'Sv

MANAGEMENT
IN SMALL DOSES
MANAGEMENT
IN SMALL DOSES
RUSSELL L. ACKOFF
The Wharton School
University of Pennsylvania
Philadelphia, Pennsylvania

JOHN WILEY & SONS

New York / Chichester / Brisbane / Toronto / Singapore
Copyright © 1986 by Russell L. Ackoff

Published by John Wiley & Sons, Inc.

All rights reserved. Published simultaneously in Canada.

Reproduction or translation of any part of this work

beyond that permitted by Section 107 or 108 of the
1976 United States Copyright Act without the permission
of the copyright owner is unlawful. Requests for
permission or further information should be addressed to
the Permissions Department, John Wiley & Sons, Inc.

This publication is designed to provide accurate and

authoritative information in regard to the subject
matter covered. It is sold with the understanding that
the publisher is not engaged in rendering legal, accounting,
or other professional service. If legal advice or other
expert assistance is required, the services of a competent
professional person should be sought. From a Declaration
of Principles jointly adopted by a Committee of the
American Bar Association and a Committee of Publishers.

Library of Congress Cataloging in Publication Data:

Ackoff, Russell Lincoln, 1919-
Management in small doses.

1. Management. 3 I. Title.
HD31.A283 1986 658 86-7835
ISBN 0-471-84822-0

Printed in the United States of America

10 987654321
To

Jim Rinehart

who manages
to educate
as well as
manage
 Foreword
"Management in Small Doses" by Russ Ackoff is an abso¬
lute delight. Here are 52 sparkling and provocative essays
on management topics, each one easily read in three or
four minutes. The author recommends one per week, so
he gives us a full year's supply. Few will follow his pre¬
scription. I found myself gobbling them down in much
more substantial dosages.

Ackoff writes with a marvelous wit and practical common

sense that will appeal to the business reader. He is highly
irreverent, a smasher of so-called "accepted wisdom," yet
never irrelevant nor given to rhetorical hyperbole. Just
read his essay on business jargon to understand this
point. This is no huckster, but rather a respected academic
who has honed his skills in the real world of business.
How else could he write an essay on "Mess Management"
in which he says, "We do not experience individual prob¬
lems, but complex systems of strongly interacting prob¬
lems. I like to call them messes." You and I know these are
the situations we face in our businesses.

Don't miss his essay on "Advertising: Wonder or Waste."

Here are insights and examples that are worth the price of
the book, besides providing some real chuckles.

vii
via FOREWORD

After delving into these essays you may agree with the au¬
thor's statement, "business schools do a much better job
of educating their faculty than their students." At least
here's one very well educated faculty member.

Reginald H. Jones

Retired Chairman and CEO,

General Electric Company
 Preface
In the autumn of 1984 Dean Russell Palmer of The
Wharton School asked me if I would do a commentary on
an about-to-be-iaunched weekly cable television program.
Management Report. I agreed, with reservations.

First, my usual presentations consist of one- to three-hour

lectures. To say something worthwhile in no more than
the four minutes allotted to me seemed virtually impossi¬
ble.

Second, television's primary function is to entertain, but I

am an educator, not an entertainer. Moreover, I am an ed¬
ucator who depends critically on feedback from a live au¬
dience. I knew I could not entertain, and I doubted that I
could even educate, when addressing a camera and its op¬
erator, who is concealed behind the camera, in an other¬
wise empty studio.

Despite my doubts and deficiencies, Chris Graves, the

program's producer, and Alice Priest, its editor, worked
long and hard on and with me. They even tried to act as an
audience. I survived about 30 telecasts, after which the
program was discontinued for lack of financial support.

IX
X PREFACE

Although most of my commentaries were delivered with¬

out a script, I worked out my thoughts on paper before¬
hand. I showed a sample of them to my old friend and col¬
laborator C. West Churchman. He told me that they made
better reading than listening and suggested their publica¬
tion. Therefore, he is to blame for this.

After Management Report was taken off the air I continued

to prepare scripts because I had grown fond of trying to
express an idea succinctly. About 20 of these "posthu¬
mous" commentaries are included in this work. Although
each script was prepared as an independent entity, I have
tried to order them to give some semblance of continuity.

Because it is my hope that these commentaries will be dis¬

cussed and played with intellectually, the recommended
dosage is one per week. If my recommendation is fol¬
lowed, this book will provide a year's supply.

The commentaries are intended to provoke thought and

discussion, not necessarily to evoke agreement. I have no
desire to think for managers, but I do enjoy thinking with
and about them, particularly with and about those who
think for themselves.

I have received a great deal of constructive criticism and

encouragement from Tom Cowan, who has been doing
this to me for about 45 years. He has taught me a lot more
than I know. I am also grateful to Professor Harvey
Hornstein for his helpful suggestions.

Russell L. Ackoff

Philadelphia, Pennsylvania
June 1986
 Contents
1. Profit 1

2. Consumer Design 4

3. Appropriate Technology 8

4. Acquisitions 12

5. The Corporation Reconceived 16

6. Mess Management 20

7. Growth Versus Development 24

8. The Value of a Corporation 27

9. The Economy of the Firm 31

10. Performance Measurement 35

11. Mission Statements 38

12. Human Effectiveness 43

13. Quality of Work Life 46

14. Incentives 49

15. The Resurrection of A&P 53

16. Paternalism 57

XI
xii CONTENTS

17. Corruption 61

18. Alcoholism and Stress 65

19. Too Much Communication? 69

20. Communicating Up 75

21. Jargon 78

22. Computer-Controlled Managers 81

23. Management Misinformation Systems 84

24. Mathemanagement 88

25. Management Consultants 92

26 Types of Problems 97

27. Problem Treatments 101

28. Creativity 105

29. Beating the System 110

30. Defending Against New Ideas 114

31. The Obvious 119

32. Objectivity 123
33. Infallibility 127
34. Who's Irrational? 132
35. Advertising: A Wonder or a Waste? 136
36. Managing Interactions 141
37. The Aesthetics of Work 146
CONTENTS xiii

38. Friendship 150

39. Respect 154

40. Consensus 158

41. Thinking, Reading, and Talking 162

42. Learning 166

43. Understanding 169

44. Management Education 173

45. Town and Gown 177

46. The Fallacy of Forecasting 181

47. Obstructions to Progress 184

48. Planning Backward 188

49. The Clark-Volvo Joint Venture 192

50. Responsiveness 196

51. Comprehensive, Coordinated, and

Participative Planning 200

52. Continuous Planning 204

 . f'., •vi-
'''■.
-■
' ■ "'-».v ''t$
i ■; '•■
■^A
Al
'AH
At.-. ,'»>.
S?'-
■ %.'.’■ \ t qa '•-...

