000 cars per year, and the first cars would be launched in the fall of
1990. In the early 1990s, GM would start a second construction phase to double
Saturn’s capacity.
What this brief overview does not make clear is that the Saturn project was
hardly an ordinary car operation.12It was designed on a “clean sheet of paper”
basis, so that its personnel could design, engineer, manufacture, distribute, and
sell cars in pathbreaking ways. GM chairman Roger Smith announced:

GM believes that what it is doing is potentially significant to anyone who

operates a plant in any industry, anywhere in the United States, because the
Knowledge Links 143

leading-edge technologies that Saturn represents could affect every one of

them. And the improvements that flow naturally from Saturn could ulti-
mately dwarf past accomplishments and establish U S . industry, once again,
as the leader in a new age of almost cosmic industrial a ~ h i e v e m e n t . ’ ~

Martin Weitzman, a labor relations expert and economics professor at MIT,

said that “if Saturn succeeds, you can legitimately call it r e v ~ l u t i o n a r y . ”Busi-
~~
nessWeek described Saturn as a “bid to d o nothing short of revolutionizing auto-
mobile manufacturing.”” Saturn would draw on technology and skills from
within GM, from EDS and Hughes Aircraft (GM’s major high-tech acquisitions
of the 1 9 8 0 ~ from
) ~ the many partnerships already described in this chapter,
and from a new set of knowledge links with the UAW, with suppliers, and with
dealers.
The alliances with the last three groups were a vital part of the transforma-
tion G M sought to achieve. For the UAW, Saturn represented a new era of collabo-
ration. The project marked the first time in G M history that the UAW had
participated in G M corporate planning. In the words of president Owen Beiber,
the union would be a “full partner” in all of Saturn’s decision making, and “no
decision could be reached without its approval.”16Saturn would be run by a stra-
tegic action council, consisting of the project’s president and staff and the top
UAW advisers. A “manufacturing action council,” also including UAW repre-
sentatives, would oversee the day-to-day operations, and work units-teams of 6
to 15 workers led by a UAW “councilor”-would perform manufacturing and as-
sembly tasks.
Saturn sought to go beyond even NUMMI in encouraging participatory la-
bor relations. G M had many motives for pursuing this new relationship, but a
crucial one involved knowledge and capabilities. Toyota’s approach to labor rela-
tions succeeds, in part, because it motivates workers to develop thousands of ideas
about ways of improving automobile manufacturing and design. It thus makes
practical use of the knowledge and experience of the workers who make the cars
and their components. Toyota’s approach also provides workers with incentives
and opportunities to communicate what they know to the rest of the organization.
Moreover, Toyota-style manufacturing builds embedded knowledge at the
individual and team levels. Its workers are cross-trained in a variety of skills.
Hence, they can detect flaws in each other’s work, suggest ways they can work to-
gether more effectively, and gradually develop higher levels of skill through expe-
rience and training. (At one G M plant, 30 cross-trained workers reduced
warranty costs on suspension systems by 400% in a mere two years.) Finally,
cross-training helped workers handle the increasingly complex, often computer-
ized equipment on the factory floor. Saturn executives wanted every member of
the organization to contribute steadily and vigorously to the creation and dissemi-
nation of knowledge. Saturn would succeed if the new partnership between GM,
the UAW, and its workers could create a powerful information-sharing team. This
approach departed radically from traditional, arm’s-length, adversarial manage-
ment-union relationships in the auto industry, and it sought to halt the decades-
144 KNOWI.EDGE htANAGEh4ENT AND ORGANIZATIONAL DESIGN

long trend toward deskilling a~itomobilemanufacturing and “dumbing down”

the tasks it required of workers.
For similar reasons, at Saturn and at many other G M facilities, G M was
seeking to create new collaborative relationships with its suppliers. In the past,
G M Iiad handled most of its suppliers by-in the words of one of its purchasing
executives-“handing them our list of requirements and asking for a sealed bid.”
In effect, GM had practiced classical, arm’s-length, market contracting. Many
supplier relationships had been renewed on a year-to-year basis, and G M had
often switched suppliers to secure lower prices. At Saturn, in contrast, supplier re-
lations would be far more intimate and collaborative. One reason was the just-in-
time system that Saturn was implementing. JIT requires deliveries on an hourly
rather than a weekly or monthly basis and hence calls for close communication
and coordination with suppliers. GM’s managers said that Saturn would establish
long-term relationships with suppliers that would meet its time and quality de-
mands, rather than shop around each year for the lowest-cost supplier. iMoreover,
suppliers could achieve “preferred supplier” status. That meant G M would work
with them on product development, soliciting ideas and assisting them with de-
signs and component production. Like GM’s relationship with the UAW, this ef-
fort enlisted suppliers in the creation and dissemination of new knowledge.
Through close collaboration, G M and its suppliers would work to develop pro-
prietary technology that would give G M distinctive advantages and to find ways
to enable G M to use the technology quickly and efficiently. G M needed these ad-
vances badly, in part to recover from the cost-cutting and parts standardization ef-
forts that had led to look-alike cars some analysts nicknamed “Oldsbuicadillacs.”
G M developed its capabilities in dara processing through a wide network of
knowledge links with its computer and computer-related suppliers. Together, they
worked to create what GM called its Manufacturing Automation Protocol
(MAP). In the past, GM’s computer suppliers had provided proprietary systems
and hardware that could not communicate with each other or could do so only
through complex, expensive interface devices. MAP asked the vendors to inte-
grate their islands of computing and to help G M create a common communica-
tion network based on a nonproprietary language-the equivalent of asking
hundreds of locomotive manufacturers to alter their products so they could all run
on the same gauge track, To develop MAP, G M engaged in ongoing technical dis-
cussions with committees representing scores of hardware and software suppliers,
ranging from giants like IBM and DEC to small hardware and software shops.
These discussions are expected to last well into the 1990s, as the committees de-
velop a succession of communication standards, each moving a step closer to full
compatibility.
G M also made minority equity investments in small, high-tech companies-
such as Automatix, View Engineering, Diffracto, and Robotic Vision Systems-
that provided opportunities for its engineers to work with and learn from their
counterparts about advanced automation technology and also gave G M and its
partners the opportunity to collaborate in designing products for the auto indus-
try. For example, in 1988, G M and Teknowledge announced that they had devel-
Knowledge Liiiks 145

oped an expert system-an artificial intelligence software program that simulates

the judgments of experts in a particular field-that would troubleshoot various
types of machine tools, metal-cutting systems, and assembly machines. It could
detect problems stemming from conditions such as bearing wear, unbalanced
parts, and misalignment. At MIT’s Media Lab, a multidisciplinary research center
studying communications and computer technology, G M joined with the U.S.
Defense Department on projects studying holography and human-machine inter-
faces.”
Finally, Saturn’s planners wanted its distribution system to break down the
arm’s-length, highly contractual, often adversarial relations between G M and its
dealers. Hence, Saturn would have its own franchise system and dealer organiza-
tion. Most important, however, the franchise agreements stated specifically that
the dealers were to be “partners” in Saturn’s operations. They would be involved
in decisions not only about dealerships but also about product planning. As with
suppliers and workers, G M sought to shift to a new relationship in which it could
learn from and work more closely with its dealerships.

CONCLUSION
This overview of the strategies of G M and IBM shows how they have used
knowledge links to meet the challenges of embedded knowledge. At the simplest
level, these alliances served as organizational devices to help them avoid the diffi-
culties of trying to gain access to or create embedded knowledge through the tra-
ditional methods of market relationships, acquisitions, or going-it-alone. But
knowledge links can make broader contributions. They can help a company ex-
tend its expertise in one or more directions. And, if a company is willing to act as
boldly as G M and IBM have, they can contribute to a strategy through which it
may be able to transform its core capabilities and perhaps even change the terms
of competition in its industry.
How likely IBM and G M are to succeed in this transformation is far from
certain. IBM’s fortunes will depend on its success at reducing costs and increasing
its speed to market, on the strength of whatever Unix coalition emerges, on the
progress of IBM’s Japanese competitors, and on government regulation of tele-
communications in many countries around the world. But IBM has built on a
solid foundation: the communications networks that it has already created for
some of the largest companies in the world; its dominant, highly profitable main-
frame business; the prospect that many smaller companies, and even some govern-
ment agencies, will accept its dominance in global networks; and its efforts
through a vast range of strategic alliances to learn skills and capabilities that it did
not have and make them available to customers throughout the world.
G M faced other difficulties. It began creating a large constellation of knowl-
edge links at the same time as it was trying to transform its internal operations.
The company spent much of the 1980s implementing and refining the most mas-
sive reorganization in 50 years. During the decade, G M also acquired Hughes Air-
146 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

craft, the world’s leading defense electronics firm, and Electronic Data Services,
the world’s leading systems integration firm, in order to secure technology and
skills that it hoped would radically change its approach to auto making. G M paid
a steep price for these simultaneous changes. Costs rose at a time when the inten-
sively competitive U.S. market was limiting price increases. As a result, in 1988
and 1989, G M earned negligible profits on its North American auto operations.
Worse, its GM-10 cars-a new series of more stylish, higher-quality, mid-sized
cars-reached market late, well after Ford’s Taurus and Sable had made deep in-
roads in GM’s market position.
The difficulties encountered by IBM and G M in the late 1980s could easily
lead to skepticism about the value of knowledge links. This reaction, however,
would be wrong. Many of the most powerful and competitive Japanese compa-
nies, such as Toyota and Matsushita, rely heavily upon knowledge links with sup-
pliers and labor unions. During its renaissance in the 1980s, Ford Motor relied on
alliances in Japan and Korea, on much closer and more cooperative ties with sup-
pliers, and on a new partnership with workers and the UAW. Michael Porter’s re-
cent study of countries with internationally competitive industries concluded that
a nation’s successful firms are often linked together in “clusters.” Within these, an
assortment of mechanisms-various forms of knowledge links-promote the flow
of knowledge among a wide range of organizations.18
In the cases of G M and IBM, efforts to look more closely at individual alli-
ances and to trace their effects are riddled with difficulty. Many G M and IBM al-
liances are quite recent, so final judgments must wait several years. Moreover,
both companies are in the midst of what may prove to be historic transforma-
t i o n s - o r perhaps declines-and they are changing in response to many factors,
not just to alliances. Consider the cases in which productivity, absenteeism, and
quality measures indicate than G M has created a successful, collaborative rela-
tionship with the UAW-at Saturn, for example, or at the Chevrolet plant that
makes Corsicas and Berrettas. How much credit goes to what G M learned from
its alliance with Toyota, how much to internal efforts beginning with the Quality
of Work Life program in the early 1970s, and how much to the shift toward more
cooperative industrial relations underway in many American companies? To the
extent that G M has failed to change its labor relations quickly enough, how much
of its tardiness comes from not knowing how managers can help their organiza-
tion learn new capabilities from an alliance, and how much from the traditional
adversarial relationship between G M and its workers? Even ostensibly precise
quantitative measures fail to tell much of a story. For example, GM’s sales of cars
made with its Asian allies have fallen short of initial targets. Why? Because of the
alliances themselves, the wrong choice of partners, or the mid-1950s market shift
toward larger cars? Or was it because Pontiac dealers earned higher margins for
selling large cars like Bonnevilles and Grand Prix’s rather than the subcompact Le-
Mans’s made by the Daewoo alliance?
Moreover, the question of comparison makes assessments of alliances even
more difficult. Alliances are often criticized as unstable because many of them last
only several years. But compared to what? Are partnerships less stable, in general,
than organizational arrangements inside firms? In the 1980s, General Electric cre-
Knowledge Links 147

ated roughly 100 strategic alliances; some prospered, others failed, and many
needed redesign during their lives. B u t in the same period, General Electric reor-
ganized itself dramatically, reduced total employment by 100,000, and bought
and sold scores of businesses. G M and IBM both overhauled their internal opera-
tions and organization in the 1980s, and then made a multitude of corrections and
refinements. During this period of organizational earthquakes and aftershocks in-
side GE, GM, and IBM, were the core operations of the companies less turbulent
than their alliances?
Even if some of the alliances created by U.S. companies in recent years have
proved difficult to manage, it does not follow that companies should avoid them.
Historical comparison is also important. Early in this century, American firms
spent decades adopting, refining, and learning to manage the multidivisional form
of organization that Du Pont, Sears, GM, and a few other companies had in-
vented. In contrast, the U.S. experiment with alliances has been brief. From the
perspective of a twenty-first century historian, product and knowledge links may
represent only the initial phases of a decades-long effort to find new and more
flexible forms of organization suited to knowledge-driven, global markets.
While it is difficult to foresee and assess the ultimate effects of these changes,
one conclusion is quite firm: the pace and magnitude of the changes in GM’s and
IBM’s boundaries have been astonishing. Ultimately, the success of G M and IBM
will depend on many factors, and not simply on the contributions that knowledge
links can make. How much these alliances contribute to the two companies’ ef-
forts to secure embedded knowledge, extend their capabilities, and transform
themselves will depend on how well these alliances are managed and how quickly
these longtime citadels can learn from close relationships with outside organiza-
tions.

REFERENCES

1. Paul B. Carroll, “IBM Joins With Siemens AG,” The Wall Strcet]otirnal, January 2 5 ,
1990, p. B4.
2. Michael Polanyi, Personal Knowledge: Towards a Post-Critic01 Philosophy (Chicago:
University of Chicago Press, 1948), p. 53.
3. This description of Fanuc and G M F Robotics is bascd upon a n interview with Eric
Mittelstadt, president and C E O of G M F Robotics, November 1987; Fanuc Annual
Reports, 1983-1987; Kuni Sadamoto, Robots in the l a p a t m e Economy (Tokyo: Sur-
vey Japan, 1981); Centre for Business Research, The World Market for Itrditstrial Ro-
bots (Manchester, England: Centre for Business Research, 1986); Shinichi Kamata,
“GM-Fanuc Joint Venture Will Build Robot Factory in Michigan,” Japan Economic
Journul, M a y 3 , 1983, p. 10; “Next-Century Car Builder,” American Metal Mar-
ketlMetulworking News, February 3, 1986, pp. s8-slO; a n d Gene Bylinsky, “Japan’s
Robot King Wins Again,” Fortune, M a y 25, 1987, pp. 53-.58.
4. Ideally, a preemptive strategy secures several important advantages and not just one.
The move is difficult for competitors to copy; it exploits rivals’ weaknesses; it builds
upon the firm’s strengths; and the preempting firm can reverse its move, if necessary.
 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

Through its relationships, GM aimed to seize advantages across the full range of criti-
cal activities in the auto industry: supplier relations, product design, basic research,
manufacturing systems, customer relations, and distribution in service. An analytic
overview of preemptive strategies is Ian C. MacMillan, “Preemptive Strategies,” The
Journal o f Business Strategy (Fall 1985),pp. 16-26.
5 . Niccolo Machiavelli, The Prince, translated by George Bull (London: Penguin Books,
1981), p. 51.
6. An operating system is a collection of software that assists and in part controls a com-
puter’s basic operations. It coordinates the reading and writing of data between the in-
ternal memory and such peripheral devices as disk drives, keyboards, and printers; it
also prepares the computer to execute applications programs.
7. This overview of AT&T and its computer efforts is based on Charles P. Lecht, “The
Waves of Change,” Computer World, May 30, 1977, pp. 11-13; David A. Loehing,
“Ma Bell vs. IBM?,” Burron’s, February 9, 1976, pp. 3, 14, 16, 22; Michael A. Vere-
spej, “Clash of the Titans,” Industry Week, May 27,1985, pp. 64-68; Richard Brandt
and John W. Verity, “Unix: The Soul of a Lot of New Machines,” Businessweek,
March 14, 1988, pp. 94-96; Brenton R. Schlender, “AT&T, Sun, Looking to Open
Windows of Opportunity,” The Wall Street Journal, April 11, 1988, p. 6; and
Jonathan B. Levin, “Will Sun Melt the Software Barrier?,” Businessweek, April 18,
1988, p. 72.

IBM Japan executives and on “”

8. This account of IBM Japan’s efforts in telecommunications is based on interviews with
Signs Pacts with IBM,” Electronic News, March
4, 1985, p. 40; “Mitsubishi’s New Communication Adviser Lays Plans for Data Serv-
ices,” Data Communications, March 1984, pp. 115-116; and Carla Rapoport, “NTT,
IBM Japan Agree Software Systems Venture,” The Financial Times, September 26,
1985, p. 4.
9. Howard Anderson, “The IBM-MCI Merger: An Evaluation,” Telecommunication
Products and Technology (August 1985), pp. 21-25; and Frances Seghers, “NOWIf
MCI Can Just Keep the Party Going,” Businessweek, May 16, 1988, p. 39.
10. This overview of IBM’s efforts in software and systems integration is based on inter-
views with IBM executives; “Computer Industry Trend: System Integration,” Nikkei
Computer, April 1, 1988, pp. 42-58; Edith Myers, “Risk and Responsibility,” Com-
puter World, November 18, 1987, pp. 4 3 4 5 ; Paul Gaskell, “Alien Connection,” Sys-
tems International (September 1987), pp, 35-37; “IBM, Mitsubishi Bank in Software
Effort,” Electronic News, October 19, 1987, pp. 12, 20; Peter Freeman, “Software
Development Systems,” Computers and People (September-October 1987), pp. 11-
27; and “Nissan and IBM Japan Agree to Jointly Set Up New Company,” press re-
lease, IBM Japan, January 14, 1987.
11. For an extended and critical assessment of IBM’s long-term prospects in telecommuni-
cations, see Richard T. DeLamarter, Big Blue: IBM’s Use and Abuse of Power (New
York: Dodd Mead, 1986),Chapter 5.
12. The overview of Saturn’s technology and systems is based on an interview with Rich-
ard LeFauve, President, Saturn Corporation, October 5, 1988; personal communica-
tions with G M managers John Barry DuVall and William McPherson; “Till the Magic
Goes,” Manufacturing Systems (September 1986), pp. 32-36; Bryan H. Berry, “It’s
Now or Never for World-Class Auto Making at GM,” Iron Age, November 7, 1986,
pp. 34A1-35; Dale D. Buss and Melinda G. Guiles, “GM Slows Big Drive for Saturn
Knowledge Links 149

to Produce Small Car in Five Years,” The Wall Streetlournal, October 30, 1986, p. 1;
and “Saturn: It’s on Target for ‘90 Introduction,” Automotive News, April 25, 1988,
pp. 1, 64.
13. Roger B. Smith, “A New Age of Almost Cosmic Industrial Achievement,” The journal
ofBusiness Strategy (March 1984), p. 80.
14. Anne B. Fisher, “Behind the Hype at GM’s Saturn,” Fortune, November 11, 1985, p.
44 *
15. David Whiteside, “How GM’s Saturn Could Run Rings Around Old-Style Car Mak-
ers,” Businessweek, January 28, 1985, p. 126.
16. “Developments in Industrial Relations,” Monthly Labor Review (October 1985), p.
49.
17. Stewart Brand, The Media Lab (New York: Viking, 1987), pp. 1-16.
18, See Michael E. Porter, The Competitive Advantage ofNations (New York: Free Press,
1990), pp. 152-165.
This page intentionally left blank
 -

Organizational Learning,
and Competitive Advantage:
The HRM Agenda
Vladimir Pucik

STRATEGIC ALLIANCES AND COMPETITIVE

COLLABORATION
Partnerships and alliances between two or more multinational firms are be-
coming increasingly common. Recent examples include AT&T’s cooperation with
Olivetti and IBM’s links with Matsushita in office automation equipment; a tri-
partite venture of Honeywell, Bull, and NEC in computer mainframes; Philips and
AT&T’s alliance in telecommunications; Toyota and General Motors’ joint manu-
facturing at NUMMI; or General Electric and Fanuc’s worldwide collaborative
network in robotics. New strategic alliances are not limited to the manufacturing
sector, they are increasingly frequent in the financial sector (e.g., the joint venture
of Credit Suisse and First Boston Corporation or the tie-up of Nippon Life and
Shearson Lehman) and other service industries as well.
Some of the new alliances are clearly short-term in nature (e.g., General
Motors and Toyota); others aim for a long-term strategic synergy between the
partners (e.g., AT&T and Olivetti). While international alliances are intrinsically
difficult to manage (Killing, 1983), many experts argue that, as business risks soar
and competition grows more severe, firms are expected to rely on such alliances
with increasing frequency (Harrigan, 1985; Perlmutter and Heenan, 1986). The
rationale and the scope of international alliances are becoming increasingly com-
plex (Contractor and Lorange, 1988; Root, 1988).
In the past, alliances were seen primarily as a means to reduce capital invest-
ment and lower the risks associated with entry into new markets. Ties between
firms were also formed in order to secure fast and reliable access to previously

“Strategic Alliances, Organizational Learning, and Competitive Advantage: The H R M Agenda,”

Vladirnir Pucik, Hirman Resource Management, Vol. 27, No. 1: pp. 77-93. Copyright 0 1988 by
John Wiley & Sons, Inc. Reprinted by permission of John Wiley & Sons, Inc.
151
152 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

closed markets, or to respond to a government’s preference (formal or informal)

for local participation in the business (Leontiades, 1985). Today, the rationale be-
hind the formation of new alliances in most cases is related to the increasing speed
of technological change and the rapidly growing competitiveness in global mar-
kets. Partners join in order to diversify risks inherent in developing new technolo-
gies or to take advantage of the complementarity of each partner’s developmental
skills (Hergert and Morris, 1988).The new partnerships can also provide essential
economies of scale and market power to withstand a dominant competitor whom
neither partner can challenge individually (e.g., international alliances in the com-
puter industry targeted at IBM).
Strategic alliances can take many forms: technical exchange and cross-li-
censing, co-production and OEM agreements, sale and distribution ties, joint
product development programs, or creation of joint venture firms with equity dis-
tributed among the partners. Such a functional classification of alliances, how-
ever, does not say much about their competitive context. To understand the
strategic logic of the new partnerships and the implications for human resource
management, it is essential to consider the changing patterns of global competi-
tion. In contrast to traditional single-market joint ventures between large multina-
tional firms and much smaller local firms, the new alliances are often formed by
partners of comparable strength whose activities are often global and who are or
may become direct competitors (Contractor and Lorange, 1988).
The rapid increase in international partnerships among competitors does
not necessarily imply the heralded dawn of a new cooperative era in the global
economy (Ohmae, 1985; Perlmutter and Heenan, 1986). The change from com-
petitive to collaborative strategies is often merely a tactical adjustment aimed at
specific market conditions. Many of these new partnerships should be viewed as a
hidden substitute of market competition, not its dissipation. The objective is simi-
lar: attaining the position of global market leadership through internalization of
key value-added competencies. The potential competitive relationship between
partners distinguishes strategic alliances that involve competitive colluborution
from more traditional complementary ventures (Doz et al., 1986).
The strategic and managerial implications of the two types of alliances are
fundamentally different. In a truly cooperative relationship the underlying as-
sumption is the feasibility (and desirability) of long-term widwin outcomes. In
the partnerships that involve competitive collaboration, the strategic intent of
achieving dominance makes the long-term widwin outcome highly unlikely. This
does not imply that all partnerships between multinational firms are always com-
petitive in nature. However, many of them are, especially when seen in a long-
term dynamic context. Partnerships that involve competitive collaboration are
dynamic in nature. The relative endowment of resources, skills, and competencies
and the sources of bargaining power can change over time. For one firm to be able
to sustain its long-term competitive advantage, the organization and control of
the partnership has to reflect its competitive context.
Another way to look at strategic partnerships is to examine the source of
leverage exercised by the individual partners. In this sense, strategic alliances may
be classified into those leveraging resources and those leveraging competencies.
The H R M Agenda 153

The cross-licensing, technical agreements, joint development programs (pooling

of resources), and co-production or co-distribution (resource economies of scale)
are examples of alliances that focus primarily on resource leverage. The resources
contributed to a partnership usually have a specific market value, be it land,
equipment, labor, money, or patents. Both the contribution and withdrawal of re-
sources are explicit and thus relatively simple to control.
In contrast, competencies are fundamentally information-based invisible as-
sets (Itami, 1987) that cannot be readily purchased and their market value is diffi-
cult to ascertain. Examples are management and organizational skills, knowledge
of the market, or technological capability. Invisible assets are embodied in people
within the organization. These assets represent a tacit knowledge that is difficult
to understand and that only can be appropriated over time, if at all (Teece, 1987).
Accumulation of invisible assets is seen as the foundation for a sustainable com-
petitive advantage (Itami, 1987).
Invisible assets are closely linked to information, its stock as well as its flow.
To increase invisible assets is to increase the amount of information available in
the firm as well as its capacity to handle the information. Invisible assets can be
accumulated through an explicit action, such as training, or implicitly as a by-
product of daily operations (Itami, 1987). Alliances that leverage competencies
usually take the form of an OEM supply agreement or a joint venture aimed at a
specific market. Superior competencies in different parts of a value chain are com-
bined to achieve a distinct competitive advantage in the market, or a t least to pro-
tect market position against a superior joint competitor.
Table 9-1 summarizes a classification of international partnerships that con-
siders the competitive context and the source of leverage. While the issue of stra-
tegic control as related to the distribution of benefits from the alliances is
important in all four quadrants of the strategic alliance matrix, it is especially
critical in quadrants 111 and IV that represent conditions of competitive collabora-
tion. Implicit in a competitive collaboration is the risk that benefits from the alli-
ance may be accrued asymmetrically by the respective partners. The process of
appropriation is influenced by the characteristics of organizational assets lever-
aged in the partnership.
Obviously, in most cases, strategic alliances involve the contribution and lev-
erage of both visible and invisible assets. The type of contribution can be different
for each partner. Nevertheless, the traditional management focus is concentrated

TABLE 9-1 Strategic Alliances, Strategic Context,

and Source of Leverage
Source of Leverage

Strategic Context Resources Competencies

Complementarity I. 11.
Competition 111. IV.
154 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

on the control of visible assets. In complementary partnerships, lack of attention

to the accumulation of invisible assets may erode the competitive advantage de-
rived from the venture. However, in the context of competitive collaboration,
where competencies provide the critical leverage (quadrant IV), lack of attention
to invisible assets may result in a loss of control over the direction of the alliance.
It is for this reason that management processes that support accumulation and
control of invisible assets are of such critical importance.
The distribution of benefits related to visible assets, such as new products or
profits, is relatively easy to monitor. Protection against asymmetry can be insti-
tuted through administrative protocols and rules regarding the implementation of
the partnership agreement. However, the asymmetric appropriation of invisible
benefits-such as the acquisition of product or market know-how for use outside
of the partnership framework, or even to support a competitive strategy targeted
at the partner-cannot be easily protected. The asymmetry results from the inter-
nal dynamics of the strategic alliance. Benefits are appropriated asymmetrically
due to differences in the organizational learning capacity of the partners. The
shifts in relative power in a competitive partnership are related to the speed a t
which the partners can learn from each other. Not providing a firm strategy for
the control of invisible assets in the partnership, and delegating responsibility for
them to operating managers concerned with short-term results, is a sure formula
for failure.
A good illustration of such a process is the reversal in the competitive rela-
tionship between Japanese and Western firms in many industries over the last sev-
eral decades. The asymmetrical distribution of benefits from these alliances often
was the fundamental cause of such a reversal. Japanese firms used access to tech-
nology through licensing or joint ventures to master new competencies, and then
used the newly acquired knowledge to gain sole control of the market in Japan
and even penetrate markets previously dominated by the Western partners with
their own superior products. The list of firms (e.g., Allied/Bendix, General Elec-
tric, General Foods, International Harvester, Philips, Renault, USX, Westing-
house) that gave up more than they gained is long and is not limited to a single
country or industry.
While other factors contributed to the high failure rate of Western joint ven-
tures in Japan (Wright, 1979; Zimmerman, 1985)-such as policies of the Japa-
nese government that a t least until the mid-1970s made it difficult for Western
partners to achieve bargaining power parity-much of the imbalance in the ap-
propriation of benefits was caused by disparities in learning between Japanese and
Western partners. Many Japanese firms have developed a systematic approach to
organizational learning (Cole, 1985; Nonaka and Johansson, 1985). This ap-
proach involved more than an explicit rejection of the parochial “not-invented-
here” syndrome. Japanese firms put in place managerial systems that encourage
extensive horizontal and vertical information flow and support the transfer of
know-how from the partnership to the rest of the organization. The policies guid-
ing the management of human resources at all levels and functions constituted a
vital part of such a learning infrastructure (Pucik, 1983).
The H R M Agenda 155

The organizational capability to learn is the key to protect competitive ad-

vantage in competitive collaboration and to control the strategic direction of the
cooperative venture. An organization has many tools to manage the process of
learning (Hedberg, 1981), but in principle, the learning ability of an organization
depends on its ability to accumulate invisible assets. As invisible assets are embod-
ied in people, policies regarding human resources are critical to organizational
learning. The objective of the HRM activities is to complement line management
in providing a supporting climate and appropriate systems to guide the process of
learning. Organizational learning results from a combination of hard and soft or-
ganizational practices anchored in specific H R M techniques.

HUMAN RESOURCE MANAGEMENT A N D OBSTACLES

TO ORGANIZATIONAL LEARNING
Organizational learning is not a random process. Preventing an asymmetry
(or creating an asymmetry in one’s favor) in organizational learning must be the
key strategic priority for human resources executives in multinational firms en-
gaged in strategic alliances. Removing the organizational obstacles to learning is
closely linked to the strategic priorities of the Human Resource function and its
involvement in the design and management of the strategic partnership. However,
this strategic priority is often buried under the pressure of daily operational con-
cerns. The key obstacles to organizational learning identified from research on
Western joint ventures in Japan (Pucik, 1988) are listed in Table 9-2.
The obstacles to organizational learning reviewed in greater detail below are
not limited to a specific organizational climate that can easily be changed. Rather,
they result from a complex set of HR practices and policies that, while often ra-
tional in the short-term, may ultimately lead to a loss of control over the destiny
of the partnership, if not to the loss of the entire business. Understanding the ob-
stacles to learning is the first step in the process of restoring competitive balance.

Human Resource Planning

Strategic intent not communicated throughout the firm. Most alliances take
place in a highly complex competitive environment. The desirability of
cooperation may easily be perceived differently among various parts of
the organization, depending on their level of involvement in the creation
of the alliance and their responsibility in executing the strategy. Top man-
agement often emphasizes the cooperative nature of the new alliance,
partly to set the right tone for the partnership, partly to break down any
resistance from those opposed to the cooperative strategy. What is often
not made clear are the boundaries of cooperation and the specific nature
of the missing competencies that led to the alliance in the first place.
156 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

TABLE 9-2 Obstacles to Organizational Learning in International Strategic Alliances

HR Function K e y Obstacles
H R planning Strategic intent not communicated
Short-term and static planning horizon
Low priority of learning activities
Lack of involvement by the H R function

Staffing Insufficient lead-time for staffing decisions

Resource-poor staffing strategy
Low quality of staff assigned to the alliance
Staffing dependence on the partner

Training and development Lack of cross-cultural competence

Uni-directional training programs
Career structure not conducive to learning
Poor climate for transfer of learning

Appraisal and rewards Appraisal focused on short-term goals

No encouragement of learning
Limited incentives for transfer of know-how
Rewards not tied to global strategy

Organizational design and

control . Responsibility for learning not clear
Fragmentation of the learning process
Control over the HR function given away
N o insight into partner’s HR strategy

Short-term and static planning horizon. Planning of the alliance is often

driven by short-term contingencies, such as an improvement of profitabil-
ity by cutting production costs through an OEM arrangement, without
considering long-term effects on the sustainability of the firm’s competi-
tive advantage. General Electric’s recent withdrawal from the consumer
electronics field was forced by a series of “correct” short-term decisions
during the previous two decades that led to a transfer of critical product
and process competencies from GE to its competitors. The logic behind
many short-term decisions assumes that the existing balance of compe-
tencies in the alliance will not change with time.
Low priority given to learning activities. The traditional focus of business
plans is on the utilization of and the return on tangible assets. The pro-
jected outcomes from the partnership are scrutinized in terms of returns
on equity invested, savings from pooled research and development, cost
reductions from outsourcing components and products, and/or increases
in sales from added distribution channels. However, the accumulation of
The H R M Agendn 157

invisible assets, such as experience regarding the production process, inti-

mate knowledge of the market or relationship with customers, is not
evaluated, as traditional planning systems cannot assign a financial value
to these outcomes. Activities that can’t be evaluated in financial terms are
generally not funded. Organizational learning is left with no support.
Lack of involvement by the Human Resource function. In the rush to
launch the alliance, insufficient attention is given to a critical evaluation
of the learning capacity of the organization and to the steps necessary to
upgrade the learning skills and learning climate appropriate for the new
venture. Often, the Human Resource function does not play any role in
the negotiation process or becomes involved only at a very late stage. The
compatibility of philosophies regarding the management of human re-
sources between the partners and its implications for organizational
learning are seldom a factor in the decision-making process.
Insufficient lead-time for staffing decisions. When the alliance involves a
creation of a new organization, staffing decisions regarding the key repre-
sentatives should be made well in advance of the conclusion of the agree-
ment; all relevant future players can thus be involved in the negotiation
process. Institutional memory breaks down when negotiators are re-
placed by implementators without continuity. Insufficient lead time also
forces short-cuts in training for the managers to be assigned to the part-
nership. In general, everyone agrees with the idea of training, but many
firms are reluctant to invest in the preparation of managers for the new
venture until the outcome of the business negotiations is clear; yet after
the deal is signed, there is no time to train. As a result, what is won labo-
riously at the front end through long, arduous bargaining is often lost
through the inability to control implementation of the partnership agree-
ment.
Resource-poor staffing strategy. As the motivation for the alliance is often
driven by cost consideration, firms cut expenses by limiting the size of
managerial staff assigned to the partnership. In particular, this can be
observed in alliances that have the major location of their activities over-
seas, where the cost of expatriates seem prohibitive. Yet, while the ex-
pense of staffing a position in an overseas venture can be substantial, such
economizing does not consider two substantial benefits derived from ex-
patriate posts: improved control over the management process in the ven-
ture and ability to transfer skills from the venture into the home
organization. Organizational learning often requires at least some slack
resources. When an overextended management team just keeps on dous-
ing fires, the last thing on a manager’s mind is the transfer of know-how.
Low quality of staff assigned to manage the alliance. It is often the case that
after the initial period of high visibility for the new alliance, management
positions in the partnership become a dumping ground for sidetracked
executives. The emphasis is on “making the deal,” not on its implementa-
tion. The dispatched managers don’t have the necessary learning skills;
158 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

they are expected to “watch the books” only. Even if they gain new
knowledge, they may lack the credibility to effectively transfer the know-
how to the parent firm, especially if this involves challenging existing “sa-
cred cows.’’ The partners in the alliance are generally well aware of the
low skill and credibility level of these managers and d o not hesitate to
freeze them out of the important decisions.
Staffing dependence on the partner. When staffing is considered a cost
rather than an investment, it is very tempting to go along with the offer
by the partner to assume the responsibility for staffing the new venture.
Naturally, there is always a great concern over the composition of the top
management team. However, very little learning ever occurs in the board
room: learning takes place in the laboratories, on the production floor,
and in interactions with the customers. The partner who controls posi-
tions critical to the accumulation of invisible assets gains substantial lev-
erage over the direction of the alliance. Short-term excursions will not do,
long-term participation is essential. As G M learned at NUMMI, a vide-
otape of new work practices is a far less efficient learning tool than
hands-on experience.

Training and Development

Lack of cross-cultural competence. Many managers and staff involved in

international partnerships d o not have sufficient intercultural skills (lan-
guage competence, familiarity with partner’s culture, etc.). Expatriates
are dispatched abroad with no or limited training at best, with the as-
sumption that knowledge of the business should compensate for the lack
of cultural understanding. While the partner’s perfect fluency may not be
essential, the ability to understand the basic flow of a business conversa-
tion and to interact informally with the customers and employees should
be the minimum prerequisite for an international assignment. This is im-
portant for an expatriate’s effectiveness, even in a wholly-owned foreign
affiliate (Tung, 1984); the price to pay for the lack of cross-cultural skills
in an alliance may be higher: both inability to learn and inability to con-
trol.
Uni-directional transfer of know-how. One of the most effective means of
learning is through temporary personnel exchange between the partners.
However, this exchange is often asymmetrical, especially when the part-
nership takes the form of a joint venture. While the flow of personnel
from the Western joint ventures in Japan often includes staff temporarily
seconded from the Japanese parent, the training assignments in the oppo-
site direction are infrequent (Pucik, 1988). Even when transfer of person-
nel into the joint venture occurs on a regular basis, it is seldom for the
purpose of skill acquisition. Rather, staff is transferred either to control
or manage the joint venture or to serve as a conduit for transferring
know-how into the venture. It is often felt that there is no need to learn
The H R M Agenda 159

(and thus expend resources) 011 knowledge already possessed in the joint
venture. Yet, by gaining independent know-how, n firm can avoid becom-
ing hostage to the uncertain future of the partnership.
Career structure not conducive to learning. Personnel exchange can have a
positive impact on the amount of accumulated knowledge only if admin-
istered in a consistent and planned fashion over a period of time. Unless
the firm posts the returnees from the partnership ventures into positions
where the acquired know-how can be effectively used and disseminated,
the invisible asset accumulation will not be possible. The amount of time
spent learning and transferring know-how is the critical constraint. An ef-
fective transfer of know-how requires a long-term commitment of quali-
fied personnel, which clashes with expectations of fast mobility among
the most promising executives. While many managers (on a personal ba-
sis) may benefit even from a relatively short assignment abroad, a single
short-term assignment-especially when it comes relatively late in an ex-
ecutive’s career-will not do much for the accumulation of invisible assets
in the rest of the organization.
Poor climate for transfer of learning. A large amount of critical invisible as-
sets is embedded in the staff involved in the partnership. To what degree
these assets are shared with the parent depends largely on the parent’s re-
ceptivity to new ideas, and on the quality of the interaction between the
cooperative venture and the parent firm. When learning from the outside,
in particular from abroad, is seen as an admission of weakness, the recep-
tivity will be poor (Westney, 1988). The ossification of the learning infra-
structure reflects the low priority given to the accumulation of invisible
assets in the execution of a company’s strategy. Low receptivity to inputs
from the partnership will naturally encourage a passive attitude towards
the transfer of knowledge among the partnership staff. This tendency is
further reinforced if the socialization activities in the partnership are con-
trolled by the local parent, as is often the case in Western joint ventures in
Japan.

Appraisal and Rewards

Appraisal focused on short-term goals. Organizational learning is fun-

damentally a long-term activity, stretching far beyond a typical one-year
appraisal time-frame. Also, the costs associated with learning are im-
mediate, while the benefits (most of them difficult to quantify under
standard accounting procedures) are accrued over time. Support for or-
ganizational learning thus may have a negative impact on the short-term
measurements used to evaluate a manager’s performance. The expecta-
tion of short tenure in a given job is another critical constraint. The pres-
sure to get immediate results forces managers to economize on
expenditures with long-term payoffs, no matter how attractive such pay-
offs may be. The issue is not sacrificing profits for abstract learning, but
160 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

forfeiting a long-term superior performance in order to inflate short-term

results.
N o encouragement of learning. With little or no rewards given for contri-
butions to the accumulation of invisible assets, learning becomes a
“hobby,” not a prerequisite of the job. In many leading Japanese and Ko-
rean firms, the cross-fertilization of skills across functional areas is ac-
tively encouraged, and both foreign language ability (tested by the
company) and familiarity with principal foreign markets are considered
before promotion to an executive position. In contrast, the skill base of
typical Western managers is rather narrow, as are their intercultural skills.
Even in firms with decades of experience in the Far East, only a handful
of managers speak any of the local languages and have a first-hand
knowledge of local conditions. In a joint venture, asymmetry in the distri-
bution of skills will result in an erosion of competitive advantage and the
loss of leverage.
Limited incentives for transfer of know-how. The reward systems in many
multinational firms encourage hoarding of critical information, not shar-
ing it. Information is treated as a source of power, not as a resource.
Smart managers assigned to an international joint venture, who other-
wise may expect few opportunities for upward mobility, can make them-
selves indispensable by blocking the flow of information. Such a behavior
is not only tolerated, but these “valuable experts” are often rewarded in
terms of superior compensation and considerable operational autonomy.
Any increase in information concerning the activities of the partnership
outside of their own domain is seen by these managers as a threat to their
power. In an alliance that involves competitive collaboration, the other
parent and some of the company’s own managers may share an interest in
limiting the transfer of know-how.
Rewards not tied to global strategy. Performance of executives assigned to
manage a partnership venture is often appraised solely on the basis of re-
sults in a limited business area or market. There is very little incentive for
the “core” partnership staff to worry about the competitive conditions
facing other businesses of the distant parent. These managers have noth-
ing to gain from allocating scarce resources to organizational learning
benefiting an organization in which they have no tangible interest. This
tendency is especially pronounced if these managers are actually dis-
patched from the other “competing” parent. In such a case, their attitude
towards transfer of competencies can easily turn from conservative to
downright hostile.

Organizational Design and Control

Responsibility for learning not clear. Who gains and who loses from a stra-
tegic alliance often depends on the vantage point. A “win-win” partner-
ship strategy on a corporate level often entails a “win/lose” scenario a t
The H R M Agenda 161

the business unit or business function level. For example, a shift from
captive manufacturing to an OEM partnership may contribute to imme-
diate cost reduction and thus enhance the product’s position in the mar-
ket while the production competence is eroded. Under such conditions,
incentives and responsibility for learning may become unfocused. When
competencies are lost, operation managers blame faulty strategy while the
corporate staff cites incompetent implementation.
Fragmentation of the learning process. In diversified, complex firms, the
stakes in organizational learning may differ by business unit and func-
tion. Each subunit has only a partial view of the exchange of competen-
cies involved in the partnership. The perceptions of the potential value of
the relationship may therefore differ, as will the commitment to support
competencies needed to defend the long-term competitive advantage. In
firms with decentralized business units (e.g., SBUs), organization-wide
learning activities have low priority in comparison to a business unit’s im-
mediate needs.
Control ofthe HR function is given away. The HR function is seen as a cost
burden, not as a powerful tool of control over the strategic direction of
the partnership. In particular, when the alliance involves a venture inside
the new partner’s territory, responsibility for the Human Resource func-
tion is often delegated to the partner. In fact, the very possibility of utili-
zation of the partner’s know-how concerning the local labor market
conditions is often a factor leading to the creation of the alliance in the
first place. However, what is gained in lowering the cost of entry may be
lost over time, as control over human resource deployment enables the
partner to control the patterns of organizational learning, thus the distri-
bution of benefits from the partnership.
N o insight into partners H R strategy The learning strategies of the partner
can be monitored through the control of personnel exchange between the
joint operations and the parent. The objective is not to stop learning, but
to gain understanding about the direction of the partner’s learning strate-
gies and its long-term impact on the balance of power in the collaborative
relationship. However, when personnel control is abdicated in favor of
the partner, the logic of the learning process is obscured. The boundary
between the partner’s organization and the partnership operation be-
comes fuzzy and impossible to control. Valuable competence may leak
without notice and without reciprocity. A learning asymmetry is again
likely to occur.

HRM AGENDA IN COMPETITIVE COLLABORATION

The challenge of competitive collaboration creates a new agenda and new

priorities for the management of human resources. This challenge can’t be
avoided by staying away from strategic alliances. The economic forces in the envi-
ronment will continue to push firms into more complex sets of global relation-
162 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

ships. Those who learn from these relationships will survive; the others will per-
ish. The organization’s ability to learn (or the lack of it) will influence the shape of
the global markets for many years to come.
Experience shows that the competitive balance in strategic alliances, and in
joint ventures in particular, cannot be controlled through structural solutions. The
successes and failures of the alliances are often embedded in the same organiza-
tional context (Killing, 1983). Neither can symmetry in the appropriation of bene-
fits from a partnership be protected through legal clauses. The complexity of
international commercial law and rapid technological change make legal protec-
tion impractical. In fact, the reliance on legal means to safeguard the company in-
terests can be counterproductive as it encourages “we-are-safe” attitudes and thus
decreases the stimuli to learn.
The accumulation of invisible assets, be it manufacturing competence, mar-
ket know-how, or global coordination capability, should be explicitly recognized
as a value-enhancing activity. It is dangerous to act as if the existence of a partner-
ship permits lowering commitment to the maintenance and expansion of core
competencies. Such a strategy assumes that the partner is unwilling or unable to
learn and thus unable to alter the long-term bargaining power regarding the ap-
propriation of benefits. In the context of competitive collaboration, such an as-
sumption is unsupportable. It also does not make sense to set up barriers to
learning. Artificial constraints imposed on information flow in the partnership
may hinder its ability to sustain its competitive advantage and thus erode the
competitive position of both parents. The only sustainable response is a pro-ac-
tive policy encouraging organizational learning that, a t minimum, matches, if
not surpasses the learning ability of the partner. Everything else is an inferior
solution.
A number of specific agenda points for the HR function in firms engaged in
international strategic alliances can be drawn from the experience of firms that
continuously incorporate organizational learning into their competitive strategy:

1. Get involved early, The human resource function should be involved in

the formation of the strategic alliance from the early planning stages. In a
dialogue with the appropriate line functions, HR staff should assume re-
sponsibility for the development of a thorough organizational learning
strategy. It is essential to precisely identify the critical value-added learn-
ing activities in a given business, and the means to control them. The ob-
jective is to support and expand core competencies essential to sustain
the long-term competitive advantage of the firm.
2. Build learning into the partnership agreement. In order to maintain a
long-term symmetry in the distribution of benefits from the partnership,
both parties have to learn simultaneously. The process of parallel learn-
ing can and should be made explicit. An attempt to prevent the partner
from learning is most likely fruitless, as organizational learning is impos-
sible to police. Instead, provisions should be made in the agreement to
safeguard the reciprocity in the transfer of competencies (e.g., personnel
exchange).
The H K M Agenda 163

3. Communicate strategic intent. As a part of its responsibility for corpo-

rate communications, HR should cooperate with operational managers
to assure that the strategic intent with respect to the partnership is ade-
quately communicated to the employees. Training programs should be
developed to prepare managers to deal effectively with the ambiguity and
complexity of strategic alliances. The competitive context has to be made
explicit: hushing it up does not fool anybody but your own employees.
4. Maintain HR input into the partnership. The control of the H R function
in partnership operations, such as joint ventures, should not be bargained
away, as it is within the boundaries of such an entity that much of the
learning occurs. Once the partnership agreement is concluded, the HR
function should continuously monitor the congruence between the learn-
ing strategy and the operational HR activities related to the launching
and the implementation of the agreement. Periodic reviews of the learn-
ing process should be set up with the participation of top management.
5. Stuff to learn. The accumulation of invisible assets should be the key
principle guiding the staffing strategy. Staffing and development plans
should be established to cover the existing blind spots. Such an approach
may require a considerable investment in the development of core compe-
tencies within the parent firm through a carefully calibrated transfer pol-
icy. Some attrition must be considered inevitable. In joint ventures, this
also means the development of a local staff that is fundamentally loyal to
the joint venture entity and has no vested interest in blocking the transfer
of critical information. While the immediate costs of the “staff to learn”
program may be high, they are far smaller than the long-term negative
consequences of lost competence.
6. Set u p learning-driven career plans. From an individual perspective, ef-
fective learning and transfer of competencies span the entire career. While
cross-cultural learning is most effective during the early career stages,
functional learning and its effective application may require considerable
business experience. In the context of international partnerships, this
may imply a necessity for multiple assignments, which is seldom done at
the present time. A greater use should be made of reciprocal trainee pro-
grams. The notion that all expatriates should be managers is obsolete.
7. Use training to stimulate the learning process. Three kinds of training
activities can create a better climate for learning. First, in internal train-
ing, managers should be made aware of the subtleties involved in manag-
ing collaboration and competition at the same time. Second, open
communication and trust within the partnership is essential for the
smooth transfer of know-how. Team-building and cross-cultural commu-
nication training should be offered regularly at all management levels. Fi-
nally, any training program geared to the acquisition of a specific
competence should be, in principle, reciprocal. This diminishes the incen-
tives for opportunistic behavior.
8. Responsibility for learning should be specified. In order to create a cli-
mate receptive to learning, a specific responsibility for learning should be
164 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

written into business plans for managers transferred into the partnership
operations as well as those in the receiving units. It should be made clear
who is responsible that the information actually flows as intended, in
necessary quality and speed, and what supporting mechanisms are
needed to be put in place. Where appropriate, support for mutual learn-
ing should be made explicit in the partnership agreement.
9. Reward learning activities. Management behavior that encourages or-
ganizational learning, such as sharing and diffusion of critical informa-
tion, should be explicitly recognized and rewarded. Long-term incentives
(e.g., career opportunities) should be provided to managers actively seek-
ing to acquire new skills. The framework of expatriate transfers into
critical locations must be restructured to make them more attractive
without incurring prohibitive compensation costs. Dead-end assignments
are costly to the organization.
10. Monitor the H R practices of your partner. Throughout the duration of
the relationship, attention should be given to the partner’s H R activities.
Beginning with an H R audit prior to the establishment of the partner-
ship, much insight can be gained from the continuous monitoring of the
partner’s staffing and training. In joint ventures, the career records of
staff transferred from the partner’s organization should be carefully scru-
tinized, including their assignments after returning to the partner. It must
be assumed that the partner is doing the same, as much of the necessary
information is actually in the public domain.

In summary, the strategic agenda for the H R M function in firms involved in

international alliances must be centered around the process of learning. In the
context of competitive collaboration, the competitive advantage of a firm can be
protected only through the organization’s capability to accumulate invisible assets
by a carefully planned and executed process of organizational learning. As this
process is embedded in people, many of the necessary capabilities are closely
linked to H R M strategies and practice. The transformation of the H R system to
support the process of organizational learning is clearly the key strategic task fac-
ing the H R function in many multinational firms today.