■*n^ ,,r|- y,j

777 '"i: vj’;-^,;v:^u’>?:^''- ;'c:i*^'ir>^'> ^J’'*'.i‘ '/svOT !>

1. r’.
'K
')! i'*Ai!j.'Lii
^A.liA'1 iffit : J

V- ■ : ■ , - . ' V ' . S
Syijyt-:; :.v:oV;7^ -j

.,♦ -t.l -I--/..-

\iii •>:

'* ?i

■'-i ■u'^ .*.

t.y 1-. k •.
MANAGEMENT
IN SMALL DOSES
 7

Profit is necessary for the survival of a business enterprise

but not the reason for it. Profit is a requirement, not an
objective; a means, not an end. The American humorist
Ambrose Bierce put this clearly when he defined money as
"a blessing that is of no advantage to us excepting when
we part with it."

Efforts to maximize profit are efforts to obtain money to

use for something else. What should this something else
be? Growth? No. Growth is also a means, not an end. This
is apparent when we consider how undesirable growth is
when it is accompanied by reduced profit. Growth may, of
course, increase profit. But what should profit be used for
and why growth?

1
2 PROFIT

The objectives pursued by organizations frequently differ

from those proclaimed. I learned this from spending 45
years in universities. Like many, when I started I assumed
that the principal objective of universities was the educa¬
tion of students. Armed with this assumption, I could
make no sense of their behavior. I learned that education
of students, like profit, is at most a requirement of univer¬
sities, not an objective. I also learned that their principal ob¬
jective is to provide their faculties with the quality of work life
and standard of living they desire. This enabled me to under¬
stand why professors do so little teaching, give the same
courses over and over again, arrange classes at their own
convenience, not that of their students, teach subjects
they want to teach rather than those the students want to
learn, give examinations that are easy to grade rather than
facilitate learning, and skip classes to give lectures for a
fee.

Universities are not unique in this regard. Corporations

are much the same. Having worked in several hundred of
them, I am convinced that those who manage them do so
primarily to provide themselves with the quality of work
life and standard of living they desire. I believe their be¬
havior can be better understood by assuming this than by
assuming that their objective is to maximize profit or
growth.

I see no reason to apologize for this objective. The only

thing wrong with it is its coverage. It should be extended
to cover the quality of work life and standard of living of
all employees. This quality and standard cannot be im¬
proved without making a profit, without paying divi¬
dends large enough to attract and retain investors, with¬
out increasing productivity so as to be competitive, and
PROFIT 3

without growth to absorb those employees displaced by

increased productivity.

The termination of an enterprise has a greater effect on its

employees than on any of its other stakeholders—the
lower their level, the greater their stake. Employees have
the greatest stake and make the largest investment in their
employing organization and incur the greatest losses
when it goes out of business.

To provide employees at every level with work that is in¬

teresting and financially rewarding and that provides op¬
portunities for personal development is something to be
proud of, not to apologize for or conceal. It can also give a
company's shareholders and the community of which it is
a part something to be proud of. It can even help sales.
Volvo, for one, has found this to be the case.

It is time for corporate executives to stop pretending they are not

caring people and to start enlarging the number of people for
whom they care.

REFLECTIONS*

*My hope is that each piece in this book provokes thought about the way things
are done in your organization. So, at the end of each piece you'll find a brief
space to record your reflections. You may find it a handy place to jot down your
ideas.
 2
Consumer Design

Producers often try to find out what consumers want by

asking them. This seldom yields useful information be¬
cause consumers either don't know what they want or
they try to provide (or avoid) answers they think are ex¬
pected of them. In many cases a better way consists of
using the consumer to design products or services; for ex¬
ample, a chain of men's stores, although successful, failed
to attract the type of customer its owner wanted. He
wanted to reach upwardly mobile professionals and busi¬
nessmen by offering high-quality designer clothing at dis¬
count prices. But this method failed; rather it drew bargain
hunters from lower income segments of the population.
Repeated questionnaires addressed to potential buyers the

4
CONSUMER DESIGN 5

firm wanted to attract yielded results that, when applied,

failed to bring them into the stores.

The owner and his executives sought help from a research

group with a reputation for unconventional approaches to
marketing problems. This group selected 15 representa¬
tives of the targeted customer population and invited
them to spend a Saturday designing their ideal men's
store. The identity of the sponsoring firm was not revealed
but several of its executives took part incognito.

The representatives of the targeted population produced a

very creative design of a men's store. Once done, the iden¬
tity of the sponsor was revealed and a comparison was
made of the sponsor's stores with the one newly de¬
signed. The principal differences were wide and had not
been revealed by any of the earlier research.

Here are a few of the differences. First, the designers

made it known that they always decided how much to
spend for articles of clothing before shopping. What they
were looking for was a store that offered the highest qual¬
ity at their predetermined price. Put another way; they did
not want to minimize price for a predetermined quality;
they wanted to maximize quality for a predetermined price.
Discount prices, the chain's advertising message, turned
them off.

Second, they wanted different articles of clothing of the

same size to be segregated. They did not want grouping
by type, which requires hunting all around the store.
6 CONSUMER DESIGN

Third, they wanted to examine clothing without a sales¬

man hovering over them. They wanted call buttons in¬
stalled in each size-organized area for summoning a sales¬
man when they wanted one.

Fourth, they did not want items for women in the store—
not even gifts. They preferred to shop for women in stores
that specialize in products suitable to them, just as they
preferred to buy their own clothing in stores that special¬
ize in supplying men. But they did want a lounge in which
women could pass their time pleasantly and comfortably
and be available for consultation. They thought a mixture
of male and female salespersons would be desirable.

They wanted alteration costs included in the price of the

clothing and delivery of all altered clothing. They also
wanted more informative labels with complete informa¬
tion on the material used, where the clothing was made,
and by whom. They said they were uninterested in the de¬
signer's name; they didn't believe that the designer
identified on men's clothing had designed it anyhow.

As regular customers of a store they wanted access to its

sales before the general public. They thought that there
should be only two sales a year, presummer and
prewinter, to occur at the same time each year. They also
wanted the store to maintain and refer to records of their
sizes and style preferences.

Similar consumer design groups have been used for other

types of product and service, and even to write advertis¬
ing copy. They have always been creative and informa¬
tive.
CONSUMER DESIGN 7

It is harder for a market researcher to get inside a consumer's

mind than it is for a consumer to turn his mind inside out.