REFERENCES
Cole, R. E. The macropolitics of organizational change: A comparative analysis of the
spread of small-group activities. Administrative Science Quarterly, 1985, 30, 560-585.
Contractor, E, and Lorange, P. Why should firms cooperate? The strategy and economics
basis for cooperative ventures, in F. Contractor and P. Lorange (Eds.), Cooperative
strategies in international business. Lexington, MA: Lexington Books, 1988.
DOZ,Y.,Hamel G., and Prahalad, C. K. Strategic Partnerships: Success or surrender?, pa-
per presented a t the Conference on Cooperative Strategies in International Business, The
Wharton School and Rutgers University, October 1986.
The H R M Agenda 165

Harrigan, K. R. Strategies for joint ventures. Lexington, MA: Lexington Books, 1985.
Hedberg, B. How organizations learn and unlearn, in P. C. Nystrom and W. H. Starbuck
(Eds.), Handbook o f organizational design. New York: Oxford University Press, 1981.
Hergert, M., and Morris, D. Trends in international collaborative agreements, in F. Con-
tractor and P. Lorange (Eds.), Cooperative strategies in international business. Lex-
ington, MA: Lexington Books, 1988.
Itami, H. Mobilizing invisible assets. Cambridge, MA: Harvard University Press, 1987.
Killing, J. P. Strategies for joint venture success. New York: Praeger Publishers, 1983.
Leontiades, J. C. Multinational corporate strategy: Planning for world markets. Lexington,
MA: Lexington Books, 1985.
Nonaka, I., and Johansson, J. K. Organizational learning in Japanese companies, in R. B.
Lamb (Ed.), Advances in strategic management. Greenwich, CT:JAI Press, 1985, 3,
277-2 9 6I

Ohmae, K. Triad power: The coming shape of global competition, New York: Free Press,
1985.
Perlrnutter, H. V., and Heenan, D. A. Cooperate to compete globally. Harvard Business Re-
view. 1986, 64 (2),136-152.
Pucik, V. Management practices in Japan and their impact on business strategy, in R. B.
Lamb (Ed.), Advances in strategic management. Greenwich, CT: JAI Press, 1983, 1,
103-1 31.
Pucik, V. Joint ventures with the Japanese: Implications for human resource management,
in E Contractor and P. Lorange (Eds.), Cooperative strategies in international business.
Lexington, MA: Lexington Books, 1988.
Root, E Some taxonomies of international cooperative agreements, in F, Contractor and P
Lorange (Eds.), Cooperative strategies in international business. Lexington, MA: Lex-
ington Books, 1988.
Teece, D. J. Profiting from technological innovation: Implications for integration, collabo-
ration, licensing and public policy, in D. J. Teece (Ed.), The competitive challenge: Strate-
gies for industrial innovation and renewal. Cambridge, MA: Ballinger Publishing
Company, 1987.
Tung, R. K e y to lapan’s economic strength: Human power. Lexington, MA: Lexington
Books, D.C. Heath, 1984.
Westney, D. E. Domestic and foreign learning curves in managing international cooperative
strategies, in E Contractor and P Lorange (Eds.), Cooperative strategies in international
business. Lexington, MA: Lexington Books, 1988.
Wright, R. W. Joint venture problems in Japan. Colrtmbia lourml of World Business.
1979,20 (l),25-31.
Zimmerman, M. How to do business with theJapanese. New York: Random House, 1985.
This page intentionally left blank
 10
The Numbers
Charles Handy

The numbers are the numbers of people, the numbers working, numbers dying,
and numbers growing up. Demography is a boring word for a mesmerizing
subject .

THE NEW MINORITY

Less than half of the work force in the industrial world will be in “proper”
full-time jobs in organizations by the beginning of the twenty-first century. Those
full timers or insiders will be the new minority, just when we had begun to think
that proper jobs were the norm for everyone. The others will not all be unem-
ployed, although in every country there will be some people who belong to this
“reserve army” as Marx called it. More will be self-employed, more and more
every year; many will be part timers or temporary workers, sometimes because
that is the way they want it, sometimes because that is all that is available. And
then there is, everywhere, another reserve army of women in waiting, those whom
the OECD so accurately calls, “unpaid domestic workers,” mothers whose talents
and energies are not totally absorbed by their families. Add these disparate groups
together and already they just about equal the numbers of those with the full-time
proper jobs.
When less than half the available work force is in full-time employment, it
will no longer make sense to think of a full-time job as the norm. Continuous
change will have flipped into discontinuous change, and we shall begin to change
our views of “work,” of “the job,” and of “a career.”
The reason for the shift is the emergence of the shamrock organization. . . .
Essentially, it is a form of organization based around a core of essential executives
and workers supported by outside contractors and part-time help. This is not a
new way of organizing things-builders large and small have operated this way
for generations, as have newspapers with their printers and their stringers, or
farmers with contract harvesting and holiday labor. What is new is the growth of

Reprinted by permission of Harvard Business School Press from Chapter 2 of The Age of Unreason by
Charles Handy. Boston: 1991. Copyright 0 1994 by Charles Handy.
167
168 KNOWLEDGE LMANAGEMENT AND ORGANIZATIONAL DESIGN

this way of organizing in big businesses and in the institutions of the public sector.
All organizations will soon be shamrock organizations.
They have proliferated because their way is cheaper. Organizations have re-
alized that, while it may be convenient to have everyone around all the time, hav-
ing all of your work force’s time at your command is an extravagant way of
marshalling the necessary human resources. It is cheaper to keep them outside the
organization, employed by themselves or by specialist contractors, and to buy
their services when you need them.
When labor is plentiful, when you can pick and choose between suppliers,
the shamrock organization provides a sensible strategy. It is a sensible strategy
when your work ebbs and flows as it tends to d o in service industries. Unlike in
manufacturing, where any surplus resources of people or equipment can always
be turned to good advantage by producing things for stock for the weeks of peak
demand, the service industries cannot, or at least should not, stockpile their cus-
tomers. They must therefore flex their work force.
Both these factors currently exist. The labor supply, the potential work
force, is growing in all the industrialized countries as the baby boomers of the
1960s, and their wives, join the work force during the 1990s-an extra million or
so in Britain, for instance. At the same time the shift to the service sector continues
inexorably everywhere. Between 1960 and 1985 the share of employees in the
service sector in the USA rose from 56 to 69 percent and in Italy from 33 to 55
percent. It is unlikely to change back. The two factors work on each other; a
growing service sector offers greater opportunities to women, which increases the
potential work force, which in turn increases the potential for more flexible ways
of organizing.
It has been happening slowly, so slowly that most people have not noticed
the new dimensions. Before very long the full-time worker will be a minority of
the working population. Our assumptions about how the world works, how taxes
are collected, families supported, lives planned, and corporations organized will
have to change radically. The Universal Declaration of Human Rights, which in
1947 guaranteed a choice of job to everyone, will be a clear anachronism.

THE NEW INTELLIGENTSIA

The second statistic is alarming in a different way. A study by McKinsey’s
Amsterdam office in 1986 estimated that 70 percent of all jobs in Europe in the
year 2000 would require cerebral skills rather than manual skills. In the USA, the
figure is expected to be 80 percent. That would be a complete reversal of the
world of work some fifty years earlier, Discontinuity indeed!
It is impossible to be precise about such things. There is, to start with, no
clearcut distinction between a cerebral job, requiring brain skills, and a manual
job, needing muscles. Even simple manual jobs, like gardening, now require a de-
gree of brains to understand the proper use of fertilizers and herbicides, to distin-
guish plant varieties, and to maintain machinery.
The Numbers 169

What is more controversial and even more alarming is McKinsey’s estimate

that one-half of these brain-skill jobs will require the equivalent of a higher educa-
tion, or a professional qualification, to be done adequately. If that is even approxi-
mately true, it means that some 35 percent of an age group should today be
entering higher education or its equivalent if the labor force is going to be ade-
quately skilled by the year 2000. McKinsey’s estimate may even be on the conser-
vative side. If we look at the new jobs alone, the current expectation is that 60
percent of them will be managerial or professional of some sort.
In spite of these trends the percentage of young people in Britain going on to
higher education is currently 14 percent, rising to 18 percent by 1992, but only
because there will be fewer teenagers in total. In the rest of Europe the overall fig-
ure is around 20 percent, with small national differences. In France, for instance,
36 percent pass their baccalaureate and are therefore entitled to enter university,
but nearly half leave, or are asked to leave, at the end of the first year. Only Japan,
the USA, Taiwan, and South Korea seem to have university populations of the
right sort of size for the future, and in those countries there are concerns about the
quantity if not the quality of some of what is called “higher education.”
If these estimates of the required levels of education are even partly true, it
means that not only will we see alarming numbers of skill shortages but that,
more seriously still, we may lack the skills and the wits even to create the busi-
nesses and the opportunities which will then encounter skill shortages! It will, of
course, be an invisible discontinuity. We will not miss the organizations we have
not had, and never thought to have. . . .

THE VANISHING GENERATION

In the nineties, there will be almost 25 percent fewer young people leaving
school. At first glance this seems like a timely end to the problem of youth unem-
ployment. A second glance changes the picture because it points to even more
pressure on the relatively small percentage who have the brain skills needed by to-
day’s work force. The bulk of the newly reduced cohort of young people will still
be like those who leave British schools without a proper certificate in even one
subject, 43 percent of them in 1986.
A 1988 British report by the National Economic Development Office and
the Training Commission, “Young People and the Labour Market, A Challenge
for the 1990s,” pointed out that in 1987, less than twenty large employers took
on half of all the 27,000 school leavers with two or more A-levels who were look-
ing for work. The drop in youth population, therefore, is a problem because, if
nothing is done, it means that the supply of brain skills, already inadequate, will
be even more inadequate, and that the skills shortages referred to above will be-
come even more severe. The competition for the more educated will intensify, and
the rejection of the less educated will be felt even more cruelly. Youth unemploy-
ment will not be solved, indeed it will be raised a notch or two.
170 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

The statistic, 25 percent fewer, is an opportunity, however, if it makes it eas-

ier to tackle the task of educating more of our young men and women for life and
work in the world of brain skills. Without anyone doing anything, as every gov-
ernment has discovered, the percentages of those going on to higher and further
education are bound to improve as the base number falls. Doing rather more will,
in percentage terms, make a great deal of difference and will set markers for the
future.
Those markers are important because they must change a culture. There is
no innate reason why Britain should be sixteenth in the OECD league table of
young people in education after 16 years of age-above only Portugal and Spain.
British teenagers are not innately more stupid or less educable; they are simply the
inheritors of a tradition which held that book learning was for the few; that real
life, and real money, should begin as soon as possible, and that manual and prag-
matic skills were best learnt on the job. The past, as so often in Europe, deter-
mines the future, even though the degree to which these beliefs might have been
true in the world of work as it used to be is the degree to which they must be less
true today.
In Japan, top of the OECD table, 98 percent of young people stay on in
formal education until 18 years of age, even though that education is far from
stimulating and far from being pragmatic. They are the inheritors of a different
cultural tradition, one that just happens to be more attuned to the needs of the fu-
ture than that of Britain and most of the rest of Europe. In America, too, the
young stay on in school, but whether they learn anything there is a question of
growing concern.
The information society, after all, uses information, in the form of numbers,
words, pictures, or voices-on screens, in books, or in printouts and reports-as
its currency. The essential requirement, therefore, of all its workers is that they are
able to read, interpret, and fit together the elements of this currency, irrespective,
almost, of what the data actually relates to. That is a skill of the brain. It can be
taught or at least developed in classrooms. It does not, for most people, happen
quickly, easily, or early but requires years of practice, years which are most con-
veniently and usefully spent at the beginning of adult life rather than inconven-
iently in the middle. This general skill is akin to riding a bicycle, once learnt it is
never unlearnt, and having learnt it, one can then go on to learn its use in particu-
lar applications.
It is this conviction that brain skills are of general use and can be developed
in youth that has led places like Taiwan and South Korea, following Japan, to put
such an emphasis on the formal, even scholastic, education of their youth. It has
been said that every second person in Seoul has either been at university or is cur-
rently studying or teaching there, while in the 1970s Mr. Goh Thok Tong, then
minister for trade and industry in Singapore, was arguing that Singapore needed
“to step into the shoes left behind by countries like Germany and Japan as they
restructure, they from skill-intensive to knowledge-intensive and we from labor-
intensive to skill-intensive.” In pursuit of these objectives Singapore proceeded to
increase greatly the number of university places and to lower the entry require-
The Numbers 171

ments. Britain, who needs to be one step ahead of Singapore, has until recently
been doing the reverse.
The opportunity, however, remains and is made more accessible because of
the vanishing generation. The smaller population within an age bracket is also
good news for those who want to re-enter or enter late the work of the informa-
tion society. The squeeze on qualified youth will encourage employers to turn to
other sources of skill, particularly to women, many of whom have the necessary
early education but have been busy working to raise their families and manage
their homes. Less convenient as employees because they want and need more
flexibility, they have not been wooed too assiduously in the past. In the 1990s they
will be. They do, after all, represent nearly half of all university entrants (over half
in 1987 in the USA for the first time). They are a neglected resource which few
will be able to neglect once the -25 percent factor begins to bite. The N E D 0 Re-
port cited estimates that four out of five of the 900,000 new workers it foresees in
Britain’s work force over the next eight years will be women returning to work.
Women have re-entered the work force before, but the numbers and the con-
ditions under which they will return in the 1990s turn this into a significant dis-
continuity, which will change the way organizations are run, will affect family
structures and living patterns quite significantly . . .

THE THIRD AGE

In 1988 the social affairs ministers of the OECD met to contemplate the
time when one person in five will be a pensioner and one in ten aged over 75,
when there will be only three people of working age to support each pensioner,
and when old-age pensions may account for one-fifth of national income. It will
be even worse for Switzerland and West Germany where there will be only two
people of working age for each pensioner.
It will be 2040 before this scenario fully becomes ;f reality, but the people
who will be old then are alive now, and unless they quickly change their breeding
habits, the numbers of their children are quite predictable. This world will happen
and it will start to happen before the end of this century.
Once again, there have been old people before, but never before so many of
them. I knew only one grandparent-the others had died before I was born. My
children knew all four. Their children will almost certainly know a great-grand-
parent or two. People in their sixties and retired will still be someone’s children.
The infrequent has become the commonplace, and the world as we know it will
inevitably change in some way.
It is happening because, in the richer countries, it is becoming harder to die.
Each major cause of death is either diminished, like smallpox or polio and, one
day, cancer, or postponed for a few more years or decades, like heart disease. Of
course, nature, or man’s tampering with nature, may trigger another plague, and
some wonder whether AIDS may not be just that plague, but such disasters ex-
172 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

cepted, there seems little reason why many of today’s teenagers cannot expect to
live to 100, provided they do not drink, smoke, or drive themselves to death.
The question is, will they want to live that long? When death as an act of
God seems to be indefinitely postponed, will we want to make it increasingly an
act of mankind? Euthanasia, already quasi-legal in the Netherlands, may become
more acceptable to more societies.
More urgent are the questions, “What will they live on?” “What will they
d o ? ” “Who will care for them?” By the year 2020, if nothing changes, Italy will
be spending over a quarter of its national income on pensions, while Britain’s
health service will spend ten times as much on a patient over 7 5 as on one of
working age.
Like all discontinuities, however, this one contains opportunities as well as
problems if the changes are seen coming and if everyone concerned can indulge in
a little upside-down thinking.
The aged will not all be poor, for instance. An increasing number of them
will own their own homes, an asset which can be turned into an annual income
provided that they d o not intend to bequeath it to the next generation (who will
by then be in mid- or late career with their own homes bought and paid for). Most
of them will be healthy and active. That is, of course, why they are still alive. They
are capable of working. One British study found that 43 percent of over-65s regu-
larly helped other elderly people, 25 percent helped the disabled, 11 percent
helped neighbors. If we change our view of work to include such unpaid activity,
then these people are only retired in a legal or technical sense. After all, in the last
century no one had heard of retirement-they worked till they dropped, or, as a
farmer said once when I asked him what was the difference between farming a t 75
and farming at 50, “The same only slower!” Experience and wisdom can often
compensate for energy.
So, many older people will not go unnoticed, particularly when many more
of them will have experienced responsibility earlier in life and will not be used to
keeping quiet. If we are sensible we will want to use their talents in our organiza-
tions, but not full time or on full pay. We shall need, then, to rethink what jobs
call for part-time wisdom and experience, and what work can be done a t a dis-
tance by responsible people. We shall need to revise the tax rules for pensions to
make it economic for such work to be done. Many people, active and healthy, will
devise their own activities, organizing around their enthusiasms; we must not let
too many rules from the past stand in their way. We will need to change the way
we talk about them, words like “retirement” will become as antiquated as “ser-
vant” today. Words are so often the bridges of social change, the outward signs of
a discontinuity at work triggering some upside-down thinking.
Already the linguistic signposts are going up. The Third Age, the age of liv-
ing, as the French would have it, which follows the first age of learning and the
second of working, is already becoming a common term. There is a University of
the Third Age, a network of people exchanging their skills and their knowledge.
There will soon be more talk of Third Age careers. Soon, no doubt, there will be
Third Age societies and, ultimately, ministers for the Third Age in all OECD coun-
The Nirmbers 173

tries! The “wrinklies,” as my children fondly term us, can be assets as well as li-
abilities, if we want them to be.
If words are indeed the heralds of change, then the Third Age language sug-
gests that before too long we shall be referring to people’s job-careers as we now
do to their educations. “Where did you work?” to a 65-year-old with fifteen
years, at least, of life ahead will sound much like “Where did you go to school?”
It would all sound strange indeed to my father who died two years after retiring,
at the age of 74. For him there was no Third Age worth living, and the second age,
of job and career, had long been a burden before he could afford to leave it.
It will be different for us, his children, and for our children. It is change of a
discontinuous sort, but it need not be change for the worse if we can see it coming
and can prepare for it.

100,000(4)
2 =J

The changes coming to our ways of work and living, indeed the changes al-
ready here, are conveniently summed up by this strange equation. When it is un-
ravelled, it suggests that we have, for some time now, been engaged in a massive
job-splitting exercise in o u r society and have not even noticed it.
It will work like this. Thirty years ago when I joined an international com-
pany and started my job, I signed on, although I never realized it, for 100,000
hours of work during my lifetime, because I should, if I was anything like every-
one else in the developed world a t that time, be expected to work for 47 hours a
week, including overtime paid or unpaid, for 47 weeks a year for 47 years of my
life (from, on average, 18 to 65). 47 x 47 x 47 = 103,823 or 100,000 hours, give
or take a few.
My teenage son and daughter, a generation later, can expect their jobs to add
up, on average, to 50,000 hours. The lifetime job will have been halved in one
generation. At first sight this would imply that they would be working half as
many hours per week, for half as many weeks and half as many years. But mathe-
matics does not work like that. Just as half of 43 (64) is not z3 (8),so half of 473 is
not 23.53. In fact, rather bizarrely, half of the three 47s is three 37s, for 37 x 37 x
37 = 50,653.
It is because of this statistical sleight-of-hand that we have not noticed this
rather dramatic piece of discontinuous change. It is also, in part, because it is only
now beginning to bite as the next generation begin their second age of jobs and
careers.
The world is not so neat, however, as to switch uniformly from the three 47s
to the three 37s. That is where the (4) comes in. My daughter and my son have
four principal options to choose from.
In the first option they will follow in their father’s footsteps and look for a
full-time job, or at least a sequence of full-time jobs, in the core of an organization
or perhaps as a professional of some sort. In this case their working week will not
be that different from the one I knew. Statistically, it will average 45 hours per
174 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

week, with rather less overtime for the hourly paid and fewer Saturday mornings
for office workers. Nor will their working year be much reduced; longer annual
holidays bring it down to 45 weeks rather than 47.
What will change, however, is the length of their job life. To get one of those
increasingly rare jobs in the core or the professions (less than half of all jobs by
2000), they will need to be both well qualified and experienced. In Germany to-
day, a six- or seven-year university course is piled on top of eighteen months of
military or community service so that the average entrant into the job market is
27 years of age. In the USA, a postgraduate qualification of some sort, after a
four-year degree, is increasingly becoming a prerequisite for a good job, making
24 the normal age at which adults begin full-time jobs. Britain still has three-year
degree courses (except in Scotland) and no military service, but employers increas-
ingly look for further qualifications of a more vocational or professional nature
and for relevant experience in vacations or in “gap” years. It has, after all, been
the established practice in the older professions of medicine, architecture, and the
law for centuries-a long (seven-year) mix of education, experience, and voca-
tional training. We can expect to see it extend to many other occupations, with the
result that British parents must increasingly expect to wait until the offspring are
24 or 25 before they are established in a full-time job, if that is what they want.
It is possible that the fall in the numbers of qualified young people in all in-
dustrialized countries will tempt organizations and professions to shorten their
training requirements in order to get the best of a shrinking supply. The form this
will probably take, however, will be to finance them, perhaps under the guise of
employment, during their studies. It will be education more generously funded,
not a job.
The next generation of full-time core workers, therefore, be they profession-
als, managers, technicians, or skilled workers, can expect to start their full-time
careers later-and to leave them earlier. This is the crucial point. The core worker
will have a harder but shorter job, with more people leaving full-time employment
in their late forties or early fifties, partly because they no longer want the pressure
that such jobs will increasingly entail, but mainly because there will be younger,
more qualified, and more energetic people available for these core jobs.
It is true that early in the next century the total number of people in the
work force in every country will start to decline and the average age to rise, as the
dip in the birthrate of the 1970s works its way through life. However, with the re-
duced numbers of full timers, employers will continue to place a premium on
youth, energy, and qualifications whenever they can get them in combination. It
will be a shorter life but a more furious one for the full timers, as the new profes-
sionals in business are already discovering.
The net result of these changes will be a full-time job which, on average, will
result in 45 hours for 45 weeks for 25 years, totalling 50,000 hours. Work won’t
stop for such people after 50 but it will not be the same sort of work; it will not be
a job as they have known it. They will enter their Third Age sooner than others,
affluent, no doubt, but still with a good third of their lives to live.
It is happening already. One personnel manager was surprised to discover
that only 2 percent of his work force were, as he put it, still there a t the official
The Numbers 175

retirement age of 62. What he had done was to look back fifteen years to those
who were then 47 and had found that only a few had stayed on with the organi-
zation for the remaining fifteen years. Some had moved to new jobs and one or
two had died, but the great majority had opted for, or been persuaded into, early
retirement in their fifties. “We knew that people were leaving us early,” he said,
“but we had n o idea of the scale of it all until we started counting.” An advertis-
ing agency, aware that creativity and mental energy tend to decline with the years,
would like to see everyone under the age of 50. They have not, so far, been al-
lowed by the tax authorities to make their full pension scheme applicable under
the age of 55 but they are confident that it will come down to 50 within the next
ten years-well in time for the generation now starting their careers.
There will always be the glorious exceptions of course. Moreover, those who
control their own careers-the self-employed, the professionals, and, apparently
heads of state-will buck the trend as long as the clients and their supporters will
permit it. It is the bigger organizations, in which most full timers still work, who
will be most choosy about whom they keep on their full-time books, and they will
want the energetic, the up-to-date, the committed, and the flexible. Most of those
will be in their thirties and forties, putting in their 50,000 hours in big annual
chunks. Full-time work in organizations will, however, be only one of the options
and, if the numbers are anything like right, it will be a minority option, perhaps an
elite one. Most people will find their place outside the organization, selling their
time or their services into it, as self-employed, part-time, or temporary workers.
For them the pattern of hours will be different. They may find themselves
working 25 hours a week for 45 weeks of the year (part time) or 45 hours a week
for 25 weeks a year (temporary). In either case they will need to keep on working
as long as they can, for 45 years if possible, because they will not be able to accu-
mulate the savings through pension schemes or other mechanisms to live on. This
will suit the organizations who will look for experience and reliability in their
temporary staff, rather than for the energy and certainty of youth. In both tempo-
rary work and part-time work, the sum is still 25 x 45 x 45 = 50,000.
We may, therefore, see the national retirement age going in two very differ-
ent ways at the same time. While for the core workers it will gradually come down
toward 50 over the next twenty years, for most of the work force it will go up. For
them the questions, “What shall I d o in the missing 50,000 hours, and what shall
I live on?” cannot be postponed until the Third Age; they need to be answered
now. For these people the future is not a generation away-it started yesterday.
My children have a fourth option. They may be able to work full time for
ten years, take ten years out to raise a family, then return to the work force at, say,
45 for a further ten or even fifteen years. (45x 45 x 25 hours of paid work =
50,000,) It is an option that has traditionally been taken up by women, who have
varied the pattern by going part time for some of the intervening years, but it may
increasingly be seen as an opportunity by men to vary their lifestyle and to play a
bigger role in home and family life.
Re-entry into the full-time work force has always been difficult. It will get
easier as the shortage of qualified young people begins to bite organizations in the
1990s. The organization will then turn to that reservoir of talent, the qualified
176 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

women at home. In order to tempt them back, however, organizations may have
to learn to be more flexible, more willing to recognize that they are buying some-
one’s talents, but not necessarily all their time.

THE PRESSURES BEHIND THE NUMBERS

100,000(4)
The = J equation is, of course, spuriously precise. The numbers
will not work out precisely like that. The equation is there to make a point. The
world of jobs is changing. It is changing more dramatically than we realize be-
cause numbers like this creep up on one unexpectedly when multiplied out over a
lifetime.
No one particularly wants those numbers to happen. They are not the result
of any policy decisions by government or boards of directors. They are the cumu-
lative result of many instinctive responses to a changing environment. There is
now some general agreement about the nature of this changing environment and
an acceptance that it is not going to change back again. Some of the main features
are these:

A Move Away from Labor-Intensive Manufacturing

Thirty years ago nearly half of all workers in the industrialized countries
were making or helping to make things. In another thirty years’ time it may be
down to 10 percent (in the USA it is already down to 18 percent).
To some extent this is because we have all had to export our factories, in-
stead of our goods, to countries where labor is cheaper and more amenable to
working in factories. Even Japan has now been forced by the high price of the yen
to follow suit. When Britain did not export her factories soon enough, they were
replicated in the newly industrialized countries, and the UK lost out. Situations
such as the rapid rise of the pound sterling in the early years of Margaret
Thatcher’s government in Britain only accelerated this process, leaving swathes of
abandoned factories throughout Britain. It would have happened anyway. The
clever thing would have been to join the unbeatable (rather than competing with
them) by exporting the factories not the goods. Discontinuous change can always
be turned to advantage with a bit of forethought.
The result is not just fewer jobs, but different organizations. Labor-intensive
manufacturing was traditionally managed with a large pool of relatively cheap la-
bor, a lot of supervision, and a hierarchical management structure. There were a
lot of people around, most of them full-time employees whose time was bought to
be used at the discretion of the organization, subject increasingly to the agreement
of the union.
It was a convenient way to run things; everything and everyone you needed
was yours. If you want to control it, own it, was the message. It proved, in the
The Numbers 177

end, to be a very expensive message. The Japanese always did it differently, with a
small core staff, a raft of subcontractors, heavy investment in clever machines,
and enough clever people to instruct them and work with them. The demise of
mass manufacturing has led to the end of the mass employment organizations and
a redefinition of the job.

A Move Toward Knowledge-Based Organizations

The end of labor-intensive manufacturing leaves us with organizations

which receive their added value from the knowledge and the creativity they put in
rather than the muscle power. Fewer people, thinking better, helped by clever ma-
chines and computers, add more value than gangs or lines of unthinking “human
resources.” Manufacturing has gone this way. The more obviously knowledge-
based businesses of consultancy, finance and insurance, advertising, journalism
and publishing, television, health care, education, and entertainment, have all
flourished. Even agriculture and construction, the oldest of industries, have in-
vested in knowledge and clever machines in place of muscles.
The result is not only a requirement for different people, but different or-
ganizations, organizations that recognize that they cannot d o everything them-
selves, that they need a central group of talented and energetic people, a lot of
specialist help and ancillary agencies. They are smaller, younger organizations
than their predecessors, flatter and less hierarchical. . . . Their most immediate ef-
fect is on the numbers-fewer people inside who are better qualified, more people
outside who are contracted, not employed.

A Move Toward Service

Paradoxically, rich societies seem to breed dependency. If you are poor you
are forced into self-sufficiency. As you get rich, it is easier and more sensible to get
other people to d o what you do not want to d o or cannot do, be it fixing the roof
or digging the garden. It makes economic sense to let others make your clothes
and to buy them in the store, that way you get better clothes and more time to d o
what you are good at. It goes on and on. Convenience foods take the chore out of
cooking, and package holidays, the work out of leisure. All of us become more
specialized, better at one thing and worse at others. Like knowledge-based organi-
zations we contract out everything we are not good a t and so breed a raft of serv-
ices on which we now depend.
Affluence breeds service industries, and they in turn create affluence. Some-
times it seems as if everyone is taking in everyone else’s metaphorical washing and
making money out of it, or as in my particular case, that everyone is going to eve-
ryone else’s conference and being paid for it or paying for it. Affluence is a matter
of mood and self-confidence as much as of economics, for dependency has its own
imperatives. If you need to buy all these services, you have to find something to d o
178 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

to pay for them, hence some competitive striving. It is a self-fulfilling prophecy

which works as long as everyone believes the prophecy of continued affluence.
The service industries of affluence are therefore ephemeral creations, which
could always disappear overnight. The point, however, once again, is that the or-
ganizations they spawn are of a different kind. Because they are essentially ephem-
eral, they have to flex with every shift in demand. Small core staff and lots of
part-time and temporary help has to be the rule. Many of them are not knowl-
edge-intensive businesses, although some are, of course. Retailing, transport,
cleaning, catering, leisure, are all industries with large requirements for the com-
petent but semi-skilled. It is here that you will find most of the 30 percent who d o
not have the brain skills for the knowledge-based organizations. It is here that you
will find the bulk of the part timers and the temporary workers.
It is the growth of the service sector which has transformed the working
lives of so many people in Europe and the USA because of the kind of organiza-
tion which it needs and breeds.
These shifts are irreversible. The degree of affluence may increase or wane in
each country but labor-intensive manufacturing will not return to Europe, or to
the USA, or to Japan. Knowledge-based enterprises have to be the way forward
for all our countries, the more and the better the richer, whether they are manufac-
turing goods or providing services. The service sector will ebb and flow with local
prosperity but will never fade away.
If the shifts are irreversible, so are the changes in the patterns of work which
they induce, and therefore the numbers with which this chapter began. A dramatic
change in the economic climate may slow things down, but it will not stop them.
The world of work has changed already. We need to take notice.
 Workers-The Challenge
for the 1990s
Mahen Tampoe

As early as 1970, Peter Drucker, writing about approaches to managing technol-

ogy-based companies, raised the question of “whether traditional organization
structure is going to work tomorrow the way it has worked for the past 40 years.”
It was his view that organization structures and management styles appropriate
for manufacturing industry may be inappropriate for knowledge-based industries
which depended on the acquisition and exploitation of knowledge for competitive
advantage. It was his contention that “ideas do not observe hierarchical channels
of communication” and that therefore hierarchical structures were inappropriate
for knowledge-based organizations.
Kantrow (1984) points out that twentieth century manufacturing techniques
had succeeded in converting the “idiosyncratic work of artisans, craftsmen and
engineers into consistent, predictable and repeatable practice-to utility,” and in
this way limited the scope for innovation and creativity among its knowledge
workers. Sasaki (1991) writing on how the Japanese accelerate new car develop-
ment cycles, points out that “new product concepts wither in a company which
imposes too much bureaucracy on designers and restricts their creativity.”
At a recent workshop run by the author at Henley Management College on
the management of High-Tech projects, the participants (consultants and R & D
managers from industry and government) agreed that of fourteen problems listed,
a key problem was that of “specialist clash” which they defined as balancing the
need for autonomy of team members with the need to conform to teadorganiza-
tional norms to achieve effective outcomes. Bailyn (1985) in her research into the
management of technologists in R & D Laboratories focused on the management
dilemma of line managers between retaining control and allowing their staff ade-
quate autonomy to be creative and innovative. It would appear that new manage-
ment approaches seemed justified, this view is supported by Wilmot (1987) who

Reprinted with permission from Long Range Planning, Vol. 26, Mahen Tempoe, “Motivating Knowl-
edge Workers-The Challenge for the 1990s,” 1993, Elsevier Science Ltd., Pergamon Imprint, Ox-
ford, England.
179
180 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

points o u t that the investment gap in R & D is “in management of innovation”

and not in the availability of finance for R & D.
Beattie and Tampoe (1990), in their article on the problems of managing
and motivating technologists to achieve high performance in ICL, showed how
traditional human resource practices had to be modified or replaced with new
H R M initiatives to meet the H R M needs of ICL. Kuwahara, Okada and Horik-
oshi (1989) point out that Hitachi’s approach to this same problem was to de-
velop parallel career structures including the institution of a special “status”
structure to ensure that technologists and managers enjoyed similar status within
their R & D function. In the early 1980s authors such as Peters and Waterman
(1982),Rosabeth Moss Kanter (1983)and others focused our attention on organ-
izational cultures and empowered employees as being a means of overcoming
some of the problems of managing knowledge workers. However, empowered
employees need direction if their efforts are not to be wasted or self-indulgently
applied. It seems that the problem identified by Bailyn (1985) remains still unan-
swered. Perhaps the answer lies in drawing on the inner drive and motivation of
these knowledge workers, rather than in better methods of supervision. Herzberg
(1966) offers job enrichment as a way to solve this problem on the basis that it
will increase job satisfaction by meeting the motivational needs of employees.
This research study sets out to explore the motivational needs and organiza-
tional environments best suited to knowledge workers. It was based on motiva-
tion models proposed by previous researchers for managers and modified on the
basis of information gathered from literature reviews and recent empirical re-
search in related fields specific to knowledge workers in industry. It seemed inap-
propriate to extrapolate from the motivational needs of managers, clerical and
factory staff, as Hunt (1987) in his research has shown that the motivational
needs of staff varied with the nature of their work, their skills levels, and their do-
mestic and material circumstances. In the light of this evidence a new look at the
motivational and environmental needs of knowledge workers seem justified.
The research identified four key motivators, eight characteristics which im-
pacted on employee effectiveness at work and six predominant current manage-
ment practices. These are used as the basis for offering guidance to managers on
motivators and management methods appropriate to deriving high performance
from knowledge workers.

THE RESEARCH PROJECT

The research project began in May 1988. It drew on widely acknowledged

views on motivation theory; for example, those proposed by Herzberg (1987),
Maslow (1987), and McGregor (1960) and developments in motivation models,
for example, the work of Porter and Lawler (1968), Vroom (1964), and Hunt
(1986). It also looked at recent research into creativity (Amabile 1988), man-
agement practices in R & D establishments, (Bailyn 1985), and the career
expectations of recent graduates (Garden 1990). In addition various views on or-
ganizational design and management styles were explored. Based on what was
Motivating Knowledge Workers-The Challenge for the 1990s 181

learnt from this literature review a research model was proposed and tested using
data gathered by distributing a questionnaire which sought to collect information
on:

1. How knowledge workers perceived they were being managed in organi-

zations today; and how they thought they should be managed to be more
effective at work, and
2. What characteristics were important to this group of staff at work so that
key motivators could be deduced from their answers to the questions
posed.

THE RESEARCH MODEL

The model of motivation (Figure 11-1)which formed the basis of the em-
pirical research and the hypotheses, was based on the model proposed by Porter
and Lawler (1968) for managers. It also incorporated ideas from the research
work done by Amabile (1988).
The model postulated (a) the potency of different rewards which aroused
motivated behaviour, and ( b ) the presence of certain instrumental factors (e.g.,
task and domain relevant skills, role and goal clarity) which led to performance,
with performance leading to rewards and eventually to a sense of psychological
success. The model was built on the expectation that performance would result in
the expected rewards and that these would satisfy the motivated drives.

Value of Task &

rewards domain perceived
relevant equitable
enabling
skills
I
organization
I / rewards \
I
I TASK
d X d MOTIVATED-
I
X -WORK
AlTAl N PSYCHO-
LOGICAL
MOTIVATION
I ENERGY I GOALS SUCCESS
I I
personal
effectiveness
role & goal
clarity
\ actual
/
rewards

expectancy
A I
FIGURE 11-1 Model for motivating knowledge workers.
182 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

DATA COLLECTION AND ANALYSIS

A questionnaire consisting of 133 questions was prepared and circulated to

800 staff in R & D establishments, software development houses and engineering
firms. Smaller samples from many large multinationals in the U.K. were also can-
vassed. Three hundred and twenty-two staff responded to the questionnaire.
Three questions referred to demographic data such as age, sex and qualifications.
Ten referred to work related information such as job title, nature of work, team
size, years in employment, salary, and job satisfaction. There were 48 identical
questions each on current and preferred working conditions and 24 relating to
motivators a t work. The respondents were required t o express the absence or
presence of certain characteristics by rating each characteristic against a 7 point
Likert scale. The sample comprised 91 managers of knowledge workers, 40 pro-
ject managers, and 191 team members comprising 44 consultants, 43 researchers,
58 implernentors, and 46 designers. Their ages ranged from 20 to 59. Two hun-
dred and thirty of the respondents were below the age of 39, the average age of
the sample was 32 years. Two hundred and twenty-one were married, 91 were
single and 10 divorced. There were 60 female staff in the sample representing a
higher proportion than in industry in general. One hundred and fifty-five of the
staff had less than 5 years service with their present employers. Two hundred and
seventy had spent less than 5 years in their current jobs. The majority of the sam-
ple worked in teams of less than 10 staff, which probably representative of IT and
R & D teams.
The responses received were analysed using factor analysis as the main ana-
lytical tool along with discriminant and conjoint analysis to differentiate and pri-
oritize the motivators. The data analysis revealed information on what motivators
were relevant to knowledge workers and two secondary but related aspects of
worklife which the research set out to investigate were also highlighted namely;

1. The way knowledge workers perceived they were managed.

2. The organizational climate which knowledge workers felt they needed if
they were to be effective at work.

THE EMPLOYEE
In interpreting the data collected from this research it is important to con-
sider the employees themselves. It would be wrong to assume that they were all
interested in the same motivators or that their preferences met the generalized
model. It is likely that staff were at different stages in their careers and domestic
circumstances and that these factors would affect the strength of their need for the
four motivators identified. For the purpose of this article we can propose a differ-
ent classification of staff based on their stage of personal and career development
(see Figure 11-2).We can hypothesize that employees will be at one of four stages,
namely;
Motivating Knowledge Workers-The Challenge for the 1990s 183

Stage 1-Fulfilment. These employees will feel a high sense of job satisfac-
tion derived from having achieved a preferred balance of those motiva-
tional and reward factors of importance to them. This could lead to a
transitional stage.
Stage 2-Transition. These employees have arrived a t a crossroads in their
career and personal development and are seeking to reposition themselves
for the future. This can happen to an employee at any age but is more
likely to occur in their late thirties to mid-forties. This can lead to either 3
or 4 below.
Stage 3-Developmental. These employees are seeking to reach their state
of equilibrium. They are likely to be in their twenties and early thirties, or
moving from one state of equilibrium to another irrespective of age or ca-
reer position.
Stage 4-Plateaued. These employees are likely to have decided that their
level of achievement and personal growth meets their motivational and
rewards needs and they are not seeking new challenges.

Developmental
stage 2

1
FULFILMENT

Career
stages
personal

- Transitional
stage

Developmental
stage
STAGE &
Plateau
stage

Motivational states

Movement is from fulfilment to transition leading to Q Mahen Tampoe

either development stage 2 or plateaued stage which in turn
may lead to a state of fulfilment

FIGURE 11-2
184 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

THE MOTIVATORS

Four key motivators were identified:

1. Personal growth-the opportunity for individuals to realize their poten-

tial, supporting the hypothesis that knowledge workers were seeking in-
tellectual, personal and career growth.
2. Operational autonomy-a work environment in which knowledge
workers can achieve the tasks assigned to them within the constraints of
strategic direction and self-measurement indices.
3. Task achievement-the achievement of producing work to a standard
and quality of which the individual can be proud. There is a need for the
task undertaken to be relevant to the organization.
4. Money-earning an income which is a just reward for the contribution
made and enables employees to share in the wealth created by them,
through incentive schemes geared to their company’s success but related
to their personal performance.

To help prioritize these four motivators a new questionnaire was prepared

and circulated to 75 knowledge workers. Figure 11-3 below shows how these mo-
tivators were prioritized by the sample. It must be pointed out that whilst the ma-
jority fitted this general pattern, some showed a very strong preference for one out
of the four motivators, emphasising the fact that this generalized data cannot be
used without the priorities for each individual being identified and catered to.
This ranking highlights the significantly lower importance placed by the re-
spondents on monetary rewards. No doubt this is due in part to the fact that they
all earned well above the national average wage and that money in its varying
forms must be considered as having little incremental value as a motivator, even
if it is related to individual performance, unless the potential earnings are very
significant.
Once the financial rewards offered meet those prevailing in the industry it is
the other three motivators, namely personal growth, operational autonomy, and

Motivator Percentage Preferred option

Personal growth 33.74% Significant growth
Operational autonomy 30.51% Freedom to work within the rules
Task achievement 28.69% Very high
Money 7.07% Salary + bonus on personal effort
100.01~0

FIGURE 11-3 Principal motivators of respondents.

Motivating Knowledge Workers-The Challenge for the 1990s 185

task achievement, which managers should use to achieve motivated behaviour

from their staff. When these three motivators are combined in a three circle dia-
gram (Figure 11-4) four motivated states can be identified. These are:

Segment A-motivated behaviour: Achieved by meeting the combination of

all three motivators desired by all knowledge workers.
Segment B-Supervised behaviour: This combination of personal growth
and task achievement excludes operational autonomy, thus implying that

A = motivated behaviour 0 Mahen Tarnpoe

B = supervised behaviour
C = employee-centred behaviour
D = organization-centred behaviour

FIGURE 11-4
186 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

staff are within a supervised environment. As two o u t of the three re-

quirements are met, however, it is still seen as motivational, especially for
staff who see themselves as being in a developmental stage or employees
who feel they have plateaued.
Segment C-Employee-centred behaviour: This combination of opera-
tional autonomy and personal growth excludes task achievement, sug-
gesting that the employee is probably not doing work which he or she
feels meets organizational needs. This may be demotivating for some
staff, especially those in their developmental phase, but not for those who
are in a transitional stage where they may find the independence from
strong task dominance useful to reorientate their careers.
Segment D-Organization-centred behaviour: This combination of opera-
tional autonomy and task achievement excludes personal growth and is
likely to meet the needs of the plateaued employee, especially the more
senior ones who have the experience and ability to deliver outcomes, pro-
vided they are given clear goals and objectives. They may, for example,
trade-off personal growth for job security. The degree of freedom this
would give them will compensate for losing out on promotions to the
managerial ranks. Others who will benefit from these motivators are
likely to be staff in their developmental stage where they may seek to
prove themselves at the temporary sacrifice of personal growth.

HOW SHOULD MANAGERS USE THIS DATA?

In this article we have described four states of motivated behaviour and four
stages in the career of a n employee and have related them to each other offering
scope for their effective use. However, progress can only be made if managers are
able to identify what career stage employees have or perceive they have reached
and then apply a management style which will fit their motivational needs. This
means that team sizes and structures should be such that managers can relate to
their staff and better understand the motivational needs of their team members. In
addition, the following steps need to be taken:

1. Regular appraisals and career discussions with staff so that career stages
and motivational needs are understood by both employee and manager.
2. A working environment in which employees can achieve their preferred
motivators and rewards.

This second condition is very important as staff must believe that their per-
formance will result in the rewards they seek. This means that the work environ-
ment must not only offer these rewards but facilitate their achievement. However,
an analysis of current working practices showed they fell short of the conditions
considered necessary to satisfy the motivation needs identified earlier.
Motivating Knowledge Workers-The Cha/[eizgefor the 1990s 187

THE CURRENT WORKING ENVIRONMENT

When the data on perceived management styles was analysed, six current
management practices were identified as predominating the working environment
of the knowledge workers sampled. These were:

1. Activity orientation-a preference shown by management for activity

rather then for effective outcomes. This manifested itself in staff having
inadequate time to develop and test ideas before implementation due to
time pressures for delivery.
2. Creative autonomy-the freedom to be creative but a lack of operational
autonomy as a result of the bureaucratic bias recorded later. Although
staff could think innovatively they were unable to mobilize organiza-
tional resources to implement their ideas.
3 . Shared values-strong corporate cultures.
4. Bureaucratic bias-process dominance.
5 . Career stagnation-lack of promotion opportunities.
6 . Financial rewards focus-the use of money as a motivator in preference
to career opportunities or some of the motivators identified earlier. The
perception of staff was that they were unable to share in the financial suc-
cess of the firm.

When comparing the above list with the sample’s own preferences, only
Creative Autonomy and Shared Values were found to coincide. It would appear
that current management practices are failing to offer employees a motivating cli-
mate and that it is therefore necessary to identify those which would constitute an
effective work environment for knowledge workers. This we can do by an analysis
of the data on preferred work environments collected from the sample.

BUILDING AN EFFECTIVE WORKING ENVIRONMENT

When the data on effective working environments was analysed, eight re-
quirements were identified, namely:

1. Commercial relevance of the job-the need for the assigned task to be

relevant and important to the business.
2. Task competence-the ability to carry out assigned tasks.
3 . Task consistency-a stability and consistency in the tasks and goals set so
that staff can devote time and energy to their achievement without being
required to divert their attention to meet “management crisis” situations.
4. Directed skills-role clarity and skills, based on matching individual
skills to the assigned tasks and the opportunity to build new competen-
cies.
188 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

5. Creative autonomy-freedom to pursue the job as seen fit by the em-

ployee without detailed supervision or constraint in the way the job is
handled.
6 . Resources-the tools to d o the job.
7. Commitment-loyalty to the organization rather than to colleagues or
the profession.
8. Peer contacts-access to information and knowledge which is relevant to
the job or specialization.

These eight requirements were seen as being instrumental in enabling staff to

perform well at work, and they can be used as the basis for developing the compo-
nents of the “ideal” empowering work environment. The eight effectiveness fac-
tors can be summarized into five main requirements, namely:

Motivated and committed employees.

Individual competence-comprising task competence and creative auton-
omy.
Facilitative work environment-task consistency and resources.
Purgose-directed skills and commercial relevance of the assigned tasks.
Knowledge exchange-the genuine transmission of information up, down
and sideways in a listening rather than speaking organization.

THE MANAGEMENT MODEL

These five elements can be used to develop a management model for knowl-
edge workers as shown in Figure 11-5.

MOTIVATED
EMPLOYEE
, MOTIVATED
ENERGY
:’:
~
PROFESSIONAL
& PERSONAL
ACHIEVEMENT

FIGURE 11-5 Model for managing knowledge workers.

Motivating Knowledge Workers-The Challenge for the 1990s 189

Professional a n d personal achievement is seen as t h e trigger which, by lead-

ing t o psychological a n d material rewards, activates a n d sustains a n individual’s
motivated drive. T h e facilitative environment interacts with motivated drive a n d
individual competence t o release motivated energy. Motivated energy is directed
into professional a n d personal achievement by ensuring t h a t individuals have a
clear sense of purpose a n d a r e sustained by access t o information a n d peer c o n -
tacts. T h i s is essential for t h e sense of awareness t h a t is critical for success.

REFERENCES
Teresa M . Amabile, A model of creativity and innovation in organisations, Researchs in
Organisational Behaviour, 10 (1988).
Lotte Bailyn, Autonomy in the industrial R & D lab, H u m a n Resource Management, 2,
(1985).
D. E Beattie and E M . K. Tampoe, Human resource planning for ICL. L o n g Range Plan-
ning, 23 ( l ) ,(1990).
Peter Drucker, Technology Management and Society, Pan, London (1970).
Anna-Maria Garden, Career orientation of software developers in a sample of high tech
companies. R t 9 D Management 24 (4),(1990).
Frederick Herzberg, W o r k and the Nature of M a n , World Publishing Corporation, New
York (1966).
John W. Hunt, Managing People a t W o r k , McGraw Hill, New York (1987).
Rosabeth Moss Kanter, T h e Changemasters. George Allen and IJnwin, London (1983).
Yutaka Kuwahara, Osami Okada, and Hisashi Horikoshi, Planning research and develop-
ment at Hitachi. L o n g Range Planning 22 (3),(1989).
Abraham Maslow, Motivation and Personality (revised edn), Harper & Row, New York
(1987).
Douglas McGregor, T h e H u m a n Side o f Enterprise, Penguin, London (1960).
Lyman W. Porter and Edward E. Lawler 111, Managerial Attitudes and Performance, Rich-
ard D. Irwin, Inc., New York (1968).
Thomas J. Peters and Robert H. Waterman, I n Search of Excellcnce, Harper & Row, New
York (1982).
Toru Sasaki, How the Japanese accelerated new car development, L o n g Range Planning,
24 ( l ) ,15-25 (1991).
V. H. Vroom. W o r k and Motivation, Wiley, New York (1964).
Robb W. Wilmot, Change in management and the management of change, L o n g Range
Planning, 20 (6),(1987).
This page intentionally left blank
 Shop Level
Michel Crozier

THE SYSTEM OF POWER RELATIONSHIPS IN THE

SHOP: ITS EQUILIBRIUM AND ITS LIMITS

The behavior and attitudes of people and groups within an organization

cannot be explained without reference to the power relationships existing among
them. All the lessons of the past fifteen years’ research in organization have
brought to light more and more the importance of those problems of power and
control that the first attempts at understanding leadership had neglected.”
The situation in the Monopoly is especially interesting seen in this perspec-
tive. A main rationale of bureaucratic development is the elimination of power re-
lationships and personal dependencies-to administer things instead of governing
men. The ideal of bureaucracy is a world where people are bound by impersonal
rules and not by personal influence and arbitrary command. The organizational
system of the Monopoly has gone a long way toward realizing this ideal. It is pri-
marily characterized by the extent of impersonal ruling. People at the posts of
command d o not have much leeway. Their response to most eventualities has been
fixed in advance; their subordinates know this and can, therefore, act accordingly.
The seniority system makes it impossible for the higher-ups to intervene in the ca-
reers of their subordinates. Of course, there are many constraints and the usual
concomitant punishments; but what is important for our purpose is the impossi-
bility of arbitrary punishment rather than the leniency of the system.
If we re-examine the functioning of the plants we have described in this new
light, the system appears, on the one hand, to be very rational in this respect, yet,
on the other, to yield unintended results that change the meaning completely, at
least at the shop level.

“The Social System at the Shop Level,” an excerpt from Chapter 4 in The Bureaucratic Phenomenon
by Michel Crozier, 1964, pp. 107-111. Copyright 0 1964 by The University of Chicago Press. Re-
printed with permission.