REFLECTIONS
 3
Appropriate
Technology

As labor-intensive industries migrate from more to less de¬

veloped countries (from MDCs to LDCs) the LDCs need
easy-to-maintain production equipment that is appropri¬
ate to labor-intensive processes. Most of the production
equipment currently available to LDCs is inappropriate
because it was designed for use in MDCs. Equipment used
in LDCs should be designed with their costs of labor, en¬
ergy, and capital in mind, not those of MDCs; for exam¬
ple, old and new trucks designed in MDCs incorporate an
assumed economic value of the drivers' lives. This as¬
sumption determines how much safety is designed into
trucks. The life of a driver cannot be assumed to be
infinitely valuable because an infinite amount of money
cannot be spent to protect it. The maximum average

8
APPROPRIATE TECHNOLOGY 9

amount that a society can spend to protect the lives of its

members cannot exceed the present value of their ex¬
pected net contribution to their society's wealth over their
lives. This amount obviously varies from society to soci¬
ety; it is proportional to the wealth of countries. Therefore,
a truck designed with safety features appropriate to an
MDC is considerably overdesigned for use in an LDC.
This is equally true for most manufacturing equipment.

Furthermore, the appropriateness of equipment depends

on climatic, regulatory, and cultural, as well as economic,
conditions.

Markets in LDCs may not be large enough now to justify

the design and production of some types of equipment ap¬
propriate to them. In many such cases compromises are
possible; for example, trucks and production equipment
currently produced in MDCs often can be modified and
stripped down to make them more appropriate to LDCs.

New factories are being built in rural areas of LDCs to alle¬

viate overcrowded conditions in their major cities and to
promote rural development. Factories that are appropriate
in cities in MDCs are not appropriate in rural areas of
LDCs. Rural factories could be designed to serve as water¬
pumping and electricity-generating stations even if they
performed these functions only during off-hours.
Material-handling equipment could be designed to be use¬
ful on farms when not in use in factories.

Mobile facilities are particularly appropriate in many

LDCs. In Mexico, for example, CONASUPO uses mobile
10 APPROPRIATE TECHNOLOGY

markets to provide food for small, remote rural

communities. There is no reason why factories cannot also
be made mobile. Mobile food-processing plants, that
move from place to place as sequentially planned crops be¬
come ready for processing, might be useful in some LDCs.
One chemical company in an MDC has built a factory on a
barge which they tow to different ports as needed and a
factory that manufactures houses has been built into a set
of truck trailers for transportation from one construction
site to another.

Many consumer products supplied to LDCs by MDCs are

also inappropriate. Even when these products are manu¬
factured in LDCs they usually retain designs that are not
totally suitable. Western clothing is an example. This
clothing, which is often not functional even in MDCs, is
seldom as functional in LDCs as native dress. Yet as many
LDCs develop they adopt Western dress because it is
taken as a symbol of development.

The same is true of automobiles. Most cars designed in

MDCs are inappropriate in LDCs. Their cost is excessive
because they have properties of little or no value to LDCs;
for example, studies done in Mexico City show that the av¬
erage number of occupants in automobiles, including
taxis, is 1.2. Less than 15 percent of the cars carry more
than two people, which suggests that a two-passenger ve¬
hicle would serve LDCs better even than the "small" four-
passenger vehicles produced in MDCs. Two-passenger
automobiles sensibly designed for urban use in LDCs
would cost no more than half as much as the least expen¬
sive automobile currently available in MDCs. This cost
could be reduced further by designing automobiles that
could be assembled by purchasers.
APPROPRIATE TECHNOLOGY 11

Use of small urban automobiles would considerably re¬

duce traffic congestion, air pollution, and energy con¬
sumption, which are critical problems in many LDCs. Re¬
search has shown that if these vehicles were designed to
seat the passenger behind (not beside) the driver, and
they were used exclusively, 2.2 times as many people
could be carried on expressways and 2.7 to 5.4 times as
many on city streets, depending on parking arrange¬
ments.

By developing equipment, facilities, and products appro¬

priate to LDCs, companies in MDCs could not only create
new business opportunities but could contribute
significantly to the development of the less developed
countries of the world.

REFLECTIONS
 4
Acquisitions

A great deal has been written about how to make a suc¬

cessful acquisition and much of it is useful. Despite this,
however, some aspects of making an acquisition appear to
me to receive too little attention in practice.

More attention is usually given to what the acquired com¬

pany can contribute to the acquiring company than the
other way around. The premiums required to make most
acquisitions should be justified by potential additions to
the value of the company acquired. Put another way: the
principal question should not be “What can the acquisi¬
tion candidate do for us?" but "What can we do for it?" A
corporation that consists of more than one business can¬
not be justified unless the whole is worth more than the

12
ACQUISITIONS 13

sum of its parts. Therefore, unless the acquiring company

can add value to the company to be acquired, the acquisi¬
tion should not take place.

More than the usual amount of attention should be given

to the quality of the candidate's facilities and equipment.
These assets should be thoroughly examined by compe¬
tent members of the acquiring firm. If they are below that
firm's standards, estimates should be made of the cost of
"elevating" them. This cost is frequently overlooked or
underestimated.

However attractive an acquisition candidate may appear,

acquiring it is likely to succeed only if the managers of the
two companies like, respect, and trust one another. To
find out whether they do, they should interact in and out
of the business environment as much as possible before
the acquisition is made.

Despite all precautions, once an acquisition is made the

acquiring company may find it necessary to manage and
operate the acquired company. Are the technologies,
skills, and knowledge required to do so available in the
acquiring company? If not, the candidate should not be
considered.

More attention is usually given to an acquisition candi¬

date's past performance than to its potential. The value of
an acquired company lies in its future, not its past. There¬
fore, a company contemplating an acquisition should
make itself thoroughly familiar with, and carefully evalu¬
ate, the candidate's strategy for coping with its competi-
14 ACQUISITIONS

tive environment and the dynamics of this environment. If

management of the acquiring company is not capable of
evaluating that strategy, or fixing it if necessary, the acqui¬
sition should not be made.

Does the possibility of entering the acquisition candidate's

business excite and present a challenge to the manage¬
ment of the acquiring company? Will involvement in its
business be a source of pride and fun? Has it conducted its
business ethically and been socially responsible? Are its
employment practices compatible with those of the
acquiring firm? How do the candidate's employees at all
levels feel about their employer and the acquiring firm?

I have found that most acquiring companies do not know

enough about acquisition candidates to answer these
questions. They seldom identify or speak to former em¬
ployees of the candidate, nor do they consult its suppliers,
customers, security analysts, and bankers. Too low a
value is placed on comprehensive "intelligence."