191
192 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

The over-all extension of the rules, the stability and predictability of all oc-
cupational behavior, and the lack of interference across hierarchical echelons, all
weaken the chain of command considerably. Power is weak down the line, and in
its absence there is relative cordiality and lack of concern. Supervisors are passive
and workers tolerant. Some provision is still necessary for giving sanctions to the
system as a whole, and the roles of the director and assistant director still carry
some power. However, it is important that the centralization that has occurred has
caused the power of decision to be located a t a level where personal influence
is difficult to exercise, because of the number of people involved and the lack
of immediate reliable information. Thus even the real key relationship-be-
tween the workers and the director-is stripped of its power and control function.
The director, acting as a judge, remains impersonal. He cannot combine his de-
sires for action and power with his duty as the man who interprets and applies the
law.
But if power problems seem eliminated from the official line of command,
there remains one group relationship with all the connotations of dependency and
attendant emotional feelings. This is the relationship between the maintenance
workers and the production workers. We must try to understand how it has come
about and how it relates to the system as a whole.
This relationship is centered around the problem of machine stoppages. Ma-
chine stoppages , . . occur unusually often because of the difficulties in condition-
ing the raw material. This is a sore spot in the technological system. However,
comparable problems seem to be handled better in other factories in France, and
in similar factories working with the same technology in other countries. Else-
where, a t least, they are not considered the crucial events they have become in the
Monopoly.
There are apparently two complementary reasons for their being crucial in
the bureaucratic organizational setup of the Monopoly. First, machine stoppages
are the only major happenings that cannot be predicted and to which impersonal
rulings cannot apply. The rules govern the consequences of the stoppages, the re-
allocation of jobs, and the adjustment of the work load and of pay; but they can-
not indicate when the stoppage will occur and how long it will take to repair. The
contrast between the detailed rigidity of all other prescribed behavior and the
complete uncertainty of mechanical functioning gives this problem disproportion-
ate importance. Second, the people who are in charge of maintenance and repair
are the only ones who can cope with machine stoppage. They cannot be overseen
by anyone in the shop. No one can understand what they are doing and check on
them. Furthermore, a department-a rather abstract services unit-is not respon-
sible. Instead, men are individually responsible, each of them for a number of ma-
chines, Thus there is another contrast between impersonality and abstractness on
the one side, and individual responsibility on the other.
Production workers are displeased by the consequences of a machine stop-
page. It disrupts their work; it is likely to make it necessary for them to work
harder to compensate for lost time; and if it lasts long enough, they will be dis-
placed, losing friendship ties and even status.18
The Social System at the Shop Level 193

With machine stoppages, a general uncertainty about what will happen next
develops in a world totally dominated by the value of security. It is not surprising5
therefore, that the behavior of the maintenance man-the man who alone can
handle the situation, and who by preventing these unpleasant consequences gives
workers the necessary security-has a tremendous importance for production
workers, and that they try to please him and he to influence them. From this state
of affairs, a power relationship develops.
The contrast between the power wielded by the maintenance men and the
lack of influence of the supervisors explains the advantage that the former have
over the latter. Supervisors cannot check on maintenance. They may be competent
in the various aspects of their work, but their competence does not extend to the
only problem about which the workers care, because only its outcome is uncer-
tain. A supervisor cannot reprimand the mechanics who work in his shop. There
is likely to be a perpetual fight for control, and the supervisors will usually be the
losers. It is, therefore, natural for them to have low morale, and to adjust to their
situation only after having resigned themselves to being the losers-using what-
ever rationalization they please.
Maintenance workers, on the other hand, have the best of this situation; but
their power is contested. It is not an overt, legitimate power. It does not fit the
usual expectations of industrial leadership. As a result, maintenance workers still
feel insecure. One can understand that their aggressiveness is a way of warding off
any attack, of cementing the group solidarity and making individual compromise
impossible. It is a value necessary to group struggle-and effective in it. Soul-
searching and moderation are qualities the group will definitely refuse to consider;
and these qualities tend to make people marginal, if not outcasts.
Production workers resent their dependence, but cannot express their hostil-
ity openly, because they need the maintenance men’s help and good will individu-
ally at the shop, and because, collectively, they know that they can keep their
privileges only by maintaining a common front with the other workers’ group.
Union solidarity and working-class unity are the values in the names of which
production workers accept the maintenance workers’ leadership. These values are
important to them, because of their feelings of insecurity. They feel that they have
rights and privileges that are not customary in the usual industrial setups in
France, and that they must protect themselves. They fear that they will not be able
to keep these assets unless they are prepared to fight. Since the production work-
ers are in this state of mind, the threats of abandoning them which maintenance
men make are always suc~essful.’~
The system of organization we have described may appear quite unwork-
able. Groups fight endlessly. It seems that there is no way of making changes and
adjusting to new conditions. The system appears completely static. Yet it works,
with a low but adequate degree of efficiency, and it has incorporated, in one way
or another, every stage of technological progress.
One should not, therefore, translate the burden of the opinions expressed
and the attitudes revealed into too black-and-white a picture. Conflict, forces dis-
couraging growth, a general conservative system of human adjustment, all put a
194 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

premium on conformity and conservative values. However, these tendencies can-

not develop ad infinitum. There are constant checks that prevent them from going
too far and threatening the permanence of the system itself. Behind the struggles
there is also, as in any sort of organization, some kind of consensus and organiza-
tional commitment.
Keeping these qualifications in mind, we can make a few final remarks
about the consequences of the bureaucratic patterns we have observed.
As in the Clerical Agency, the combination of impersonal rules, the absence
of promotion from one role to another, and the seniority system, tend to make the
hierarchical line progressively weaker. This preserves the personal independence
of each employee in respect to the higher-ups, but it produces new kinds of frus-
tration inasmuch as it provides no way of solving immediate problems.
Second, the development of the holders of each separate role into a stratum
or an estate-like group has, as its consequence, the submission of each employee
to considerable group pressure. In its own way, group pressure replaces the dwin-
dling hierarchical pressure.
In addition, there is a decline in the importance of instrumentality in all
personal judgments, accompanied by an increase in the role of affectivity. Human
relations, and especially group relations, are likely to be more acute in a bureauc-
racy than in an organization where sanction is found in relevant measurable re-
sults.
Fourth, the initiation of change is made very difficult within an organization
where the only kind of leadership is an administrative judicial one. If fairness is
the only legitimate value a director can advocate, then whenever there is resistance
he is likely to abandon his role as an agent of change. Change will occur only
when external pressure becomes impossible to withstand. The company directors
will then administer such change in a very impersonal way, without paying due re-
gard-and rightly, if they want to succeed-to the specific requirements of each
plant. To counter this kind of action, subordinates will develop a very distrustful
and demanding attitude, which will enable them to take full advantage of all the
consequent inadequacies.
Finally, new power relationships develop around the loopholes in the regula-
tory system. Groups fight for control of the ultimate strategic sources of uncer-
tainties, and their fates in the group struggle depend on their ability to control
these. New power relationships will have, as a consequence, new kinds of depend-
encies and frustrations, which will exert pressure for more centralization and rein-
force the demand aspect of the subordinate-superior relationship, creating a sort
of vicious circle that, at least a t this level, it will be impossible to evade.
The Social System at the Shop Level 195

NOTES
17. This opinion has been more and more frequently expressed since the middle fifties by
social psychologists and specialists in the theory of organizations.
18. They are held personally responsible for all stoppages of less than one hour and a half
and must compensate for the loss of production; if the stoppage is longer, they will be
displaced or may be sent around to d o menial jobs if there is no possibility of bumping
less senior fellow workers.
19. One of these threats is for a minority of the group to start an autonomous union of
maintenance workers. Such attempts are usually temporary, but they impress produc-
tion workers very much.
This page intentionally left blank
 Industrial Work
Shoshana Zuboff

In the older pulp and paper mills of Piney Wood and Tiger Creek, where a highly
experienced work force was making the transition to a new computer-based tech-
nology, operators had many ways of using their bodies to achieve precise knowl-
edge. One man judged the condition of paper coming off a dry roller by the
sensitivity of his hair to electricity in the atmosphere around the machine. Another
could judge the moisture content of a roll of pulp by a quick slap of his hand. Im-
mediacy was the mode in which things were known; it provided a feeling of cer-
tainty, of knowing “what’s going on.” One worker in Piney Wood described how
it felt to be removed from the physical presence of the process equipment and
asked to perform his tasks from a computerized control room:

It is very different now. . . . It is hard to get used to not being out there with
the process. I miss it a lot. I miss being able to see it. You can see when the
pulp runs over a vat. You know what’s happening.

The worker’s capacity “to know” has been lodged in sentience and dis-
played in action. The physical presence of the process equipment has been the set-
ting that corresponded to this knowledge, which could, in turn, be displayed only
in that context. As long as the action context remained intact, it was possible for
knowledge to remain implicit. In this sense, the worker knew a great deal, but
very little of that knowledge was ever articulated, written down, or made explicit
in any fashion. Instead, operators went about their business, displaying their
know-how and rarely attempting to translate that knowledge into terms that were
publicly accessible. This is what managers mean when they speak of the “art” in-
volved in operating these plants. As one manager at Piney Wood described it:

There are a lot of operators working here who cannot verbally give a de-
scription of some piece of the process. I can ask them what is going on a t the

Excerpts from Chapter 2 from In the Age ofthe Smart Machine: The Future of Work and Power by
Shoshand Zuboff. Copyright 63 1988 by Basic Books, Inc. Reprinted by permission of Basic Books, a
division of HarperCollins Publishers, Inc.
197
198 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

far end of the plant, and they can’t tell me, but they can draw it for me. By
taking away this physical contact that he understands, it’s like w e have
taken away his blueprint. He can’t verbalize his w a y around the process.

In this regard, the pulp and paper mills embody a historical sweep that is un-
available in many other forms of work. Unlike other continuous-process indus-
tries, such as oil refining or chemical production, the pulp-and-paper-making
process has not yet yielded a full scientific explication. This has retarded the
spread of automation and also has worked to preserve the integrity of a certain
amount of craft know-how among those operators with lengthy experience in the
industry. Like other continuous-process operations, the technological environ-
ment in these mills has created work that was more mediated by equipment and
dependent upon indirect data than, say, work on an assembly line. However, dis-
crete instrumentation typically was located on or close to the actual operating
equipment, allowing the operator to combine data from an instrument reading
with data from his or her own senses. Most workers believed that they “knew”
what was going on at any particular moment because of what they saw and felt,
and they used past experience to relate these perceptions to a set of likely conse-
quences. The required sequences and routines necessary to control certain parts of
the process and to make proper adjustments for achieving the best results repre-
sented a form of knowledge that the worker displayed in action as a continual re-
flection of this sentient involvement. Acquired experience made it possible to
relate current conditions to past events; thus, an operator’s competence increased
as the passing of time enabled him or her to experience the action possibilities of
a wide variety of operating conditions.
In Piney Wood and Tiger Creek, the technology change did not mean simply
trading one form of instrumentation for another. Because the traditional basis of
competence, like skilled work in most industries, was still heavily dependent upon
sentient involvement, information technology was experienced as a radical depar-
ture from the taken-for-granted approach to daily work. In this sense, workers’
experiences in these mills bridge two manufacturing domains. They not only illus-
trate the next phase of technological change within the continuous-process indus-
tries but also foreshadow the dilemmas that will emerge in other industrial
organizations (for example, batch and assembly-line production) with the transi-
tion from machine to computer mediation.
When a process engineer attempts to construct a set of algorithms that will
be the basis for automating some portion of the production process, he or she first
interviews those individuals who currently perform the tasks that will be auto-
mated. The process engineer must learn the detail of their actions in order to
translate their practice into the terms of a mathematical model. The algorithms in
such a model explicate, rationalize, and institutionalize know-how. In the course
of these interviews, the process engineer is likely to run up against the limits of im-
plicit knowledge. A worker may perform competently yet be unable to communi-
cate the structure of his or her actions. As one engineer discovered:
The Abstraction of Industrial Work 199

There are operators who can run the paper machine with tremendous eff-
ciency, but they cannot describe to you how they do it. They have built-in
actions and senses that they are not aware of. One operation required pull-
ing two levers simultaneously, and they were not conscious of the fact they
they were pulling two levers. They said they were pulling one. The operators
run the mill, but they don’t understand how. There are operators who know
exactly what to do, but they cannot tell you how they do it.’

Though every operator with similar responsibilities performs the same func-
tions, each will perform them in a unique way, fashioned according to a personal
interpretation of what works best. A process engineer contrasted the personal ren-
dering of skill with the impersonal but consistently optimal performance of the
computer:

There is no question that the computer takes the human factor out of run-
ning the machine. Each new person who comes on shift will make their own
distinct changes, according to their sense of what is the best setting. In con-
trast, the computer runs exactly the same way all the time. Each operator
thinks he does a better job, each one thinks he has a better intimate under-
standing of the equipment than another operator. But none of them can
compete with the computer.

These comments describe a particular quality of skill that I refer to as ac-

tion-centered. Four components of action-centered skill are high-lighted in the ex-
periences of these workers:

1. Sentience. Action-centered skill is based upon sentient information de-

rived from physical cues.
2. Action-dependence. Action-centered skill is developed in physical per-
formance. Although in principle it may be made explicit in language, it
typically remains unexplicated-implicit in action.
3. Context-dependence. Action-centered skill only has meaning within the
context in which its associated physical activities can occur.
4. Personalism. It is the individual body that takes in the situation and an
individual’s actions that display the required competence. There is a felt
linkage between the knower and the known. The implicit quality of
knowledge provides it with a sense of interiority, much like physical ex-
perience.

THE DISSOCIATION OF SENTIENCE AND KNOWLEDGE

Computerization brings about an essential change in the way the worker can
know the world and, with it, a crisis of confidence in the possibility of certain
knowledge. For the workers of Piney Wood and Tiger Creek, achieving a sense of
200 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

knowing the world was rarely problematical in their conventional environments.

Certain knowledge was conveyed through the immediacy of their sensory experi-
ence. Instead of Descartes’s “I think, therefore I am,” these workers might say, “I
see, I touch, I smell, I hear; therefore, I know.” Their capacity to trust their knowl-
edge was reflected in the assumption of its validity. In the precomputerized envi-
ronment, belief was a seamless extension of sensory experience.
As the medium of knowing was transformed by computerization, the placid
unity of experience and knowledge was disturbed. Accomplishing work depended
upon the ability to manipulate symbolic, electronically presented data. Instead of
using their bodies as instruments of acting-on equipment and materials, the task
relationship became mediated by the information system. Operators had to work
through the medium of what I will call the “data interface,” represented most vis-
ibly by the computer terminals they monitored from central control rooms. The
workers in this transition were at first overwhelmed with the feeling that they
could no longer see or touch their work, as if it has been made both invisible and
intangible by computer mediation.

It’s just different getting this information in the control room. The man in
here can’t see. Out there you can look around until you find something.

The chlorine has overflowed, and it’s all over the third floor. You see, this is
what 1 mean. . . it’s all over the floor, but you can’t see it. You have to re-
member how to get into the system to do something about it. Before you
could see it and you knew what was happening-you just knew.

The hardest thing for us operators is not to have the physical part. I can
chew pulp and tell you its physical properties. We knew things from experi-
ence. Now we have to try and figure out what is happening. The hardest
part is to give up that physical control.

In a world in which skills were honed over long years of physical experience,
work was associated with concrete objects and the cues they provided. A worker’s
sense of occupational identity was deeply marked by his or her understanding of
and attachment to discrete tangible entities, such as a piece of operating equip-
ment. Years of service meant continued opportunities to master new objects. It
was the immediate knowledge one could gain of these tangible objects that engen-
dered feelings of competence and control. For workers, the new computer-medi-
ated relationship to work often felt like being yanked away from a world that
could be known because it could be sensed.

Our operators did their job by feeling a pipe-“ls it hot?” We can’t just tell
them it’s 150 degrees. They have to believe it.

With computerization I am further away from my job than I have ever been
before. I used to listen to the sounds the boiler makes and know just how it
The Abstraction of Industrial Work 201

was running. I could look at the fire in the furnace and tell by its color how
it was burning. 1 knew what kinds of adjustments were needed by the shades
of color 1 saw. A lot of the men also said that there were smells that told you
different things about bow it was running. 1 feel uncomfortable being away
from these sights and smells. Now 1 only have numbers to go by. 1 am scared
of that boiler, and 1 feel that I should be closer to it in order to control it.

It is as if one’s job had vanished into a two-dimensional space of abstrac-

tions, where digital symbols replace a concrete reality. Workers reiterated a spon-
taneous emotional response countless times-defined by feelings of loss of
control, of vulnerability, and of frustration. It was sharpened with a sense of crisis
and a need for steeling oneself with courage and not a little adrenaline in order to
meet the challenge. It was shot through with the bewilderment of a man suddenly
blind, groping with his hands outstretched in a vast, unfamiliar space. “We are in
uncharted water now,” they said. “We have to control o u r operations blind.” This
oft-repeated metaphor spoke of being robbed of one’s senses and plunged into
darkness. The tangible world had always been thick with landmarks; it was diffi-
cult to cast off from these familiar moorings with only abstractions as guides.
One operator described learning to work with the new computer system in
Tiger Creek’s pulping area. “The difficulty,” he said, “is not being able to touch
things.” As he spoke, his hands shot out before him and he wiggled all his fingers,
as if to emphasize the sense of incompleteness and loss. He continued:

When I go out and touch something, I know what will happen. There is a
fear of not being out on the froor watching things. It is like turning your
back in a dark alley. You don’t know what is behind you; you don’t know
what might be happening. It all becomes remote from you, and it makes you
feel vulnerable. It was like being a new operator all over again. Today 1 push
buttons instead of opening valves on the digester. I f I push the wrong but-
ton, will I screw up? Will anything happen?

Many other descriptions conveyed a similar feeling:

With the change to the computer it’s like driving down the highway with
your lights out and someone else pushing the acceleratox

It’s like flying an airplane and taking all the instruments out so you can’t see.
It’s like if you bad an airplane and you put pieces over each instrument to
hide it. Then, if something went wrong, you have to uncover the right one in
a split second.

Doing my job through the computer, it feels different. It is like you are rid-
ing a big, powerful horse, but someone is sitting behind you on the saddle
holding the reins, and you just have to be on that ride and hold on. You see
what is coming, but you can’t do anything to control it. You can’t steer your-
202 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

self left and right; you can’t control that horse that you are on. You have got
to do whatever the guy behind you holding the reins wants you to do. Well,
1 would rather be holding the reins than have someone behind me holding
the reins.

The feeling of being in control and the willingness to be held accountable re-
quire a reservoir of critical judgment with which to initiate informed action. In the
past, operators like those a t Piney Wood derived their critical judgment from their
“gut feel” of the production process. Becoming a “good” operator-the kind that
workers and managers alike refer to as an “artist” and invest with the authority of
expertise-required the years of experience to develop a finely nuanced, felt sense
of the equipment, the product, and the overall process. With computerization,
many managers acknowledged that operators had lost their ability “to feel the
machine.” Without considering the new skill implications of this loss, many man-
agers feared it would eliminate the kind of critical judgment that would have al-
lowed operators to take action based upon an understanding that reached beyond
the computer system.
Piney Wood’s plant manager, as he presided over the massive technology
conversion, asked himself what the loss of such art might mean:

In the digester area, we used to have guys doing it who had an art. After we
put the computers in, when they went down we could go to manual backup.
People remembered how to run the digesters. Now if we try to go back, they
can’t remember what to do. They have lost the feel for it. We are really stuck
now without the computer; we can’t successfully operate that unit without
it. I f you are watching a screen, do you see the same things you would if you
were there, face-to-face with the process and the equipment? 1am concerned
we are losing the art and skills that are not replenishable.

There were many operators who agreed. In one area of Piney Wood, the
crew leader explained it this way:

The new people are not going to understand, see, or feel as well as the old
guys. Something is wrong with this fan, for example. You may not know
what; you just feel it in your feet. The sound, the tone, the volume, the vi-
brations. , , the computer will control it, but you will have lost something,
too. It’s a trade-off The computer can’t feel what is going on out there. The
new operators will need to have more written down, because they will not
know it in their guts. 1 can’t understand how new people coming in are ever
going to learn how to run a pulp mill. They are not going to know what is
going on. They will only learn what these computers tell them.

Sam Gimbel was a young production coordinator in Piney Wood. Though

trained as a chemical engineer, he had been particularly close to the operators
whom he managed. He had shepherded them through the technology conversion
The Abstraction of Indzrstrial Work 203

and construction of the new control room, and worked closely with them as they
grappled with new ways of operating:

We are losing the context where hands-on experience makes sense. I f you
don’t have actual experience, you have to believe everything the computer
says, and you can’t beat it at its own game. You can’t stand up to it. And yet
who will have the experience to make these kinds of judgments? It will
surely be a different world. You lose the checkpoints in reality to know if
you are doing it right; therefore, how will anyone be able to confront the
computer information?

Piney Wood’s management had approached the technology conversion with

the following message: “We are simply providing you with new tools to d o your
job. Your job is to operate the equipment, and this is a new tool to operate the
equipment with.” Managers repeatedly made statements such as, “We told them
this was a tool just like a hammer or a wrench.” One manager even went so far as
to say, “We hoped they wouldn’t figure out that the terminal we were giving them
was really a computer.”
As experience with the new operating conditions began to accumulate,
many managers began to see that treating the computer system like a physical ob-
ject, “just another tool,” could lead to chronic suboptimization of the technol-
ogy’s potential. A powerhouse worker with over twenty-five years of experience
had developed a special way of kicking the boiler in order to make it function
smoothly. He used the same approach with the terminal; if he hit a certain button
on the keyboard, a particular reading would change in the desired direction, but
he did not know why or how. Piney Wood’s powerhouse manager put it this way:

The guy who kicks the boiler is the same guy who mashes the button a cer-
tain way just to make the line go down. This person will never optimize the
process. He will use too much chemical and too high pressure. H e will never
make you money because he doesn’t understand the problem.

Just as the digester operators had lost their ability to cook manually, other
workers throughout the mill felt equally powerless:

In the old way, you had control over the job. The computer now tells you
what to do. There is more responsibility but less control. We lost a boiler
that was on computer control. Me just had to sit there and stare. We were all
shook up,

Sometimes I am amazed when I realize that we stare at the screen even when
it bas gone down. You get in the habit and you just keep staring even if there
is nothing there.
204 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

Ironically, as managers and operators across the mill watched the level of
artistry decline, the senior technical designers continued to assume that manual
skills would provide the necessary backup to their systems.
The problem was even more acute in Cedar Bluff, where most of the work
force lacked the experience base from which felt sense and critical judgment are
developed. Managers at Cedar Bluff engaged in a quiet debate as to how much of
a problem this lack of experience would ultimately be. On one side of the argu-
ment were the “old-timers”-managers with years of experience in the industry:

I like to smell and feel the pulp sometimes. It can be slick, it can be slimy, it
can be all different consistencies. These are the artistic aspects of making
pulp that the computer doesn’t know about. Some of the operators have
been picking up these aspects, but there are so many numbers so readily ac-
cessible, we have to shortcut it at times and solve more problems from the
office. The information is so good and rapid we have to use it. . . . You have
got to be able to recognize when you can run things from the office and
when you have to go and look. Yet, I recognize that I a m not as good a pulp
maker as the people who trained me, and the new operators are not as good
as I am. They are better managers and planners. I am very happy with the
new managers, but not with the new pulp makers.

The younger engineers, schooled in computer-based analytic techniques, had

little patience with anxious laments over the loss of the art of pulp making. They
were relentlessly confident that a good computer model could reproduce anything
that operators knew from experience-only better. Here is how the process engi-
neers articulated the argument:

Computer analysis lets us see the effects of many variables and their interac-
tions. This is a picture of truth that we could not have achieved before. I t is
superior to the experience-based knowledge of an operator. You might say
that truth replaces knowledge.

People who have this analytic power do not need to have been around to
know what is going on. All you need is to be able to formulate a model and
perform the necessary confirmation checks. With the right model you can
manage the system just fine.

Most Cedar Bluff managers agreed that the computer system made it possi-
ble to d o a better job running the plant with an inexperienced work force than
otherwise would have been possible, though some wondered whether the levels of
expertise would ever be as high as among workers with hands-on exposure to the
pulping process. Yet even as managers argued over the essentiality of action-cen-
tered skill, technology was irreversibly altering the context in which the operators
performed. The opportunities to develop such skills were becoming increasingly
rare as the action context was paved over by the data highway.
The Abstraction of Industrial Work 205

Many of Cedar Bluff’s managers believed that the traditional knowledge of

the pulp mill worker would actually inhibit the development of creativity and
flexibility. Under the new technological conditions, the young operators would
develop their capacity to “know better” than the systems with which they worked
as they struggled with the complexities of the new technology and the data it pro-
vided. The data interface would replace the physical equipment as the primary
arena for learning.
Yet as months passed, other managers observed a disturbing pattern of inter-
actions between the operators and the computer system. Some believed that the
highly computerized task environment resulted in a greater than usual bifurcation
of skills. One group of operators would use the information systems to learn an
extraordinary amount about the process, while another group would make itself
an appendage to the system, mechanically carrying out the computer’s directives.
These managers complained that the computer system was becoming a crutch that
prevented many operators from developing a superior knowledge of the process.
One “old-timer” provided an example:

When there is a shift change and new operators come on, the good operator
will take the process from the computer, put it on manual, make certain
changes that the operator thinks are necessary, and then gives it back to the
computer. The average operator will come in, see this thing on automatic
control, and leave it with the computer. Sometimes that operator won’t even
realize that things are getting bad or getting worse. They should have known
better, but they didn’t.

Most Cedar Bluff operators spoke enthusiastically about the convenience of

the computer interface, and some freely admitted what they perceived to be a de-
pendence on the computer system:

The computer provides your hands. I don’t think I could work in a conven-
tional mill. This is so much more convenient. You have so much control
without having to go out to the equipment and adjust things.

We can’t run this mill manually. There are too many controls, and it is too
complex. The average person can only run four or five variables at once in a
manual mode, and the automatic system runs it all. I f the computer goes
down, we have to sit back and wait. We sit and we stare at the screens and
we hope something pipes in.

Many managers observed with growing alarm the things that occurred when
operators neither enjoyed the traditional sources of critical judgment nor had de-
veloped enough new knowledge for informed action.

In a conventional mill, you have to go and look at the equipment because

you cannot get enough data in the control room. Here, you get all the data
206 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

you need. The computer becomes a substitute tool. It replaces all the sensual
data instead of being an addition. We had another experience with the feed-
water pumps, which supply water to the boiler to make steam. There was a
power outage. Something in the computer canceled the alarm. The operator
had a lot of trouble and did not look at the readout of the water level and
never got an alarm. The tank ran empty, the pumps tripped. The pump fi-
nally tore up because there was no water feeding it.

We have so much data from the computer, I find that hard drives out soft,
Operators are tempted not to tour the plant. They just sit at the computer
and watch for alarms. One weekend I found a tank overflowing in digest-
ing. I went to the operator and told him, and he said, “It can’t be; the com-
puter says my level is fine.” I am afraid of what happens if we trust the
computer too much.

At least since the introduction of the moving assembly line in Ford’s High-
land Park plant, it has been second nature for managers to use technology to de-
limit worker discretion and, in this process, to concentrate knowledge within the
managerial domain. The special dilemmas raised by information technology re-
quire managers to reconsider these assumptions. When information and control
technology is used to turn the worker into “just another mechanical variable,”
one immediate result is the withdrawal of the worker’s commitment to and ac-
countability for the work, This lack of care requires additional managerial vigi-
lance and leads to a need for increased automatic control. As this dynamic
unfolds, it no longer seems shocking to contemplate an image of work laced with
stupefaction and passivity, in which the human being is a hapless bystander at the
margins of productive activity. One young operator in Cedar Bluff discussed his
prior job as a bank clerk. I asked him if his two employment experiences had any-
thing in common. “Yes,” he said, “in both cases you punch the buttons and watch
it happen.’’
As automation intensifies, information technology becomes the receptacle
for larger and larger portions of the organization’s operating intelligence. Algo-
rithms become the functional equivalent of a once diffuse know-how, and the ac-
tion context in which know-how can be developed and sustained vanishes.
Because many managers assume that more technology means a diminished need
for human operating skill, they may recognize the waning of worker know-how
without becoming concerned enough to chart a different course. Left unchal-
lenged, these systems become more potent, as they are invested with an escalating
degree of authority. Technical experts temporarily serve as resources, but once
their knowledge has been depleted, and converted into systematic rules for deci-
sion making, their usefulness is attenuated. The analysts and engineers, who con-
struct programs and models, have the capacity to manipulate data and,
presumably, to make discoveries. Ultimately, they will become the most important
human presence to offer any counterpoint to the growing density and opacity of
thea.ummedsy-.
The Abstraction of Industrial Work 207

NOTES
1. It should be noted that there are other motivations that could account for an opera-
tor’s inarticulateness in the face of such questioning. Operators, like generations of
craftspeople before them, know that as their activities become more explicit, their
skills seem less significant. Explication means a loss of power. However, my work in
this mill over several years led me to believe that although many operators were aware
of these political dynamics, they tended to choose methods of resistance and counter-
offense other than deliberately undermining the process engineer’s efforts. In most
cases I was convinced that operators were not withholding information but, rather,
that they had really reached the limits of their explicit understanding.
This page intentionally left blank
 Networks
Wayne Baker

To manage a business well is to manage its future;

and to manage the future is to manage information.
Marion Harper’

Former President George Bush was much better off without John Sununu. When
Sununu was Bush’s chief of staff, he would indulge in a favorite pastime of subor-
dinates: intercepting information before it could get to the big boss.2 Helpful
John, it seems, would stop letters sent by the president’s longtime advisors and
keep them from ever reaching the Oval Office. He simply answered the letters
himself or referred George’s advisors to junior aides. George was the last to know.
To get around his chief of staff, George had to establish a back channel-a private
post office box in Maine-to which his advisors could send letters directly.
Bush’s problem wasn’t unusual. And it’s not limited to the chief executive;
people at all levels suffer from it. The problem just gets worse and worse the
higher up you go or the farther away you move from the company’s actual opera-
tions. The dilemma of indirect management means you rely more and more on
information fed to you by others. You get out of touch. And you make poor
decisions.
Managers can be their own worst enemies. As they move up, many fall for
the management folklore that says they should withdraw from operational affairs
and contemplate the big picture. “If he [or she] follows the advice to free himself
from operations,” warns Ed Wrapp in his famous Harvard Business Review arti-
cle, “he [or she] may soon find himself subsisting on a diet of abstractions, leaving
the choice of what he [or she] eats in the hands of his [or her] subordinate^."^
What can you do? You can’t rely on the formal organization to help you out.
“The very purpose of a hierarchy,” says Kenneth Boulding, “is to prevent infor-

From Chapter 3 of Networking Smart. Copyright 0 1994 McCraw-Hill, Inc. Reprinted with permis-
sion of McGraw-Hill, Inc.

209
210 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

mation from reaching top layers. It operates as an information filter, and there are
little wastebaskets all along the way.”4 Information is garbled as it wends its way
through formal channels; often it arrives too late to do any good. Conflicting and
ambiguous messages create a dull buzz of confusion. Critical details are lost in ag-
gregated summaries. At worst, reports are sanitized, numbers massaged; informa-
tion is hoarded and withheld, even fabricated.
To get the information you need, you have to take matters into your own
hands. Richard Neustadt discovered an important lesson for managers every-
where when he studied the information-collecting habits of Presidents Roosevelt,
Truman, and Eisenhower: “It is not information of a general sort that helps a
President see personal stakes; not summaries, not surveys, not the bland amal-
gams. Rather, . . it is the odds and ends of tangible detail that pieced together in
his mind illuminate the underside of issues put before him. To help himself he
must reach out as widely as he can for every scrap of fact, opinion, gossip, bearing
on his interests and relationships as President. He must become his own director
of his own central intelligence.’’s
That’s what you can do. To get the information you need, you have to build
your own intelligence network. Good managers and executives have always done
so, but it’s more important today than ever, no matter where you are in the organi-
zation. Knowledge, says futurist Alvin Toffler, is now the basis of power and
wealth creations6We are in the information age. By the mid-1980s, for example,
more than half of the U.S. work force already held jobs that were information-re-
lated.’ There’s a prodigious increase in the production of information; fast mar-
kets and rapid technological change make today’s information more perishable
than ice at the equator.
This information explosion aggravates the manager’s intelligence problem.
The information explosion is really the data explosion, says Richard Saul Wur-
man in Information Anxiety, and you have to sort and process data to get any-
thing usefuLs Just getting more data won’t help. In fact, psychologists have
learned that more data actually hinders good decision making because it makes
decision makers feel over~onfident.~ What you need is the right information, not
more data.
In this chapter, we look at how effective leaders at all levels get the informa-
tion they need. I describe how personal intelligence systems work and tell you
how to build them. Even if you’re a seasoned and successful businessperson, there
are many things to learn. And, as you’ll see, the information theme of this chapter
is woven throughout the topics covered in the rest of the book. Managing infor-
mation well is an essential part of managing relationships and networks in all
business areas.

YOUR INTERNAL INTELLIGENCE NETWORK

Effective leaders develop their own independent intelligence networks to
keep informed about a wide range of decisions, activities, people, and events.
Their personal networks help them to monitor ongoing activities and to spot in-
Building Intelligence Networks 211

cipient problems and opportunities. “[Elach of my heroes,” says Ed Wrapp, “has

a special talent for keeping himself [or herself] informed about a wide range of op-
erating decisions being made at different levels in the company. As he [or she]
moves up the ladder, he [or she] develops a network of information sources in
many different departments. He [or she] cultivates these sources and keeps them
open no matter how high he [or she] climbs in the organization. When the need
arises, he [or she] bypasses the lines on the organization chart to seek more than
one version of a situation.”I0
Consider Walter Wriston, the corporate leader who transformed First Na-
tional Citibank into Citicorp-one of the preeminent financial institutions in the
world.” Wriston was a master builder of informal intelligence networks. As a fel-
low executive described in Harry Levinson and Stuart Rosenthal’s chapter on
Wriston in CEO:

He relies very heavily on information that comes to him from different parts
of the organization. He gets and absorbs the feelings of people who are not
only department and group heads but always has a wary ear open to be alert
to situations, circumstances within the shop that might not surface in the or-
dinary routine of management information flows. . . . He3 got a very acute
sense. . . of the ideas as they float around the organization. . . .

Wriston understood very well how easy it is to get out of touch. When Jack
Welch took the top job at GE, he reports in an interview with Financial World
that Wriston warned him of this common problem: “Jack, remember one thing,
you’re always going to be the last one to know the critical things that need to be
done in your organization. Everyone else already knows.” “He was right,” says
Welch.I* Wriston’s advice applies to any one, not just those at the top. Without in-
telligence networks, you’re always out of touch, out of the swim of things, out of
the loop.
Personal contacts are your direct lines of communication with the various
parts of the organization. Building your intelligence network means initiating, cul-
tivating, and maintaining these contacts. You have plenty of opportunities to d o
so in the course of day-to-day activities. “Managing by Wandering Around”-
Tom Peters and Robert Waterman’s MBWA principle-offers chance encounters
that yield vital information.l3 Merck CEO Roy Vagelos, for example, often takes
lunch in the company cafeteria so he can talk informally with scientists.14 It’s a
great way to get informed of the latest developments and discoveries. In a similar
fashion but a different content, then-governor Bill Clinton regularly visited a local
Little Rock McDonald’s to sit and chat with people and hear their concern^.'^
Mobility and movement offer you chances to make personal contacts. Any
move you make inside the company-promotions, transfers, temporary details,
special projects, committee assignments, relocations, stints in foreign offices-
provides great opportunities to develop new information contacts. A pending re-
location or reassignment might look brighter if you consider its network-building
potential. It might be just the right move if it enables you to meet different people,
make new contacts, and build new relationships that’ll be helpful down the road.
212 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

As a rule, diverse contacts are better than similar or redundant contacts. A

large, diverse network of contacts, argues network expert Ronald Burt, gives you
the best access to information.16You can’t stay in the know with a few or a nar-
row set of contacts. With a diverse set, however, you’re better able to quickly dis-
cover new opportunities. And broader access to information has its rewards. Burt
shows in his impressive study of managers in a high-tech Fortune 500 company
that those who bridge lots of diverse groups get promoted faster and a t younger
ages than their peers.
Zigzag career paths are better than linear paths for building the diverse set
of connections you need. Vertical paths-moving up rung by rung within the same
function-help you build contacts in the same business area. These contacts are
useful, of course, but your intelligence network is a mile deep and an inch wide.
Zigzag job changes-moving laterally, jumping functions, going abroad-provide
the breadth you need. The Japanese have known for a long time about the net-
work-building benefits of zigzagging. That’s why Japanese managers have what
look to us like inefficient, meandering career paths. All the changes you make,
however, are just opportunities to build information contacts. You must seize
them and actively cultivate contacts. . . .
When I talk to businesspeople about building personal intelligence net-
works, many raise an objection. You ought to have more respect for the formal
hierarchy, they tell me; you shouldn’t advise people to go around their subordi-
nates to get to the bottom of a story. When I press them for details, they usually
tell me about superiors who circumvent them and go straight to the source. They
feel undermined, frustrated, caught in the middle. All too often their feelings are
justified-many managers throw their weight around and gather intelligence in
cavalier and callous ways. If you use your personal intelligence network this way,
you’re asking for trouble. Your aggrieved subordinates will search for some way
to thwart you, and your intelligence network will crumble.
Managing information relationships, just like any sort of relationship, en-
tails ethics and responsibilities. You may need to go around people to get informa-
tion, but those you go around need to be taken into account. Sometimes just
explaining why you go directly to an information source is all you need to do.
When appropriate, share what you find out. Information dissemination is part of
your job. Sharing includes confidential information, though you have to d o so ju-
diciously. “The manager is challenged,” says management expert Henry Min-
tzberg, “to find systematic ways to share privileged information.”” With your
diverse set of contacts, you’re in a great position to piece together information
that can help others do their jobs. Remember that sharing information empowers
people. A person’s power-the ability to get the job done-depends directly on his
or her access t o information.18 You empower your people-and yourself-by col-
lecting and sharing information.
Your responsibility to share information extends to your information
sources. Reciprocity, the natural give-and-take between people, is one of the basic
rules in all cultures and societies.19 It’s essential for building information relation-
ships (or any other kind of relationship for that matter). The reciprocity rule links
Building Intelligence Networks 213

a future action (repayment)with a present action (a favor, gift, etc.). By linking the
present and the future, it activates the fourth networking principle: Repeated in-
teraction encourages cooperation. If you sponge information and never give, your
sources will dry up. But give and you shall receive. This doesn’t mean you should
be an inveterate gossip. Be a tactful, judicious supplier of information and a trust-
worthy, responsible user of information.
Above all, take what you learn from your personal intelligence system with
a grain of salt. You must work hard to gather every tidbit of data, gossip, rumor,
and innuendo. Some tidbits are the most timely and accurate pieces of genuine in-
formation you’ll ever get.20A few help you peer into the future. But many pieces
of information are irrelevant, others are innocent errors, and more than a few are
deliberate disinformation and malicious lies. All, however, are parts of some larger
story. You’re faced with a balancing act. If you wait until you get the whole story,
it’ll be too late to act; but if you don’t get enough of the story before you act,
you’ll make bad decisions, alienate your people, and let your organization be
ruled by the tyranny of rumor and gossip.

DONUTS WITH DITCH

Donuts with Ditch? OK, here’s a more formal-sounding title: information-
exchange forum. Donuts with Ditch is the informal communication sessions Allan
Ditchfield created ten years ago a t A T ~ C T .It~ ’was so effective and popular that he
imported the practice to MCI when he became chief information officer (CIO) of
the long-distance communications company, and then imported it once again
when he became CIO of Progressive Corporation, the Ohio-based national auto
insurance company. “I d o it to break down the hierarchical barriers,” he told me.
“It’s basically a communications meeting.” Allan uses these informal meetings to
hear everyone’s concerns, get and give information, and keep everyone in touch.
It’s a great way to get at the real issues, the real problems. It gives people a voice
and a forum, many of whom have no other means to communicate with or hear
from upper-level managers.
Allan holds a Donuts-with-Ditch session at Progressive every two weeks.
(He would hold weekly sessions at MCI because the department was much big-
ger.) The typical meeting lasts about 2 or 3 hours. Attendance is limited to a small
number of people. “I have no more than 10 people at a time,” he says. “If you
have more than 10, people don’t like to talk.” People are chosen on a random ba-
sis for Donuts with Ditch, “but some people ask to be invited, especially if they
have a burning issue to discuss.” Trust is the most important ingredient for suc-
cess. “I have a rule-it’s a sacred open door-that there will be no retaliation, no
one’s going to be hurt by [what they say]. I don’t tell management.”
How has Donuts with Ditch changed over the years? “I have a lot more fruit
these days,” he says.
Donuts with Ditch is a great example of creating conditions that encourage
information flow. Donuts with Ditch comes in all shapes and sizes:
214 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

American Airlines CEO Robert Crandall holds conferences around the

country to talk directly with employees.22
Federal Express uses its Open Door policy to encourage employees to
communicate directly to management their ideas, questions, or comments
about the industry or company; the express courier’s Guaranteed Fair
Treatment policy makes sure employees can get a fair hearing for any
concern or complaint about fair treatment (e.g., questions about a per-
formance review or other personnel matter).23
Royal Bank of Canada establishes conferences to help area managers
share expertise, data, and i n t e l l i g e n ~ e . ~ ~
GE uses its Corporate Executive Council as a forum for leaders from GE’s
various businesses to get together, exchange information and advice, and
to integrate what could be disparate units.2”
The U.S. State Department uses the secretary’s open forum to encourage
“differences of opinion by publishing papers, sponsoring discussions, and
inviting critics to speak.”26State also has a “dissent channel” that lower-
level managers can use to send messages directly to the secretary, bypass-
ing the Sununu-type blockade I described earlier.
Donuts with Ditch, Japanese style, is the regular Friday meeting (kinyo-
kai) of the presidents of the member companies in the Mitsubishi
keiretsu, one of Japan’s vast and close-knit groups of companies. The Fri-
day meeting has taken place every month for over forty years!27

Regular operational meetings are another effective variation on the Donuts

with Ditch theme. Such meetings are common at Silicon Valley firms, which are
role models of the new network organization. Zap Computers, for example,
Kathleen Eisenhardt’s code name for the large computer maker she studied, relies
on frequent operational meetings-two or three a week-to share, relay, and dis-
cuss information about sales, inventory, backlog, engineering schedules, new re-
leases, product introductions by competitors, technical developments in the
industry, and so on.2xZap’s top managers use these meetings to relay information
they glean from constant phone calls, travel, and business and university contacts.
Lots of meetings, face-to-face interaction-all the variations of Donuts with
Ditch-create real-time intelligence networks. Real-time networks help managers
accelerate decision making, says Eisenhardt, which is essential wherever the pace
of change is fast and furious.29

CAN COMPUTER NETWORKING HELP?

Computers are often heralded as the technological cure to the information
problem. In a computer-networked company, for example, you c m find the pro-
verbial needle in a haystack. With electronic mail or an electronic bulletin board,
you can query hundreds or even thousands of people all at once. You’d get a few
wrong answers, but the odds of getting the right one are tremendous when com-
pared with the luck you’d have with the telephone.
Building Intelligence Networks 215

Searching the haystack is just one example of the many ways in which com-
puter technology is revolutionizing c o m m u n i ~ a t i o n Computer
.~~ technologies free
us from the friction of time and space. In 1991, for example, more than 5.5 million
people worked as “telecommuters,” commuting to work via phone and modem in-
stead of a car. This is a 38 percent increase over the previous year, according to Link
Resources National Work at Home S ~ r v e y . ~The ’ new advances in networking
technologies can help transform and redesign traditional organizations, yielding
quicker, more responsive, more effective decision making.32Flexible computer net-
working is often used to support the new network organizations.
But there are serious limitations. Despite great advantages, electronic com-
munication cannot become the cornerstone of your independent intelligence sys-
tem. To see why, let’s begin by looking at the older computer technologies,
management information systems (MIS). Part of the folklore about managers, ar-
gues Henry Mintzberg, is that senior managers need aggregated information like
that produced by MIS. The fact is good managers eschew formal systems that
spew forth abstracts and stylized facts.33 Instead, they obtain and transmit infor-
mation via all sorts of verbal media: face-to-face conversations, phone calls, trav-
eling, meetings, spontaneous and impromptu discussions. “I was struck during
my study,” Mintzberg said, “by the fact that the executives I was observing-all
very competent-are fundamentally indistinguishable from their counterparts of a
hundred years ago (or a thousand years ago). The information they need differs,
but they seek it in the same way-by word of mouth.”34
Why d o managers prefer word of mouth? MIS provides only aggregated in-
formation and old news. Word-of-mouth networks are on-line, real-time systems
that give you live, rich, quick, timely information about what’s going on. You
need this kind of information to get news as it happens, to spot opportunities and
problems early. “Every bit of evidence,” says Mintzberg, “suggests that the man-
ager identifies decision situations and builds models [mental maps of the organi-
zation] not with the aggregated abstractions an MIS provides, but with specific
tidbits of data.”3S
Is computer networking better? In some ways, yes. Greater efficiency is one
of the benefits companies find when they establish electronic networks: shorter
elapsed time for transactions, quicker turnaround, f,aster group c o m m ~ n i c a t i o n . ~ ~
Saving time is important, of course, but electronic networks aren’t the appropriate
medium for the complex, rich, nuanced communication essential for intelligence
gathering. “Proponents of the efficiency benefits of computer-based communica-
tion often assume that it delivers the same message as any other medium but sim-
ply does so more rapidly,” write networking experts Lee Sproull and Sara Kielser
in Connections. “That view is misleading because a message-even the same ‘mes-
sage’-changes its meaning depending on the forum within which people convey
it.” Compared with face-to-face interaction, they say, “today’s electronic technol-
ogy is impoverished in social cues and shared e~perience.,’~’
Consider the results of a study by James McKenney and associates at Har-
vard Business School on how managers use electronic mail versus face-to-face
c o m m ~ n i c a t i o n They
. ~ ~ discovered conspicuous differences in how managers use
these media:
216 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

Managers use electronic mail for efficient communication in well-defined

contexts-monitoring task status, coordinating efforts, exchanging fac-
tual information, sending alerts, and broadcasting information.
Managers use face-to-face interaction for defining and discussing prob-
lems and solutions, building a shared understanding of the situation, dis-
cussing shifting priorities and external pressures, interpreting ambiguous
signals, and socialization of members.

Electronic communication is efficient, but only face-to-face interaction pro-

vides the richness, interactivity, immediacy of feedback, and social context needed
for complex problem solving and fostering a shared set of values, beliefs, and
meanings. Former ITT CEO Harold Geneen recognized the difference: “In New
York, I might read a request and say no. But in Europe, I could see that an answer
to the same question might be yes . . . it became our policy to deal with problems
on the spot, f a ~ e - t o - f a c e . Even
” ~ ~ Allan Ditchfield, who as chief information offi-
cer is a champion of electronic communication, relies on informal, face-to-face
sessions via his regular Donuts with Ditch.
It is difficult (if not impossible) to initiate or nurture meaningful relation-
ships via electronic interaction. (It’s true that some people who meet electronically
become friends and even marry, but these are exceptions.) The ritual and ceremo-
nial value of group meetings cannot be simulated with electronic communica-
tionS4OGroup meetings, says Stanford University professor Jane Hannaway, “keep
a sense of community alive in the organization , . , [and] affirm the place of the
individual and others in that community. . . .”41
Computer networking can save you a lot of time communicating routine in-
formation. But such networking is ill-suited to the rich, sensitive, live, private in-
formation you need to be an effective manager. For that, you must build an
independent intelligence network based on personal contacts.

YOUR EXTERNAL INTELLIGENCE NETWORK

Along with internal intelligence networks, effective managers build external

networks of personal contacts. External contacts help you stay informed about
what’s going on “out there”--changing customer preferences, a competitor’s
plans and actions, social and economic trends, pending regulations, emergent
technologies, and so forth. One reason Ned Tanen of Paramount Pictures was so
effective as president of production, for example, was his “dozen Rolodexes of
contacts.”42A colleague who studies the publishing business told me a story about
an editor who returned to his office to find his desk taped shut. It seems the pub-
lishing house had been taken over abruptly, and the new owners were feeling a
touch protective about their new assets. He spotted his rolodex on the desk,
snatched it, and walked away saying, “That’s all I need.”43
External intelligence networks help you manage the information problem,
the problem of too much data and not enough real information. A few years ago,
I was talking with the treasurer of a global high-tech firm about his relationships
Building Intelligence Networks 217

with investment banks. I commented on the foot-high pile of proposals and bro-
chures on his desk. “It’s a problem coping with it all,’’ he said. “I’ll look at it even-
tually; there may be a nugget buried in there. But I don’t have time to go through
it now.” He then told me how he finds out what he really needs to know: the com-
pany’s senior vice president of finance. “He’s talking with Goldman Sachs, Salo-
mon Brothers, Merrill Lynch, Morgan Stanley, and who all else,” the treasurer
said. “That’s what he’s paid for-his network of contacts.” External contacts pro-
vide the right information.
“External networks are reality checks,” explains Howard Haas, who was
CEO of Sealy Inc. for 19 years. “How d o you know the numbers you’ve got in
your own company are correct? You don’t know unless you have some way to
prove it. Or, I know how my business is, but I don’t know how my business is in
relation to somebody else’s business. How d o I find out about that? I say to a com-
petitor, ‘How’s your business?’ He says, ‘Terrific! We’re 35 percent ahead!’ I know
the guy’s lying through his teeth. So I go out and I call a bunch of dealers and talk
to a lot of people. And finally I put together a consensus of the information and
say, well, ours is up 5 percent so we’re doing a lot better. That’s a reality
People throughout your organization have special access to the outside
world-and it pays to link up with them. “The MIS professional,” says Tom Pe-
ters, “will be the first to hear that a competitor is developing a sweeping new elec-
tronic linkup with hundreds of major c u ~ t o m e r s . ’ Your
’ ~ ~ people in finance have
contacts throughout the financial world. Your sales representatives meet your
competitor’s sales reps every day in customer waiting rooms or a t trade shows.
They swap stories and gossip (and a few fibs, no doubt!) about you, each other,
customers, other competitors, the industry, and so on. Do you have a way of
learning what they know? Your scientists and engineers have innumerable con-
tacts in professional and learned societies; they always know what’s going on,
what’s hot, what’s new. Do you talk with them? Remember that everyone in your
organization has a life outside the company; everyone can be the company’s eyes
and ears. Recruit them into your external intelligence system.
Just as the daily rounds of business help you make internal contacts, your
everyday work life offers plenty of opportunities to make external contacts. Most
people think Peters and Waterman’s MBWA principle applies only inside the com-
pany. But it’s just as useful for making outside contacts. Peters advises marketers,
for example, to spend more time “hanging out” in the m a r k e t p l a ~ e It’s
. ~ ~an idea
that Japanese companies have practiced for years. The invention of the Sony
Walkman is a great example.47The original idea came right from Sony CEO and
chairman Akio Morita. But he didn’t dream it up sitting in his office. Morita spent
lots of time hanging out, observing young people and getting to know their life-
styles and tastes in music. And what he saw was a huge untapped demand for a
personal, compact, portable tape player. That’s what he conveyed back to Sony
designers. The result, as you know, was a smashing success.
Recently, GE instituted the hanging-out principle in its QMI-quick market
intelligence technique.48 QMI is an excellent way to get fast market intelligence
and make quick decisions in response. As GE describes in its 1992 Annual Report,
“Quick Market Intelligence is our term for the magnificent boundary-busting
218 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

technique pioneered by Wal-Mart that allows the entire Company to understand,

to sense, to touch the changing desires of the customer and to act on them in al-
most real time.”49 QMI taps the knowledge and insights of people in the field,
those who see, touch, and listen to the market daily. Once a week, GE salespeople
and managers come in from the field and report what they’ve learned. “It is a
process that gives every salesperson direct access, every Friday, to the key manag-
ers and the CEO of the business, to lay out customer problems and needs. The
product of the meeting is not deep or strategic in nature, but action-a response
to the customer right away.”s0That’s networking smart.
Virtually any outside event or gathering-professional meetings, trade con-
ferences and shows, business roundtables, civic activities, charitable work, and so
on-provide tremendous network-building opportunities. Michael Mach, CEO of
Capital Partners, a very successful commercial real estate development firm, put it
this way: “ I involve myself in a number of things. Most of them are directly or in-
directly related to our business. One reaches the point where one does not deliber-
ately go out and search for things that help one’s business. But the important
factor is the networking that we all d o when you take on one of these other chal-
lenges. I take on civic challenges in order to get some form of extra edge in our
business, since they are interrelated in some way.”