The best way I know to learn about a candidate before

acquiring it is to set up a joint task force, which would
consist of managers of both companies, to prepare a de¬
tailed plan for the integration of the acquired company
into the one acquiring it. This, of course, can be done only
when the acquisition is "friendly." An "unfriendly" acqui¬
sition usually precludes the possibility of getting enough
information about the candidate to make good decisions.

Acquisition of a company is like adopting a child. How

that child develops depends on its genes as well as on the
adopting parents. One cannot learn too much about the
genetic characteristics of an adopted child or company.
ACQUISITIONS 15

To use another metaphor: the management of a company

in search of an acquisition should think of itself as piloting
a plane about to take off. It should not rely on its memory
to do all the right things in the right order. It should have a
checklist of things to be done and know what to do if any
of its actions fail to bring about the expected result. In par¬
ticular, it should know when to abort the takeoff.

REFLECTIONS
 5
The Corporation
Reconceived

The Industrial Revolution took place in the United States

in the last century. As industrial enterprises grew in size
and number a conceptual model was needed. Not surpris¬
ingly, the concept that emerged corresponded to the pre¬
vailing concept of the world. Following Newton, the
world was conceptualized as a machine created by God to
do His work; the function of man, as part of that machine,
was to serve God's purposes. Therefore, industrial enter¬
prises were first thought of as machines created by their
"gods," the owners, to serve their purpose in creating
these enterprises, which was to make a profit. An enter¬
prise was seen as having no purpose of its own but as
serving that of its owner. Within an enterprise the owner
was a virtual god, with almost limitless power, subject to

16
THE CORPORATION RECONCEIVED 17

almost no externally imposed constraints. Like a machine,

the ideal enterprise was one that could operate independ¬
ently of its environment.

Workers were viewed and treated as replaceable machine

parts. This treatment was possible because (1) workers
had virtually no source of income other than employment,
(2) they were poorly educated and had low levels of aspi¬
ration, (3) few skills were required, and (4) replacements
were plentiful.

By World War I social and economic development re¬

quired a change in the concept of an enterprise. If enter¬
prises were to take advantage of the opportunities for
growth that had become available to them, they required
external financing. Owners had to choose between retain¬
ing complete control and restricting growth, and sharing
ownership to obtain the resources required to grow as rap¬
idly as possible. Many chose to share ownership and
"went public." "God" disappeared, was dispersed, and
became an abstract spirit. As had occurred when the God
of the Western World disappeared 20 centuries earlier, a
clergy—management—was created to serve as an interme¬
diary between the workers and "god." The managers
knew the will of the owners as the clergy knew that of
God, by revelation, and transmitted it to the workers.

Meanwhile the workers had become more educated, had

the beginnings of social security, were increasingly pro¬
tected by unions, and were more difficult to replace be¬
cause of the greater skills required of them. They could no
longer be treated as replaceable machine parts.
18 THE CORPORATION RECONCEIVED

These conditions led to the reconceptualization of the

business enterprise as a corporation—derived from corpus,
body—therefore, as an organism. Like all organisms, cor¬
porations had an overriding purpose of their own, sur¬
vival, for which growth was taken to be essential. As Peter
Drucker said, profit came to be viewed as oxygen is for a
human being: necessary for its survival but not the reason
for it. The CEO became the head of the firm and manage¬
ment, its brain. Departments were thought of much as
bodily organs and workers, as cells. Their health and
safety became major concerns that were reflected in labor
legislation and contracts. The environment was viewed as
a purposeless self-renewing source of resources and a re¬
ceptacle for waste.

World War II and a permissively raised generation of well-

educated and socially secure workers changed all this.
Government and such special interest groups as ecologists
and consumer advocates began to demand responsible be¬
havior from corporations. Well-educated workers, increas¬
ingly alienated by the machinelike labor required of them,
began to demand more satisfying and challenging work
and opportunities for personal development.

As a result, industrial enterprises are once again being

reconceptualized, this time as social systems. An enterprise
viewed as a social system is considered (1) to be a part of a
larger social system (society) that has purposes of its own
and (2) to contain individuals who have purposes of their
own. The purpose of an enterprise is apparently becoming
one of serving the needs and desires of all its stakeholders,
not the stockholders alone. Survival and growth are in¬
creasingly becoming a means to this end, not ends in
THE CORPORATION RECONCEIVED 19

themselves. When an enterprise increases its ability and

its desire to serve its stakeholders, it develops. Development
is more and more commonly taken to be the appropriate
purpose of an enterprise.

This social-systemic, service-oriented view of a corpora¬

tion is completely different from the organismic view of it
as an entity served by passive environments and stake¬
holders. Today, employees, including managers, are in¬
creasingly believed to have the largest investment in firms
and therefore, to be their most important stakeholders.
From society's point of view corporations are instruments
for producing and distributing wealth, primarily by employ¬
ment. Therefore, the provision of jobs is perceived as one
of their major social responsibilities.

Of what significance is this evolution of the conception of

industrial enterprises? It is this; in a society that views and
treats corporations as social systems, as instruments for serving
their stakeholders' interests, corporations that are managed as
though they were machines or organisms are not likely to survive
and grow.

REFLECTIONS
 6
Mess
Management

Problems are to reality what atoms are to tables. We expe¬

rience tables, not atoms. Problems are abstracted from ex¬
perience by analysis. We do not experience individual
problems but complex systems of those that are strongly
interacting. I call them messes.

Because messes are systems of problems, they lose their es¬

sential properties when they are taken apart. Therefore, if
a mess is disassembled, it loses its essential properties.
Furthermore, as in any system, if each part taken sepa¬
rately is treated as well as possible, the whole is not treated
as well as possible. A system is more than the sum of its
parts; it is the product of their interactions. If taken apart,
it simply disappears. Then how can we formulate a mess
without taking it apart?

20
MESS MANAGEMENT 21

It can be done by the use of reference projections. These are

projections of the performance of an enterprise that are
based on two false assumptions. First, it is assumed that
the organization involved will not change any of its cur¬
rent plans, policies, or practices. If this were true, the or¬
ganization would not be trying to formulate its mess. Sec¬
ond, it is assumed that the organization's environment
will change only as expected; this is obviously false. Under
these assumptions the performance of the organization is
projected into the future. These projections reveal the fu¬
ture implied by the organization's current plans, policies,
and practices: the future it is in.