CUSTOMERS AND SUPPLIERS AS EYES AND EARS

Customers and suppliers are some of your best contacts in the outside
world. Their networks reach into different nooks and crannies; they have different
perspectives; they’re steps closer to vital news as it breaks. Your customers, for ex-
ample, are called on regularly by your competitors. Competitors offer them all
sorts of enticements to woo them away from you-new and innovative ideas, spe-
cial promotions, new marketing strategies, As a result, your customers almost al-
ways know before you d o what your competitors are up to. Of course, finding out
what your customers know is a delicate matter. There are no easy rules. You don’t
want to put them in an uncomfortable or compromising position; you need to re-
spect their other relationships. * Your customers can’t always tell you everything
you want to know, but they often can let you know what’s in the wind.
Using your customers’ eyes and ears is critical if you don’t deal directly with
the final consumer. If your customers are a link in the chain leading to the final
consumer, they are closer than you to information you need. In the soft drink in-
dustry, for example, Pepsi, Coca Cola, and other concentrate producers sell most
of their output to the bottlers. Bottlers sell to retail stores, who in turn sell to you
and me. Because the bottlers stock local retail shelves, they know the nitty gritty
details about local markets. Is a competitor making special deals to grab more

*You also don’t want to solicit price lists or other pricing information which could be construed as
attempted price fixing.
Building Intelligence Networks 219

shelf space? Is there a last-second change in a competitor’s marketing policy? Is a

new brand or flavor appearing on the shelves?
Your suppliers are also excellent sources of outside information. Former
Sealy CEO Howard Haas, for example, tapped advertising and media types to
garner information about competitors. “The people [who] sell us space in the
trade papers,” he told me, “are the greatest contact people in the world. I would
pump them as to what was happening with my competitor^."^' A Midwestern
auto dealer uses suppliers’ eyes and ears in a very creative way to get customer
feedback.52 He provides free taxicab service for customers who need a ride after
dropping off their cars. The cab drivers (his suppliers) converse with riders about
their car troubles and the auto dealer’s service. The drivers relay this critical infor-
mation back to the auto dealer.
Lawyers, accountants, bankers, management consultants, advertising agen-
cies, architects, and engineers are all prime information sources. These people sit
at the crossroads of complex information flows. By virtue of a diverse client base,
they are vast repositories of data. They can provide invaluable information and
insights about ongoing developments, events, and business trends. Such informa-
tional benefits add a new twist to outsourcing decisions. Many companies try to
save money by dropping outside service suppliers and doing it themselves. You
can save money this way. The cost of in-house lawyers, for example, is 40 percent
lower than the cost of using an outside law firm for the same workqS3What is not
included in these cost calculations, however, is the opportunity cost of lost infor-
mation: the value of information received from outside suppliers. The lost infor-
mation can be critical and irreplaceable. Cutting relationships severs information
links with the outside world.
You can also incorporate professional intelligence suppliers into your exter-
nal intelligence system. Chicago-based CombsMoorhead Associates, Inc., for ex-
ample, is a professional intelligence-gathering service that produces analyses for
its clients on such subjects as product trends, environmental issues, sales projec-
tions, competitor information, and patents.‘ The firm systematically combs an
array of information sources-industry and government contacts, trade associa-
tions, the media, computerized databases, and so on. Companies like this special-
ize in building information networks, and you tap their extensive networks when
you employ them.

HIGH-LEVEL EYES A N D EARS

Your company’s board of directors is a set of high-level links to the outside

world. One of the best-documented findings in organizational research is that top

“The principals of the firm, Richard E. Combs and John D. Moorhead, have written The Competi-
tive Intelligence Handbook (Metuchen, N.J., Scarecrow Press, 1992) which descrihes competitive in-
telligence techniques.
220 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

executives and managers use corporate boards to collect critical information from
the company’s environment. In my study of 1530 companies and their investment
banks, for example, I found that many companies invite investment bankers on
their boards to gain access to financial advice, ideas, new product developments,
investor attitudes, and market inte1ligen~e.j~ Northwestern Business School pro-
fessor Gerald Davis discovered that executives learn the ins and outs of poison
pills, golden parachutes, and other tricks from their outside directors.5sIn fact, he
learned, if your outside directors already have experience with these tactics, then
your company is much more likely to use them as well. It’s easy to see why. Your
outside directors have direct experience and can tell you exactly how and when to
use them.
Because directors are used as high-level eyes and ears, the composition of a
company’s board can tell you a lot about the kind of environment the company
operates in and what information executives feel is important to get. Hospitals,
for example, put local community and civic leaders on their boards because they
operate in highly politicized environments. Community and civic leaders help
them gather the specific kind of political intelligence they need. If financing is es-
pecially important, companies populate their boards with commercial bankers,
investment bankers, and insurance executives (insurance companies are big
lenders).
Some companies invite executives from key customers or suppliers to sit as
board directors. * These high-level links can provide information about the out-
side director’s company that helps coordinate and improve the relationship. . . .
Other companies invite executives from critical industries to serve on their
boards. This kind of link provides unbiased information about events, trends, and
happenings in the industry.j6
Trade associations and lobbying organizations can also be your high-level
eyes and ears. Sociologists Edward Laumann and David Knoke report a great
story about how this can When the executive director of a petroleum-in-
dustry trade association spotted a Federal Register announcement by the Federal
Aviation Administration (FAA), he immediately realized the danger to his member
firms. The FAA wanted to require the filing of detailed flight plans of noncom-
mercial aircraft as a way to help find downed planes. Under the Freedom of In-
formation Act, however, companies could obtain competitors’ flight plans and
learn where they were exploring for natural resources. The executive director
quickly alerted the membership and mobilized efforts to exempt member
organizations.

*Of course, one must be careful to avoid so-called tying arrangements, wherein the sale of one prod-
uct or service is tied to the sale of another. A tying arrangement would occur, for example, if a com-
puter maker hired an investment bank on the condition that the bank agrees to buy the computer
maker’s products.
Building Intelligence Networks 221

WHAT YOU CAN DO NOW

What you can do now is assess the current state of your personal intelligence
system. Are you in the know, or are you the last to know? Do you get news in time to
act? Or are you surprised time and time again when decisions are made that affect
you and your group? In short, are you the director of your own intelligence network?
The following short quiz will help you assess the current state of your intel-
ligence system. Answer each question yes or no depending on which answer best
describes your situation. To score the quiz, simply circle the number of each ques-
tion to which you answer yes. Count the number of questions that you answered
yes. If you answered yes to 15 or more questions, your personal intelligence sys-
tem's in good shape (though there's always room for improvement!). If you an-
swered yes to 1 0 or more but fewer than 15, you have substantial room for
improvement. And if you answered yes to fewer than 1 0 questions, well, you've
some work to do.

Evaluating Your Intelligence System

1. Do you feel that you're generally 'in the know' and typically find out about key decisions,
events, and activities inside the organization?
2. Have you stayed in touch with operations as you've moved up?
3. Have you maintained your contacts in other groups or departments as you've moved
around the company?
4. Do you regularly supplement reports from management information systems (MIS) with in-
formal word-of-mouth information?
5. Do you prefer to talk face-to-face to define and discuss complex problems and shifting
priorities?
6. Do you usually accept contact-building opportunities-transfers, temporary details, com-
mittee assignments, relocations?
7. Have you (or would you) accept an assignment abroad?
8. Do you have personal contacts in a wide range of different groups (as opposed to con-
tacts concentrated within the same group)?
9. Do you share actively information with your subordinates, peers, and superiors?
10. Do you provide information to your sources? Do you reciprocate?
11. Have you developed real-time intelligence networks, such as your own version of Donuts
with Ditch?
12. Do you know people in other groups or departments that have special access to external
information that would be useful to you?
13. Do you maintain contacts with lawyers, accountants, bankers, consultants, advertising
agencies, and other outside sources?
14. Do you use professional information suppliers?
15. Do you stay in close touch with your customers (including final consumers)?
16. Do you tap suppliers as sources of information?
17. Have you developed fast internal channels to transmit information gathered from outside
sources?
18. Do you regularly 'wander around' in the outside world, attending trade shows, meetings,
civic and charitable events, and so on?
19. Do you know and talk with your peers in other organizations?
20. Do you use board directors, lobbying organizations, and trade associations as high-level
eyes and ears?
222 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

WHAT YOU CAN DO SOON

What you can d o soon is figure out how you can improve your personal in-
telligence network. Even if your personal network’s in good shape, still work on
it: The best time to build your intelligence network is before you really need it.
Times and situations change, and you never really know where the next piece of
critical information will come from. By continually augmenting your network,
you improve the odds that you’ll get the news you need when it breaks. If you
wait until you need information, it’s too late to develop your network.
The key to creating your personal network is knowing where to build con-
tacts, not just how. You must figure out what critical information you need and
where it’s produced before investing in the establishment of an information net-
work infrastructure. First, look inward and consider your internal information
needs:

0 What kinds of information d o you need? What information is critical to

your ability to d o your job? What’s critical to your group’s ability to d o
its job?
0 Where are you (and your group) in the flow of internal information?
W h o are your internal customers? Who are your internal suppliers?
0 Where are your key information uncertainties and threats?$*W h o makes
decisions that affect your fate or the fate of your group? Where is infor-
mation generated that you really need to know?

Now look outward and consider the types and sources of information you
need from the organization’s wider environment:

What kind of information do you need? What information is critical to

your success, your group’s, and your organization’s?
0 Where’s your organization located in the production-consumption chain?
How far are you from the ultimate consumer?
0 Where are your critical uncertainties and threats-customers, suppliers,
competitors, regulatory actions, emergent technologies, and so on?

You may find that you can’t pin down precisely all the types and sources of
information you need. That’s OK, because you never really know where the next
piece of critical information will come from. You may not even know that a bit of
information is critical until some time after you get it. People who network smart
report that chance encounters, free-ranging and seemingly aimless conversations,
tidbits dropped and overheard are frequently the sources of what, in retrospect,
was vital information. There’s a healthy element of chance in intelligence gather-
ing. That’s why a diverse set of information contacts is necessary. Your objective is
to be in the right place at the right time-wherever and whenever that is-and you
must cast a broad net to make sure you are.
Building Intelligence Networks 223

Once you have some idea of the critical types and sources of information,
your task is to build contacts. Remember the fifth networking principle: It’s a
small world. . . . It’s your ally in the network-building process. The small-world
phenomenon means you’re never that far from the information you need. Every
time you increase your network of direct contacts by a single person, you tap into
a vast network of indirect contacts. This principle may help you see massive re-
structurings as blessings in disguise. Why? Restructuring means much wider spans
of management, and wider spans augment your information network: You get
more direct information sources (your additional direct contacts) and many more
indirect sources (the personal networks your direct contacts bring with them).
Look back at your answers to the 20-question quiz. Is there a pattern to
your yes and no answers? If you tended to answer no to questions 2 through 12,
then your internal intelligence system is deficient and you should start there. Here
are three suggestions for getting started:

0 Start your own version of Donuts with Ditch. You don’t have to make a
big deal about it by making formal announcements. Just extend a casual
invitation to chat about “things.” Keep it open and free-wheeling. (Be
sure to bring your favorite nosh-food is always hard to resist!)
0 Reactivate one or two old contacts in groups or departments in which
you once worked. Pick up the phone and extend an invitation to coffee or
lunch. Or just stop by.
Share information with a subordinate or team member; let him or her
know something that you know. Remember that sharing information em-
powers people. A subordinate’s or team member’s power-the ability to
get the job done-depends directly on his or her access to i n f o r m a t i ~ n . ~ ~

If you tended to answer n o to questions 13 through 20, then your external

intelligence system is deficient and you should concentrate there. (If your no an-
swers appear in both sections, start building your internal network first and then
proceed to work on external networks.) Here are three suggestions that can help
you begin the process of building your external intelligence system:

Think of who inside the organization is a natural bridge to a part of the

outside world you want more information about. The bridger could be a
salesperson, scientist, district manager, computer person, secretary, and
so on. Invite that person to your next Donuts with Ditch session.
Identify an outside supplier who might be a good source of information.
It could be your company’s law firm, ad agency, banker, and so on. Call
them and invite them to lunch.
0 Take a trip to an outside conference or trade show. It could be directly re-
lated to your business but it doesn’t have to be. Attend the sessions, go to
cocktail parties, don’t eat alone if you can help it. Volunteer to d o some-
thing at the next meeting.
224 KNOWLEDGE MANAGEiMENT AND ORGANIZATIONAL DESIGN

WHAT YOU CAN DO IN THE LONG RUN

What you can d o in the long term is manage conditions: create the right
context that will help you and your organization develop and refine the overall in-
telligence system. Remember, the more you help others build their personal intel-
ligence networks, the more you multiply your own.
Develop a reward system that encourages network building. Establish a
travel budget that lets people attend trade shows, conferences, professional meet-
ings, learned societies, educational seminars, and so on. In fact, make it a formal
requirement that everyone goes to one such event at least every six months. If you
can’t afford to send everyone out of town, have them attend local events during
the day, evenings, or weekends. And give everyone this chapter to read before they
go*
Develop fast internal channels for processing information and getting it to
the right people. Internal channels at Tennant Company are a good example. This
leading manufacturer of floor maintenance equipment uses the sales force as
“quality eyes and ears” in the company’s total quality improvement program.
“Every time a machine is delivered,” say CEO Roger L. Hale and associates, “the
salesperson fills out an installation report. If any defects are present, each one is
the subject of a separate report. These reports are sent to the warranty and quality
departments at company headquarters.”60
When hiring, ascertain each candidate’s network assets (the contacts a can-
didate brings with him or her) and networking capabilities (a candidate’s motiva-
tion and ability to make new contacts). Everyone’s personal network is portable.
Every time you hire a well-connected person you annex a new network of direct
and indirect contacts, new sources of information. (This is one reason behind the
federal government’s so-called revolving door restriction: a one-year waiting pe-
riod after leaving a government post before an ex-federal employee can return and
lobby for private interests.) When firing, or thinking about it, be sure to consider
a person’s network. If you don’t, you might unwittingly sever important links to
the outside.
When accounts are up for review, don’t forget that your suppliers have net-
work assets and networking capabilities as well. Are you getting the information
you need from your lawyers, bankers, consultants, engineers, advertising agen-
cies, and other professional service suppliers? Because these are easier to change
than suppliers of goods. . . ,you may want to consider adding or switching sup-
pliers to boost your intelligence networks.
Establish a long-term intelligence trajectory-the directions into which you
want to expand the organization’s intelligence system. Pick a key uncertainty and
establish a plan to build networks in that direction. If you’re in an industry noted
for fast technological change, for example, work to build information networks of
scientists, engineers, university contacts, and so on. If you’re in a highly regulated
and politicized environment, think of adding political and regulatory contacts to
your network. If supply of critical raw materials is a recurrent problem, build net-
works that will help you monitor supplies, substitutes, and suppliers.
Building Intelligence Networks 225

Physical location can be used to create the right intelligence conditions.

Edison Electric Institute didn’t relocate from New York to Washington, D.C., be-
cause real estate was cheaper in the nation’s capital.6’ It relocated as a way to be-
come more central in political communication networks. As the major trade
association for the electrical utility industry, the Institute had to enlarge its infor-
mation-collecting and processing capabilities in response to the increasing politici-
zation of energy policy making. Companies locate in close proximity to facilitate
the face-to-face exchange of information too ambiguous or sensitive to transmit
via electronic media.62 General Motors’ long-time advertising agency, McCann
Erickson (now part of The Interpublic Group of Companies), has offices located
right in GM’s Detroit office building. Similarly, Capital Partners’ main architec-
tural firm has its headquarters in Capital Partners’ home office.
A few final words. You’ll never really be done building your personal intelli-
gence network. Times and situations change. Always look for ways to supplement
your intelligence contacts. From time to time, reread this chapter, reassess your
situation, and work to become the director of your own intelligence network.

REFERENCES
1. Quoted in Philip Kotler, Marketing Management, 6th ed. (Englewood Cliffs, N.J.:
Prentice-Hall, 1988), p. 101.
2. Story in Time, September 30, 1991, p. 19.
3. H. Edward Wrapp, “Good Managers Don’t Make Policy Decisions,” Haruard Busi-
ness Review, July-August 1984, pp. 4-11. Quotation from p. 5 .
4. From his speech at the Crowell Collier Institute of Continuing Education (New York);
reported in Business Week, February 18, 1967, p. 202.
5 . Richard Neustadt, Presidential Power: The Politics of Leadership (New York: John
Wiley, 1960).
6 . Alvin Toffler, Powersh;fr (New York: Bantam Books, 1990).
7 . Loy Singleton, Telecommunications in the Information Age (New York: Ballinger,
1984).
8. Richard Saul Wurman, Infomation Anxiety (New York: Doubleday, 1989).
9. See, for example, J. Edward Russo and Paul J. H. Schoemaker, Decision Traps (New
York: Doubleday, 1989).
10. Wrapp, “Good Managers,” p. 4.
11. Chapter 3 in Harry Levinson and Stuart Rosenthal, CEO: Corporate Leadership in
Action (New York: Basic Books, 1984). Quotation from p. 68.
12. Stephen W. Quickel, “Welch on Welch,” CEO of the Year, Financial World, April 3,
1990, pp. 62-67.
13. Thomas J. Peters and Robert H. Waterman, In Search of Excellence: Lessons from
America’s Best-Run Companies (New York: Harper & Row, 1982).
14. “The CEO Disease/Avoiding the Pitfalls of Power,” Australian Business, May 1, 1991,
pp. 36-41 (which is reprinted from Business Week, April 1, 1991).
226 KNOWLEDGE MANAGEMENT AND ORGANIZATTONAL DESIGN

15. Mitchell Locin, “Little Rock Hops for Local Hero,” Chicago Tribune, November 4,
1992, sec. 1, p. 5 .
16. Ronald S. Burt, Structural Holes (Cambridge, Mass.: Harvard University Press, 1992).
17. Henry Mintzberg, “The Manager’s Job: Folklore and Fact,” Harvard Business Review,
July-August 1975, pp. 49-61.
18. See, for example, Rosabeth Moss Kanter, “Power Failure in Management Circuits,”
Harvard Business Review, July-August 1979, pp. 65-75.
19. See an excellent discussion of reciprocity in Chapter 2 of Robert B. Cialdini, Influence:
Science and Practice (Glenview, Ill.: Scott, Foresman, 1985). Anthropologists and soci-
ologists have documented that reciprocity is a rule in all cultures and societies.
20. For insights on gossip and decision making, see James G. March and Guje Sevon,
“Gossip, Information and Decision-Making,’’ in Lee S. Sproull and J. Patrick Crecine
(eds.), Advances in Information Processing in Organizations, vol. 1 (Greenwich,
Conn.: JAI Press, 1984), pp. 95-107.
21. From interview with Allan Ditchfield, April 23, 1993. Used with permission.
22. “The CEO Disease,” p. 40.
23. Described in letter from Kathryn E. Milano, Federal Express, January 18, 1993.
24. Ram Charan, “How Networks Reshape Organizations-for Results,” Harvard Busi-
ness Review, September-October 1991, p. 110.
25. Described in Noel M. Tichy and Stratford Sherman, Control Your Own Destiny or
Someone Else Will (New York: Doubleday, 1993), pp. 156-166.
26. Jane Hannaway, Managers Managing (New York: Oxford University Press, 1989), pp.
104-105.
27. See, e.g., “Japan: All in the Family,” Newsweek, June 10, 1991, pp. 37-39.
28. Kathleen M. Eisenhardt, “Speed and Strategic Choice: How Managers Accelerate De-
cision Making,” California Management Review, vol. 32, 1990, pp. 1-16.
29. Eisenhardt, “Speed and Strategic Choice,” discusses in detail the concepts of real time
information networks and fast decision making.
30. See, for example, Charles Savage, Fifth Generation Mamgement: Integrating Enter-
prises through Human Networking (Digital Press, 1990); and Lee Sproull and Sara Ki-
esler, Connections: New Ways of Working in the Networked Organization
(Cambridge, Mass.: MIT Press, 1992).
31. In Brad Schepp, “The 10-Second Commute,” Home Office Computing, December
1991, pp. 45-48.
32. For example, see Richard L. Nolan, Alex J. Pollock, James P. Ware, “Toward the De-
sign of Network Organizations,” Stage by Stage, vol. 9, 1989, pp. 1-12, published by
Nolan, Norton, & Co.
33. Henry Mintzberg, “The Manager’s Job: Folklore and Fact,” Harvard Business Review,
July-August 1975, pp. 49-61.
34. Ibid., p. 167.
35. Ibid.
36. See, for example, Sproull and Kiesler, Connections, p. 15, Chapter 2.
37. Ibid., pp. 37, 39.
38. James, L. McKenney, Michael H. Zack, and Victor S. Doherty, “Complementary
Communication Media: A Comparison of Electronic Mail and Face-to-Face Comrnu-
nication in a Programming Team,” in Nitin Nohria and Robert G. Eccles (eds.), Net-
Building Intelligence Networks 227

works and Organizations: Structure, F o ~ m and , Action (Boston: Harvard Business

School Press, 1992), Chapter 10.
39. Quoted in Sproull and Kiesler, Connections, p. 40; originally cited in L. K. Trevino, R.
Lengel, and R. L. Daft, “Media Symbolism, Media Richness, and Media Choice in Or-
ganizations,” Communication Research, vol. 14, 1987, pp. 553-574.
40. O n the importance of symbols, ceremony, and management, see, for example, Harri-
son M. Trice, James Belasco, and Joseph A. Alutto, “The Role of Ceremonials in Or-
ganizational Behavior,” Industrial Management Review, vol. 23, 1969, pp. 40-51.
41. Jane Hannaway, Managers Managing (New York: Oxford University Press, 1989), p.
144.
42. Joshua Hammer, “The Fall of Frank Mancuso,” Newsweek, May 6,1991.
43. O n networks in publishing, see Walter W. Powell, Getting into Print (Chicago: Univer-
sity of Chicago Press, 1985) and Lewis Coser, Charles Kadushin, and Walter W. Pow-
ell, Books: The Culture and Commerce of Publishing (New York: Basic, 1982),
especially Chapter 3, “Networks, Connections, and Circles.”
44. Personal interview, Summer 1991.
45. Tom Peters, Thriving on Chaos (New York: Harper & Row, 1987),p. 283.
46. Ibid., p. 184.
47. Described, for example, in James Brian Quinn, Henry Mintzberg, and Robert M.
James, The Strategy Process (Englewood Cliffs, N.J.: Prentice Hall, 1988), pp. 734-
735.
48. Described in “To Our Share Owners,” 1992 Annual Report, General Electric Co.
49. Ibid, p. 4.
50. Ibid.
51. Personal interview, Summer 1991.
52. Described to me by one of my Executive MBA students.
53. Estimate of cost savings from Anne B. Fisher, “HOWto Cut Your Legal Costs,” For-
tune, April 23,1990, pp. 185-192. Based, in part, o n Wayne E. Baker and Robert R.
Faulkner, “Strategies for Managing Suppliers of Professional Services,” California
Management Review, vol. 33, 1991, pp. 33-45.
54. Wayne E. Baker, “Market Networks and Corporate Behavior,” American Journal o f
Socio1ogy, vol. 96, 1990, pp. 589-625.
55. Gerald E Davis, “Agents without Principles? The Spread of the Poison Pill through the
Intercorporate Network,” Administrative Science Quarterly, vol. 36, 1991, pp. 583-
613.
56. See, for example, Linda Brewster Stearns and Mark S. Mizruchi, “Broken-Tie Recon-
stitution and the Functions of Interorganizational Interlocks: A Re-Examination,” Ad-
ministrative Science Quarterly, vol. 31, 1986, pp. 522-588.
57. Edward 0 . Laumann and David Knoke, The Organizational State (Madison, Wis.:
University of Wisconsin Press, 1987).
58. Sociologist Arthur L. Stinchcombe argues in Information and Organizations
(Berkeley, Calif.: University of California Press, 1990) that the organization’s main in-
formation task is to manage such uncertainties. The organization must be where the
news breaks, whenever it breaks.
59. See, for example, Rosabeth Moss Kanter, “Power Failure in Management Circuits,”
Harvard Business Review, July-August 1979, pp. 65-75.
228 KNOWLEDGE MANAGEMENT AND ORGANIZATIONAL DESIGN