No matter how successful an organization is, reference

projections taken collectively reveal how it would destroy
itself if it were not to change. These projections reveal the
Achilles heel of the organization. They do this because the
no-change assumption implies no adaptation even to a
predictably changing environment; for example, projec¬
tions were made in 1959 which revealed the impending
crisis of the American automotive industry. By using data
from the preceding 40 years projections were made of (1)
the number of people of driving age in the United States in
the year 2000, (2) the number of cars per person of driving
age, (3) the number of miles driven per car per year, and
(4) the percentage of these miles driven within cities. By
combining these projections an estimate was prepared of
the number of urban automobile miles that would be
driven in the United States in the year 2000 if the industry
continued on its then current path and its environment
changed only as expected.

Next, by using these projections an estimate was made of

22 MESS MANAGEMENT

the additional parking space and lane-miles of streets and

highways that would be needed in the year 2000 to main¬
tain 1960 levels of congestion. Then the cost of their con¬
struction, estimated by using projected construction costs,
revealed that more than 12 times the maximum amount
ever spent per year in the United States for such construc¬
tion would be required for each of the next 40 years. Al¬
though these expenditures were unlikely, they were, in
fact, implicitly assumed in the plans then in force in the
industry.

This was not the mess facing the industry, however. The
mess was revealed by assuming that these large expendi¬
tures would be made. If they were, 117 percent of the sur¬
faces of American cities would be covered by streets, high¬
ways, and parking lots by the year 2000. This, of course,
could not happen. Therefore, continued growth of the au¬
tomotive industry as it had been was not possible. This
was the mess.

What would prevent cities from being covered by streets,

highways, and parking lots? The answer rested in deci¬
sions still to be made. Studies showed that one way to
avoid the mess would be to reduce the size of automo¬
biles. The American automotive industry chose not to do
so at that time. It waited for more than a decade before the
cost of oil, foreign competition, and government require¬
ments forced it to move slightly in that direction. The con¬
sequences of the industry's failure to pay attention to its
mess are well known.

We have to know where we are headed before we can take

action to avoid getting there. Such redirection of an enter-
MESS MANAGEMENT 23

prise requires mess management, not problem solving, and

mess management requires creative and comprehensive
planning.

REFLECTIONS
 7
Growth Versus
Development

Growth and development are not the same thing. Neither is

necessary for the other. A rubbish heap can grow but it
doesn't develop. Artists can develop without growing.
Nevertheless, many managers take development to be the
same as growth. Most efforts directed at corporate devel¬
opment are actually directed at corporate growth.

To grow is to increase in size or number. To develop is to

increase one's ability and desire to satisfy one's own needs and
legitimate desires and those of others. A legitimate desire is
one that, when satisfied, does not impede the develop¬
ment of anyone else.

Development is an increase in capability and competence.

24
GROWTH VERSUS DEVELOPMENT 25

Development of individuals and corporations is more a

matter of learning than earning. It has less to do with how
much one has than how much one can do with whatever
one has. For this reason Robinson Crusoe is a better model
of development than Jean Paul Getty.

Development is better reflected in quality of life than in

standard of living. Therefore, the level of development of
a corporation is better reflected in the quality of work life it
provides its employees than in its profit-and-loss state¬
ment.

If an undeveloped country or corporation was flooded

with money it would be richer but no more developed. On
the other hand, if a well-developed country or corporation
was suddenly deprived of wealth, it would not be less de¬
veloped.

A well-developed country or corporation can do more

with its resources than one that is less developed. This is
not to say that the amount of resources available is irrele¬
vant. Resources can be used to accelerate development
and improve quality of life, but they can best be used for
these purposes by those who are developed.

Growth and development do not have to conflict; they can

reinforce each other. The best evidence that this is hap¬
pening is a simultaneous increase in standard of living and
quality of life. However, there is currently a widespread
belief that quality of life is being sacrificed to increase
standard of living. This belief is accompanied by a willing¬
ness to sacrifice standard of living to improve quality of
26 GROWTH VERSUS DEVELOPMENT

life, a willingness that is reflected in the environmentalist

movement.

A lack of resources can limit growth but not development.

The more developed individuals, organizations, or socie¬
ties becomes the less they depend on resources and the
more they can do with whatever resources they have.
They also have the ability and the desire to create or ac¬
quire the resources they need.

An individual can grow too much. Some people and many

societies believe that a corporation can too. But would any¬
one argue that individuals, corporations, or countries can de¬
velop too much?

REFLECTIONS
 8
The Value of a
Corporation

What justification can there be for combining two or more

businesses in one corporation? The answer is obvious: the
corporation adds value to its parts or, put another way,
the value of the corporation taken as a whole is greater
than the sum of the values of its parts taken separately.
That many corporations do not meet this condition was
made apparent in a recent article in BusinessWeek Quly 8,
1985, pp. 80f):

Breakup value is the new buzzword on Wall Street. It is a

measure of the separate prices that the market would place
on the parts of a company. And the new hot stocks are
those where the sum of the parts amounts to far more than
the current stock price.

27
28 THE VALUE OF A CORPORATION

The same article identified a number of companies for

which a sample of security analysts believed this to be
true. Included were such well known corporations as
Colgate-Palmolive, General Foods, General Mills, Parker
Pen, Ralston Purina, and Revlon.

A corporation can add to the value of its separate busi¬

nesses in a number of ways, of which the following is only
a small sample.

First, it can provide synergy, that is, opportunities for two

or more businesses to cooperate in ways that would not be
possible if they were separate; for example, economies of
scale and increases in effectiveness can be obtained by
sharing legal services, research and development, educa¬
tion and training, purchasing, distribution, and market¬
ing. A corporation may also combine the products of dif¬
ferent businesses into marketable systems.

Second, a corporation can often provide its parts with

more capital at a lower cost than they could obtain if they
were separate businesses.

Third, corporate management can provide guidance that

improves the management of its business units.

Fourth, a corporation may be able to exert greater favora¬

ble influence over the political and social environment of
its individual businesses than they could separately.

One could go on listing the ways in which a corporation

can add value to its parts. For each, however, there is a
corresponding way in which it can decrease that value; for
THE VALUE OF A CORPORATION 29

example, it can impose costly and ineffective services on

them, it can deprive them of the capital they would other¬
wise be able to invest in their own development and
growth, and it can reduce the effectiveness of their man¬
agements by excessive control from the top.

To avoid decreasing the value of its parts, corporate mana¬

gers should engage in periodic dialogue with the mana¬
gers of its parts to determine whether they feel that they
would be more valuable if they were operating independ¬
ently or as part of another corporation. If the management
of a part believes this to be the case, the reasons for it
should be discussed with corporate management. Steps
should be taken to change the views of the part's manage¬
ment or it should be allowed to secede from the union.