60. Roger L. Hale, Douglas R. Hoelscher, and Ronald E. Kowal, Quest for Quality, 2d ed.
(Minneapolis: Tennant Company, 1 9 8 9 ) , p. 56. Used with permission.
61. Based on Edward 0. Laumann and David Knoke, The Organizational State (Madi-
son, Wis.: The University of Wisconsin Press, 1987), p. 208.
62. I learned this from an excellent study conducted by one of my MBA students, Kenneth
A. Posner (MBA ‘91), ‘‘Using Hedonic Models to Examine the Value of Location”
(MBA Honors paper, Graduate School of Business, University of Chicago). In the pa-
per, he summarizes Edwin s. Mills’ argument in “Sources of Metropolitan Growth and
Development” (The Institute for Urban Economic Development, 1991) that firms
cluster to exchange ambiguous information via face-to-face interaction.
 Index

ABB, 59 Apple Computer, 63, 99-100, 1 1 4

“Abstraction of Industrial Work, The” alliances, 66
(Zuboff), 197-207 cultural pluralism in, 63-64
Action-centered skill, 199 inter-unit cooperation in, 60
Activation of innovation, Assembled knowledge, 21-22
100-1 06 AT&T, 3 9 4 0 , 8 0 , 1 1 6 , 2 1 3
Administrative innovation, 95 product development teams, 51
Alexandria Torpedo Factory and Art and Unix, 140
Center, 117-118 Automation
Alienability, 17. See also Decision rights and workers’ confidence, 199-206
CEOs and, 26-27 workers’ attitudes toward, 197-198
costs of suppressing, 26 Autonomy
functions of, 23-26 and innovation, 118-1 1 9
Alliances, 56 as motivator, 184
assessment of, 146-147
and competitive collaboration,
151-155 Badaracco, Joseph, 5
forms of, 152-153 “Knowledge Links,” 133-149
human resources considerations in, 66, Bahrami, Homa, 4
155-1 64 “The Emerging Flexible Organization:
imbalance in, 153-155 Perspectives from Silicon Valley,”
knowledge links as, 133-149 55-75
reasons for, 151-152 Bailyn, Lotte, 179, 180
staffing considerations in, 155-158, Baker, Wayne, 6
163 “Building Intelligence Networks,”
training/development issues in, 209-228
158-1 59,163-1 64 Barnevik, Percy, 59
Agency costs, 27-28 Beiber, Owen, 143
Ambassadors, 113. See also Boundary Bell, Graham, 80
management Bell System Practices, 40
American Airlines, 21, 2 14 Bi-modal organizations, 67-70
Antitrust laws, 125 Black, Joseph, 81

229
230 Index

Boards of directors, networking with, Clinton, Bill, 21 1

219-220 Coalition building, 106-1 12
Botkin, Jim, 2, 6ref members for, 109-110
Boulding, Kenneth, 209-21 0 Cohen, Susan, 6ref
Boundary management, 113-1 15 Cohn, S.F., 120
Boundary spanners, 98 Collaborative partnerships, 66. See also
Bridge structures, 121-123 Alliances
British Association, 82 CombsMoorhead Associates, 219
British Petroleum, 59 Communication. See also Networks
Budgets, 3 1. See also Resources across field boundaries, 98-99
for innovation, 94, 117-1 18 and decentralization, 68
in performance measurement, 33-34 density, 110
“Building Intelligence Networks” (Baker), face-to-face vs. mechanistic, 110,
209-228 215-216
Bureaucracy, 39-53, 191-194 of innovations, 115
chain of command in, 40-42 internal, 224
and changing nature of work, 46-53 integrationkoordination through, 51,
characteristics of, 40 100
coordination from above in, 45 open, 110
limits of, 46-53 and physical arrangements, 110
impersonality in, 45 of technological information, 81-86,
limits of, 46-53 87
professional career track in, 44-45 technology, impact of, 56
specialization in, 4 2 4 3 through the price system, 13
standardized procedures in, 44 with employees, 211,213-214
uniform rules/policies in, 43 Communism, 22,23
Burns, Tom, 4-5 Competencies, 153
“Organization of Innovation,” 77-92 Competition, and planning, 8
Burt, Ronald, 212 Complexity, and innovation, 100-101,
Bush, George, 209 105
Conner Peripherals, 69
Consultants, 69, 122
Caldwell, D., 113 Control, separation between ownership
Capitalism, 22-23, 35 and, 89
Cardwell, D.S.L., 86 Control systems, 24-25,32-35. See also
Career paths, 44-45, 182-183,212 Performance
Caring, 47-48 Controversy, and innovation, 94-95
Cash, James I., 2,6ref Cooperation
Chain of command, 4 0 4 2 , 1 9 2 interorganizational, 125
Champy, James, 2, 6ref inter-unit, 60
Change, 194 Coordination
belief in unimportance of, 10-1 1 among peers, 51-52
as cause of economic problems, 10 from above, 45
kaleidoscopic, 58 Cost centers, 32-33
managing, 55 Cost overruns, 117-1 18
and organizational obsolesence, 65 Craftsman entrepreneurs, 41-42
and risk, 78-79 Craftsman system
stability and, 68-69, 115-117 industrial technology and, 85
technology as cause/effect of, 77-79 limits of, 4 1-42
Citicorp, 211 Cross-fertilization of ideas, 98-100
Index 231

Cross-functional teams, 49, I 1 1 Dualistic systems, 60-62

collocation of knowledge and decision Durkheim, Emile, 77
rights in, 21
Cross-training, 50,65, 111
Crozier, Michel, 5-6 Eccles, Robert G., 6ref
“The Social System at the Shop Level,” Economic calculus, 1 2
191-1 95 Economy
Cultural pluralism, 63-64 change, and problems in, 10-1 1
and alliances, 158 global, ix
balancing with uniformity, 69-70 organization of, 8-15
and innovation, 95 Edinburgh, University of, 81
Culture, organizational, 104-105 Edison, Thomas, 80, 86, 91
Customers, 50-51,62 Edison Electric Institute, 225
as innovation activators, 97-98 Education
networking with, 218-219 in Britain, 85-86, 169, 170
for innovation and caring, 48
in Japan, 170
Daewoo, 136 and scientific knowledge, 85-86
Data General, 113, 116 in Singapore, 170-1 71
Davenport, Thomas H., 2, 6ref Eisenhardt, Kathleen, 214
Davis, Gerald, 220 Electrical industry, invention in, 79-81
Davis, Stan, 2, 6ref Electronic Data Services, 146
Decentralization Electronic mail, 215, 216
balancing with centralization, 67-68 Embedded knowledge, ix
determining optimal levels of, 27-28 compared with particular knowledge,
factors affecting, 29-30 5
of knowledge, 8,11, 18,22-26 and knowledge links, 134-135
Decision control rights, 29 Emery, Fred, 40, SO
Decision management rights, 29 Empowerment, 4 2 , 1 8 0 , 2 1 2
Decision rights Entrepreneurs, 95
alienability of, 17, 23-26 craftsman, 41-42
budgets and, 31 Environment, external, 4
collocating with knowledge, 18-22, Evolutionary innovations, 95, 120
24 Expertise, see Knowledge
control systems of, 22-23 Externalities, 24, 35
and costs, 26-30
decentralization and, 27-30
firms and, 25-32,35-36 Federal Express, 214
intra-firm transfers of, 24-25 Ferranti, Sebastian Ziani de, 84
the law and, 23 “Fertile fields,” 124
organizational systems for assigning, Flexibility
30-3 5 in bi-modal organizations, 67-70
and performance systems, 24-25, building blocks of, 58-67
32-35 definition of, 57-58
Delayering, 56 for innovation, 117-1 1 8
Ditchfield, Allan, 213, 216 Food Marketing Institute, 122-123
Diversity, 69-70, 100-101. See also Ford Motor Company, 45, 146
Cultural pluralism Franchise organizations, 26
Down-sizing, 56 Friedlander, E, 115
Drucker, Peter, 46, 96, 179 Front-line orientation, 62-63
232 Index

Galbraith, J., 2n, 6ref, 106, 112 Hertzberg, Frederick, 180

Geneen, Harold, 216 Hewlett, Bill, 6 9
General knowledge, 21 Hewlett Packard, 65, 101
definition of, 17 Hierarchical relationships, 109, 1 9 4
General Electric, 156, 214 High-tech firms
quick market intelligence technique, characteristics of, 58
2 17-2 18 organizational design in, 56-57, 59-70
General Motors organizational tensions in, 58-59
and Daewoo, 136 Hoechst Celanese, 49
and Fanuc, 137-1 38 Hollomon, J.H., 123
knowledge links of, 133-134, Honda Motor, 94
13S-136,137-138,142-147 Honeywell, 104,122
Manufacturing Automation Protocol Hughes Aircraft, 145-146
(MAP), 144 Human resources, 5,180. See also Labor
reorganization of, 145-146 and network building, 224
Saturn, 124, 142-145 and strategic alliances, 66, 155-164
and Teknowledge, 144-145
and Toyota, 136
and U.S. Defense Dept., 145 IBM, 4 0 , 6 1 , 6 5
German industry, 85-86 knowledge links of, 133-134,136,
Gerstein, Marc S., 6ref 139-1 42
Glasgow, University of, 81 and MCI Communications, 140
Gladstein, D., 113 and Mitsubishi, 140
Globalization and Motorola, 141
of culture, 63-64 and Nissan, 141
of the economy, ix and NIT, 140
GMFanuc Robotics Corporation (GMF), and Rolm, 141
137-1 38 and Sematech, 1 4 1
Government regulation and Siemens, 1 3 6 , 1 4 1
and decentralization, 29 Idea fairs, 1 2 2
and diffusion of innovation, 123-124 Idea generation, 96-100
Grove, Andrew, 64 Inaba, Sieuemon, 137
Group pressure, 194 Ince, Francis, 84
GTE Telemessenger, 117, 118 Individual knowledge, see Particular
Gumpert, D., 94, 119 knowledge
Industrial age, the
bureaucracy and, 40-45
Haas, Howard, 217,219 and organizational design, 4
Hale, Roger L., 224 Information. See also Knowledge;
Hammer, Michael, 2, 6ref Networks
Handy, Charles, 5 as power, 108
“The Numbers, ” 16 7-1 7 8 reliance on others for, 209-210
Hannaway, Jane, 216 sharing, 212-213
Harper, Marion, 209 Innovation, 4 7
Hayami, Y., 124 accountability/autonomy in, 118-1 1 9
Hayek, Frederick A., 4 , 1 8 , 2 0 balancing with controVstability, 58-62
“The Use of Knowledge in Society,” boundary management in, 113-115
7-1 5 by invasion, 99-100
Hertz, Heinrich, 80 characteristics of, 93-95
Index 233

characteristics of ideas likely to Kaleidoscopic change, 58

succeed, 107-108 Kaleidoscopic thinking, 98-1 00
and coalition building, 106-112 Kanter, Rosabeth Moss, I n , 5, 180
commitment to, 116-1 17 “When a Thousand Flowers Bloom,”
conditions supporting, 95 93-131
education for, 48 Karel, 138
and flexibility, 117-1 1 8 Kielser, Sara, 215
idea generationlactivation in, 96-106 Knoke, David, 220
isolationlintegration and, 105-106, Knowledge
112-113 costs of transferring, 19-22
job definition and, 101-103 current interest in, ix-x
and organizational context, 5 and decision authority, 18, 22-26
opposition to, 114-1 15 distribution of, 11, 1 8
organizational expectations for, embedded, ix, 5,134-135
103-1 05 human limits on, 18-1 9
personnel continuity in, 115-117 individual requirements for, 11-12
physical separation for, 112-113 kinds of, 8-10
realizatiodproduction of, 112-119 obstacles to organizational learning,
socialization of, 83 154-155,156
tasks in, 96 organizational, definition of, 2
transfer/diffusion of, 119-125 particular, 9-10
types of, 95 representation of, 2
Intel Corporation, 64, 69 scientific, 9, 21-22, 81-86, 8 7
Intellectual capital, see Knowledge society’s use of, 7-15
Interactive learning, 115 unquantifiability of, 11, 2 0
Interdependence, 109 Knowledge-based industries,
Interface structures, 121-123 55-70
Internal common law, 30-31 Knowledge links, 133-149
Intrapreneurs, 47-48, 96 assessment of, 146-147
Invention changing capabilities through,
communication in, 81-86, 88 138-145
genius and, 80-81 characteristics of, 134-136
research and development and, 88-91 extending capabilities through,
social context of, 79-88 136-138
as social phenomenon, 79 G M and, 133,142-145
“Invisible hand,” 22-23 IBM and, 133,139-142
Isolation, and innovation, 112-1 1 3 and product links, 134-135
Knowledge management, research review
of, 2-6
Jensen, Michael C., 4 Knowledge work, 46-47
“Specific and General Knowledge, and Kobayashi, Kojii, 139
Organizational Structure,” 17-38 Kuehler, Jack, 133
Job descriptions
broadly defined, and innovation,
101-103 Labor
and decision rights, 30-31 changing job structures and, 167-168,
Jobs, Steve, 100 173-1 74
Joy, Bill, 55 changing skill requirements for,
Just-in-time (JIT) manufacturing, 144 168-169,174
234 Index

Labor (Cont.): Mechanistic organizations, 4

division of Meckling, William H., 4
in the price system, 14-15 “Specific and General Knowledge, and
through specialization, 42-43 Organizational Structure,” 17-38
implicit lifetime contract with, 56-57 Mintzberg, Henry, 212, 215
and motivation, 179-189 Mitsubishi, 140
multiskilled, 50, 64-65, 111, 135 Mohrman, Allan, 2n
organization of, and innovation Mohrman, Susan A., 2n, 3, 6ref
adoption, 123 Money, as motivator, 184
participatory relations with, 143-144 Monopolies, 191-192
staff/line distinctions in, 62-63 Morita, Akio, 217
temporary, 69 “Motivating Knowledge Workers”
turnover of, 45, 115-116 (Tampoe), 179-189
work force, aging of, 171-173, Motivation, 5, 179-1 89
174-1 75 and career stages, 182-183
work force size, 169-171 career track as, 44-45
Lancourt, Joan, 2,6ref and management styles, 1 8 7
Lange, Oscar, 15 model of, 181
Lateral relationships, 109 and motivating factors, 184-185
Laumann, Edward, 220 organizational culture and, 104-105
Lawler, Edward E., 111, 181 and perception of environment, 1 8 7
Lean manufacturing, 4 9 and performance, 2
Lechmere, Inc., 50 working environment for, 187-1 88
Lerner, Abba P., 15 Motorola, 141
Levinson, Harry, 21 1 Mrs. Fields Cookies, 29
Lewis, Jordan, 6ref Multiskilled workers, 50, 64-65, 111,
Lockheed, 112 135
Lunar Society, 82 Myers, Paul S., 1-6

Mach, Michael, 21 8 Nadler, David A., 2n, 6ref

Machiavelli, Niccolo, 139 Nepotism, 44, 4 5
McGill, Michael E., 2, 6ref Networks, 209-228
McKenney, James, 215-216 building external, 216-219
McLean, Malcolm, 98 building internal, 210-213
Maintenance workers, and production coalition building as, 106-1 12
workers, 192-1 94 computer, ix, 214-216
Management information systems (MIS), determining information needs and,
215 2 2 2-2 23
Management of Innovation, The (Burns, eval ua tinglimproving, 22 1-225
Stalker), 4-5 in high-tech companies, 64-65
“Managing by wandering around,” 21 1, of interdependent members, 109
217 personal contacts, 98-99, 210,
Marconi, Guglielmo, 80 216-21 7
Marx, Karl, 7 7 with employees, 213-214
Matrix organization structure, 100-101, with superiors, 219-220
111-1 12 Nevis, Edwin C., 6ref
Matsushita, 146 Nissan, 141
Mayo, Elton, 83 Nohria, Nitin, 6ref
MCI Communications, 140 Nolan, Richard L., 6ref
Index 235

Nonaka, Ikujiro, 2 , 4 , 6ref Pay-for-performance systems, 34-35

“Not invented here” syndrome, 120 Pay-for-skill systems, 102
NTT, 140 People Express Airlines, 97
NUMMI, 136,158 Performance
and alienability of decision rights,
24-25,32-35
Oblique relationships, 109 budgets and, 33-34
Occupational psychosis, 99 and business strategy, 2
Ocean Spray, 99 control systems, 24-25, 32-35
Orchestrators, 106. See also Coalition cost centers/profit centers in, 32-33
building and group cohesion, 105
Organic organizations, 4 individual vs. group, 34-35
Organizational design Personal growth, as motivator, 184
agency costs vs. information costs in, Peters, Tom, 211,217
26-30 Pinchot, Elizabeth, 4
alliances/partnerships and, 66-67 “Rise and Fall of Bureaucracy, The,”
assigning decision rights in, 30-32 39-5 3
bi-modal, 67-70 Pinchot, Gifford, 4
complexity and innovation and, “Rise and Fall of Bureaucracy, The,”
100-101,105 39-53
control systems in, 32-36 Planning
cultural pluralism and, 63-64 centralized vs. decentralized, 8, 11
decision rights in, 30-35 human resource, in alliances, 155-1 58
determining optimal decentralization reliance on statistics in, 11
in, 27-30,6748 Polanyi, Michael, 134-135
drawbacks of traditional, 59-60 Policieslrules, in bureaucracy, 4 3 4 4
dualistic, 60-62 Porter, Lyman W., 181
high-tech firms experiments in, Power, 22. See also Decision rights
56-57 in bureaucracy, 4 3 , 4 5
for innovation, 94-131 separation of ownership from, 89
matrix, 100-101,111-112 tools of, 108, 111
multi-polar, 59-60 Price system, 12-15, 22. See also Decision
multi-skilled employees in, 64-65 rights, alienability of
and power relationships, 191-194 Process innovations, 95
“rules of the game,” 30, 36 Process model of innovation, 96
stafflline distinction and, 62-63 Product centers, 33
and structural constraints to Product innovations, 95
innovation transfer, 120-121 Production workers, and maintenance
Organizational learning, see Knowledge workers, 192-194
Outsiders, and innovation, 99-100 Professional career track, 44-45
Outsourcing, 25 Profit centers, 32-33
Profit-sharing plans, 34,35
Property right, 17. See also Alienability,
Pacific Telesis, 124 of decision rights
Packard, David, 69 Prusak, Laurence, ix-x
Parallel organizations, 122 Publicity value, 107
Paramount Pictures, 216 Pucik, Vladimir, 5
Particular knowledge, 5, 9-10 “Strategic Alliances, Organizational
costs of transferring, 20-22 Learning, and Competitive
disdain for, 9-1 1 Advantage,” 151-1 65
236 Index

Quick market intelligence technique, Sasaki, Toru, 179

2 17-21 8 Saturn (car), 124, 142-145
Quinn, J.B., 97, 118 Schon, D., 99
Scientific and Industrial Research, Dept.
of (Britain), 87-88
Radio, invention of, 80 Scientific journalism, 82-83
Rangan, U. Srinivasa, 6ref Scientific knowledge, 9, 21-22
Rates of equivalence, 12 communication of, 81-86, 87
Raytheon, New Products Center, 97, Scientific management, 44
106 Scott, B., 115
Redundancy, SO Scouts, 113. See also Boundary
Reeingineering, 2 , 4 9 management
Relationships, work, 5 1 , 5 3 Scully, John, 6 3
employer-employee to peer-to-peer, Sea-Land, 98
57,60 Sematech, 1 4 1
interdependent, 109-1 10 Sentries, 113. See also Boundary
power, 191-194 management
Report o n Scientific and Engineering Shackle, G.L.S., 58
Manpower in Great Britain, 87 Shaw, Robert B., 6ref
Research and development, 88-91. See Shoemaker, F.F., 1 2 2
also Innovation Siemens, 136, 141
Reservations, 112 Skunkworks, 112
Resilience, 57. See also Flexibility Slocum Jr., John W., 2, 6ref
Resources Smiles, S., 82
and innovation, 95, 103-104, Smith, Adam, 15
117-118,121 Smith, Roger, 142-143
multi-stage commitment of, 119 Socialism, 22, 23, 24-25
as power, 108 Social science research, 3
price system’s utilization of, 1 4 “Social System at the Shop Level, The”
and staffing in alliances, 157-158 (Crozier), 191-195
staff time as, 104 Society
Restructuring, 223 economic problem of, 7-8
Revolutionary innovations, 95, and innovation adoption, 123-125
120 as technological process, 78
Rewards, 1 81, 184-1 85. See also technology and changes in, 77-79
Motivation Sony, 9 7 , 2 1 7
“Rise and Fall of Bureaucracy, The” Specialist clash, 179
(Pinchot, Pinchot), 39-53 Specialization, 42-43
Robustness, 57. See also Flexibility of scientific knowledge, 84-86
Rogers, E.M., 122 and teams, 49-50,49
ROLM Corporation, 61, 6 9 “Specific and General Knowledge, and
and IBM, 1 4 1 Organizational Structure” (Jensen,
System Development Group, 6 2 Meckling), 17-38
Rosenthal, Stuart, 211 Specific knowledge
Royal Bank of Canada, 214 definition of, 17
Royal Dutch Shell, 6 5 distribution of, 28
Royal Society of Edinburgh, 82 Sponsor role, 106. See also Coalition
“Rules of the game,” 30, 36 building
implementing, 30-35 Sproull, Lee, 215
Ruttan, V.W., 1 2 4 Stability, 60-62, 115-117
Index 237

Staff time, 104 Trained incapacity, 99

Stalker, G.M., 4-5 Trist, E., 124
Starbuck, W.H., 103 Trotsky, Leon, 1 5
Stevenson, H., 94, 119 Turnover, employee, 45, 115-1 1 6
Strategic alignment, 119-121 Turyn, R.M., 120
Strategic alliances, see Alliances
“Strategic Alliances, Organizational
Learning, and Competitive Unions, labor, 50, 124, 135
Advantage” (Pucik), 151-1 65 United Airlines, 21
Strategic business planning, 34 United Auto Workers, 124, 143, 146
Sub-contracting, 56. See also Alliances United States, State Department, 214
Sun Microsystems, 55 Universal price codes, 123
Sununu, John, 209 Unix, 140
Support, as power, 108. See also User groups, 97-98
Coalition building “Use of Knowledge in Society, The”
Suzuki, G M alliance with, 135-136 (Hayek), 7-15

Takeuchi, Hirotaka, 2, 4, 6ref Vagelos, Roy, 21 1

Tampoe, Mahen, 5 Van de Ven, A.H., 97, 107, 118
“Motivating Knowledge Workers,” von Mises, Ludwig, 15
179-1 89
Tanen, Ned, 216
Task achievement, as motivator, 184 Walton, R., 122
Taylor, Frederick, 44 Waterman, Robert, 211, 217
Teams, 48-49 Watt, James, 81
accountability and autonomy of, Weber, Max, 39, 40
118-119 Weitzman, Martin, 143
and boundary management, Welch, Jack, 21 1
113-115 “When a Thousand Flowers Bloom”
cross-disciplinary, 49 (Kanter), 93-131
flexibility of, 56 Whitehead, Alfred, 14, 79
ideal longevity of, 117 Wilmot, Robb W., 179-1 80
integrative mechanisms for, 111-112 Work, changing nature of, 46-53,
personnel continuity in, 115-1 17 167-170,173-174
project, 49-50 Workforce diversity, 64. See also Cultural
temporary, 6 1 pluralism
Technological feasibility, 1 9 Wozniak, Steve, 100
Technology, social change and, 77-79 Wrapp, Ed, 209
Teknowledge, 144-145 Wriston, Walter, 21 1
Tennant Company, 224 Wurman, Richard Saul, 210
Third Age, The, 172-1 73
Thompson, Alfred, 84
Third Wave, The (Toffler), 48 Yoshino, Michael Y., 6ref
3M, 104, 122
Toffler, Alvin, 48, 210
Tonnies, E, 78 Zap Computers, 214
Total Quality programs, 4 8 4 9 Zuboff, Shoshana, 6
Toyota, 136, 143, 146 “The Abstraction of Industrial Work,”
Trade associations, 122-123 197-207