Those of us who live in a democracy believe that the right

to emigrate is one of the most important its members en¬
joy. Without it, individuals become instruments of the
state. In a democracy the state should be the instrument of
its members. When individuals no longer feel that the
state or the part of which they are members is adequately
serving their interests, they should have the right to move
to another part or emigrate from the whole. The right to
emigrate, to quit, is given to employees of corporations,
but in many their ability to move between parts is very
much constrained.

Corporations have the right to emigrate from one country

to another, but because a corporation owns its parts, these
parts lack the right to move to another corporation. Parts
of a state are similarly denied the right to withdraw from it
because they too are believed to be owned by it. Yet most
30 THE VALUE OF A CORPORATION

Americans believe that Poland, for example, should have

the right to secede from the Soviet Union, the Kurds from
Iran, or the Basques from Spain. Is this hypocrisy?

Unless parts of a corporate or political union have the right

to secede, they can be enslaved, and those who manage or
govern the whole can be deprived of the feedback that en¬
ables them to evaluate performance effectively. What bet¬
ter and more effective union can there be than one whose
parts have chosen to belong to that union?

REFLECTIONS
 9
The Economy
of the Firm

A curious contradiction is present in the ways we run our

national macroeconomy and the microeconomies of the
business enterprises within it. At the national level we fa¬
vor a market economy that is regulated as little as is com¬
patible with national interests. The economies of most of
our business enterprises, however, are run much like the
national economy of the Soviet Union. They are centrally
planned and controlled, transfer pricing is imposed on the
parts, and when internal sources of products or services
are available they are usually run as bureaucratic monopo¬
lies.

What would a corporation look like if it were run like the

American economy?

31
32 THE ECONOMY OF THE FIRM

Each component would be a profit center free to buy and

sell products and services wherever it wanted to—free,
but subject to as little regulation by corporate management
as required for the good of the whole. Corporate manage¬
ment would have the function of government within the
firm. Moreover, each unit would have to pay corporately
imposed taxes on its profits, interest, or dividends on the
capital obtained from the corporation, and, unless the cor¬
poration believed in free trade, duties on some imports to,
and exports from, the corporation.

Only a few corporations approximate this way of

operating. Why?

Two reasons are usually given. First, centralized planning

and control are said to maximize synergy between the
parts of a corporation. But there is no lack of synergy be¬
tween suppliers, producers, and consumers in the na¬
tional economy. Many corporate producers experience
less conflict and more cooperation with external suppliers
and customers than with those that are internal.

Second, it is argued that some economies of scale enjoyed

by corporations would be lost in a 'Tree corporate
economy"; but such losses are not necessary. Parts of a
corporation can form buying, producing, and selling coop¬
eratives just as independent enterprises do in the national
economy. This cooperation can be encouraged by corpo¬
rate management, just as it is by governments of most free
national economies.

Moreover, a market economy within the firm can avoid

some diseconomies of scale. Large internal monopolies are
THE ECONOMY OF THE FIRM 33

often no more efficient than those that are external and

tend to be less sensitive and responsive to their customers
than smaller internal units that must compete with exter¬
nal suppliers for internal business.

It is possible for corporate management to intervene in the

transactions of the units reporting to it as our government
can but seldom does. Suppose headquarters wants one in¬
ternal unit to buy from another internal unit but the
buying unit does not want to because the internal supplier
demands a higher price than an external source. Corpo¬
rate management can pay the buying unit the difference.
Unless it did, the buying unit would be free to use an ex¬
ternal source. This requirement would discourage corpo¬
rate management from arbitrarily restraining external
trade. It would have to pay for these restraints and there¬
fore would not be likely to impose them unless it believed
the corporation as a whole would benefit.

In a corporate market economy corporate headquarters

would receive income from taxes on unit profits, import
and export duties, and interest or dividends from units for
funds it provided. It would pay for any profit-reducing
constraints it imposed on internal units and for any ser¬
vices it obtained from them. This would make it possible
for the headquarters itself to be a profit center. Its profits
or losses would not be the same as the corporation's.
Therefore, the effectiveness of corporate management
could be evaluated in the same way as that of business
units.

Isn't it time for corporations to practice the same type of economy

they preach to the nation?
34 THE ECONOMY OE THE FIRM

REFLECTIONS
 10
Performance
Measurement

Those corporate managers who can't measure what they

want frequently settle for wanting what they can measure.

Earnings, return on investment or assets, sales volume,

price-earning ratios, cash flow, and market share are
among the more commonly used measures of annual cor¬
porate performance. The deficiencies of these measures,
used separately or in combination, are well known. Nev¬
ertheless, they are widely used because better measures
do not appear to be available.

Efforts to maximize any measure of annual performance

usually sacrifice future performance; for example, one
company in the food business paid its regional managers.

35
36 PERFORMANCE MEASUREMENT

who were responsible for production and sales, a bonus

based on their net profit in the preceding year. These pay¬
ments were as much as five times their-base salaries. To
maximize their annual bonuses these managers increased
prices and reduced product quality, expenditures on
maintenance, and investments in plant modernization.
After several years of these practices the company's price
spreads became so large and its quality differentials so
small that its market shares and volumes suffered
significantly. The regional managers could no longer
maintain their prices, let alone reduce them to meet com¬
petition. Profits went down the drain.

Obviously companies do not want to sacrifice the future

for better current performance, but they do not know how
to measure their future performance. Therefore, they
want what they can measure; last year's performance as
conventionally evaluated. There is an alternative.

The current value of a company is reflected, however im¬

perfectly, in the highest price someone is willing to pay to
acquire it. One who acquires a company is buying its fu¬
ture, not its past. Thus, estimates of the current market
value of an enterprise are based on its expected perform¬
ance, its potential. Companies and their units should try to
increase this potential and their performance should be
measured by changes in it.

A company that invests heavily in its future and pays for it

with a reduction in current profits is much to be preferred
to one that has allowed its future to deteriorate to increase
current profits.
PERFORMANCE MEASUREMENT 37

Clearly, a company's performance potential cannot always

be estimated accurately. Nevertheless, it is estimated by
anyone who contemplates acquiring it. Experts are some¬
times used to prepare these estimates. Whether done in¬
ternally or externally, they often turn out to be quite good.
Companies can and should use these estimates to evaluate
their own and the performances of their parts.

These measures may not be as accurate as those com¬

monly used to evaluate past performance, but it is better to
use imprecise measures of what is wanted than precise measures
of what is not.

REFLECTIONS
 11
Mission
Statements

Most corporate mission statements are worthless. They

consist largely of pious platitudes such as: "We will hold
ourselves to the highest standards of professionalism and
ethical behavior." They often formulate necessities as ob¬
jectives; for example, "to achieve sufficient profit." This is
like a person saying his mission is to breathe sufficiently.
A mission statement should not commit a firm to what it
must do in order to survive but to what it chooses to do in
order to thrive. Nor should it be filled with operationally
meaningless superlatives such as biggest, best, optimum,
and maximum; for example, one company says it wants to
"maximize its growth potential," another "to provide
products of the highest quality." How in the world can a

38
MISSION STATEMENTS 39

company determine whether it has attained its maximum

growth potential or highest quality?

To test for the appropriateness of an assertion in a mission

statement, determine whether it can be disagreed with
reasonably. If not, it should be excluded. Can you imagine
any company disagreeing with the objective “to provide
the best value for the money." If you can't, it's not worth
saying.

What characteristics should a mission statement have?

First, it should contain a formulation of the firm's objectives that
enables progress toward them to be measured. To state objec¬
tives that cannot be used to evaluate performance is hy¬
pocrisy. Unless the adoption of a mission statement
changes the behavior of the firm that makes it, it has no
value.

The behavior of a Mexican firm was profoundly affected

by the following passage from its mission statement:

To create a wholesome, varied, pluralistic, multiclass rec¬

reational area incorporating tourist facilities and perma¬
nent residences, and to produce locally as much of the
goods and services required by the area as possible, so as
to improve the standard of living and quality of life of its
inhabitants.

Second, a company's mission statement should differentiate it

from other companies. It should establish the individuality, if
not the uniqueness, of the firm. A company that wants
only what most other companies want—for example, “to
manufacture products in an efficient manner, at costs that
40 MISSION STATEMENTS

help yield adequate profits"—wastes its time in

formulating a mission statement. Individuality can be at¬
tained in many ways, including that in which a company's
business is identified.

Third, a mission statement should define the business that the

company wants to be in, not necessarily is in. However di¬
verse its current businesses, it should try to find a unifying
concept that enlarges its view of itself and brings it into
focus; for example, a company that produces beverages,
snacks, and baked goods and operates a variety of dining,
recreational, and entertainment facilities identified its
business as "increasing the satisfaction people derive from
use of their discretionary time." This suggested com¬
pletely new directions for its diversification and growth.
The same was true of a company that said it was in the
"sticking" business, enabling objects and materials to stick
together.

Fourth, a mission statement should be relevant to all the firm's

stakeholders. These include its customers, suppliers, the
public, shareholders, and employees. The mission should
state how the company intends to serve each of them; for
example, one company committed itself "to providing all
its employees with adequate and fair compensation, safe
working conditions, stable employment, challenging
work, opportunities for personal development, and a sat¬
isfying quality of working life." It also wanted "to provide
those who supply the material used in the business with
continuing, if not expanding, sources of business, and
with incentives to improve their products and services and
their use through research and development."
MISSION STATEMENTS 41

Most mission statements address only shareholders and

managers. Their most serious deficiency is their failure to
motivate nonmanagerial employees. Without their com¬
mitment, a company's mission has little chance of being
fulfilled, whatever its managers and shareholders do.

Finally, and of greatest importance, a mission statement

should be exciting and inspiring. It should motivate all those
whose participation in its pursuit is sought; for example,
one Latin American company committed itself to being
"an active force for economic and social development, fos¬
tering economic integration of Latin America and, within
each country, collaboration between government, indus¬
try, labor and the public." A mission should play the same
role in a company that the Holy Grail did in the Crusades.
It does not have to appear to be feasible: it only has to be
desirable:

. . . man has been able to grow enthusiastic over his vision

of . . . unconvincing enterprises. He has put himself to
work for the sake of an idea, seeking by magnificent exer¬
tions to arrive at the incredible. And in the end he has
arrived there. Beyond all doubt it is one of the vital sources
of man's power, to be thus able to kindle enthusiasm from
the mere glimmer of something improbable, difficult, re¬
mote."^

If your firm has a mission statement, test it against these

five criteria. If it fails to meet any of them, it should be
redone.

*Jose Ortega y Gasset, Mission of the University, Norton, New York, 1966, p. 1.
42 MISSION STATEMENTS

If your firm has no mission statement, one should be pre¬

pared and as participatively as possible. An organization
without a shared vision of what it wants to be is like a trav¬
eler without a destination. It has no way of determining
whether it is making progress.

REFLECTIONS
 12
Human
Effectiveness

The effectiveness of human behavior depends on the

value of its outcome. Since different parties can place dif¬
ferent values on the same outcome, individuals may be ef¬
fective in relation to their own objectives but not to the ob¬
jectives of the organization of which they are part, or vice
versa.

Consequently, when evaluating a person's effectiveness it

is essential to ask: Effectiveness for whom? If the objec¬
tives of employees are in conflict with those of the organi¬
zation of which they are part, then they behave effectively
for themselves or the organization but not for both. There¬
fore, the effectiveness of employees can be maximized only if their
values and those of the organization are not in conflict.

43
44 HUMAN EFFECTIVENESS

An industrial enterprise places value on employees' pro¬

ductivity and the quality of their output. Unless employees
themselves value the amount and quality of their output,
they may not behave as effectively as they might from the
organization's point of view. Maximization of employees'
effectiveness, however, does not require that they value
productivity and product quality as ends in themselves.
This is not necessary if they value the organization and believe
that it values productivity and product quality. They will try to
provide the organization with what it wants. People do
things for friends that they prefer not to do, but they do
them because of the satisfaction they believe their actions
give their friends.

The amount of satisfaction we derive from satisfying oth¬

ers is directly proportional to the amount of satisfaction
they provide us. This is the reason that quality-of-work-
life programs are directed at making work more satisfying
and fulfilling. Such work makes employees value the or¬
ganization that provides it, even if they do not value its
ends.

Because industrial enterprises want more productivity and

quality of product from employees, measures of these var¬
iables are commonly used as measures of human effec¬
tiveness. Such measurement can be difficult, however.
Productivity and quality of output are affected by many
things other than human effort; for example, the technol¬
ogy and the quality of the raw materials used. Separation
of an individual's contribution from that of a change in
technology or raw material may not be easy but it is sel¬
dom impossible.

Measurement of the effectiveness of an organization from

HUMAN EFFECTIVENESS 45

the point of view of its employees is more difficult. It is

usually done indirectly by use of such indicators as absen¬
teeism, lateness, incidents of sabotage, attrition rates, ac¬
tions requiring discipline, and accidents. In recent years
surveys of the attitudes of employees have been used in¬
creasingly. These surveys are often conducted annually to
detect changes in attitudes and their sources.

Quality-of-work-life programs are directed toward im¬

proving attitudes and, by such improvement, toward
increasing productivity and quality of product. These pro¬
grams are not cost-free. Although their advocates claim
their costs are more than justified, every organization
should determine for itself whether this is the case. It can
be done by small-scale experiments in which troublesome
work groups, white- or blue-collar, are exposed to these
programs. Then measures of effectiveness of the individu¬
als involved from the point of view of the organization
and of the organization from the point of view of the indi¬
viduals can be made before and after initiation of the pro¬
grams. Repeating these measurements periodically and
analyzing them helps an organization learn how to in¬
crease its own effectiveness and that of its employees.

REFLECTIONS
 13
Quality of Work Life

Many corporate efforts are being made to increase the pro¬

ductivity of labor and the quality of its output. Some of
these efforts have been successful; some have not. My ex¬
perience suggests that certain characteristics make the dif¬
ference, characteristics that convert into 10 command¬
ments. Violation of any one of them may be enough to
spell failure.

First, labor-oriented programs should not focus on increasing

productivity and product quality but on the satisfaction derived
from work. They should focus on the quality of work life, not
the quantity and quality of work. If job satisfaction is in¬
creased, productivity and product quality usually increase

46
QUALITY OF WORK LIFE 47

by considerably more than the programs that focus on

them can bring about.

Second, the quality of the work life of managers at all levels

should be improved before labor is dealt with. Wherever this
has been done subsequent programs for the workforce
have been better supported by managers at all levels.
Wherever this has not been done middle and lower levels
of management often obstruct programs directed at labor.

Third, once managers are taken care of the quality-of-work-life

program should be extended to cover all white- and blue-collar
workers. Participation in these programs should be volun¬
tary for everyone but managers; their participation should
be a condition of their employment.

Fourth, managers and workers should be prepared for these pro¬

grams by receiving instruction in cooperative group processes,
and this should be done on company time. Effective groups sel¬
dom come about naturally.

Fifth, once the programs are initiated, discipline should be di¬

rected at correcting undesirable behavior rather than punishing
it. Punishment seldom produces good behavior that lasts,
and even less frequently produces a satisfied worker.

Sixth, discipline of those who break the rules should be applied

regardless of rank. No double standards; the discipline ap¬
plied to managers and nonmanagerial personnel should
be the same.

Seventh, commitments made by any participant in a quality-of-

48 QUALITY OF WORK LIFE

work-life program should be followed up and met. If a partici¬

pant cannot meet a commitment, a complete explanation
for the failure should be given to the other parties.

Eighth, the breaking up of labor-management meetings into cau¬

cuses should be discouraged. To the extent possible, all
discussion should take place openly in joint meetings.

Ninth, an experienced and competent third party should be en¬

gaged, initially at least, to facilitate joint meetings and to pro¬
vide general guidance in the process. The third party should
be subject to approval by the other two and dismissible by
either one.

Tenth and most important, the quality-of-work-life programs

should not be designed or redesigned by experts but by those di¬
rectly affected by them. Lower level employees are wary of
experts, especially those employed by management, and
suspect their programs of being instruments of exploita¬
tion. Unfortunately, their suspicions are often justified.

REFLECTIONS
 14
Incentives

Incentives are widely used to influence behavior that can¬

not be controlled. Frequently, however, they do not work
or work in unintended ways; for example, a large pro¬
ducer of home appliances also installs and services these
appliances. Company employed servicemen equipped
with vans call at the homes of users requiring service. The
company's management became concerned about the
large parts inventory carried in these vans, particularly be¬
cause servicemen normally used only a few parts per day.

A research organization was employed to try to reduce

this mobile inventory. Researchers learned that the serv¬
icemen were paid for each repair call they made. If they
did not have the parts to complete an installation or repair.

49
50 INCENTIVES

they had to go to a warehouse to get them. This could take

two to four hours for which they received no compensa¬
tion. No wonder they carried every part they might possi¬
bly need. The way the repairmen were paid was an
unintended incentive to their carrying large numbers of
parts.

Some incentive systems are seriously misconceived be¬

cause their objectives are not clearly formulated; for exam¬
ple, two major cities decided independently to reduce
traffic congestion by charging automobiles and trucks for
using the streets. They developed very different proposals
for imposing these charges. One planned to base them on
the number of miles driven within the city, the other, on
the frequency with which vehicles appropriately marked
passed electronic eyes that could read identification num¬
bers inscribed on their sides. An argument developed be¬
tween the transportation planners of these cities regarding
the merits of each system. They decided to employ a re¬
search organization to make a comparative evaluation of
their designs.

At the planners' first meeting with the researchers one re¬

searcher asked for the purpose of the proposed systems.
Surprised by an apparently naive question, a planner ex¬
plained that the objective was obviously to reduce conges¬
tion. The researcher then asked, "What is congestion?"
An increasingly annoyed planner replied, "Everybody
knows what congestion is." The persistent researcher said
"Not everybody; I don't." The thoroughly impatient plan¬
ner rose, went to a window, pointed down to the street,
and said, "Look, you can see it." The researcher then
INCENTIVES 51

asked the planner what he saw. The planner blurted, "For

God's sake, the cars aren't moving." "Ah," said the re¬
searcher, "then what you want to reduce is the number of
cars not moving on the streets. Then why are you
proposing to charge them for moving?"

The researcher then pointed out to the flabbergasted plan¬

ner that if he wanted to reduce congestion, it would be
better to charge drivers for the number of times they
stopped than for how much they moved. Such a charge,
he pointed out, would induce them to use their vehicles at
off hours and on less congested routes. Moreover, stops
could be counted less expensively (by a simple inertial me¬
ter) than moving could be measured.

Another city, connected to most of its suburbs by bridges

and tunnels, wanted to reduce the number of cars driven
by commuters. It proposed to increase the average charges
by imposing a fee per car plus a fee for each passenger in
each car. A consultant pointed out that this would induce
drivers not to carry passengers, hence would increase the
number of cars. To reduce the number of cars using
bridges and tunnels, he said, a fee should be charged for
each empty seat. This would induce car pooling and the use
of smaller cars. (Cars carrying three or more passengers
can now travel at no cost in express lanes on the San
Francisco Bay Bridge.)

Because incentives so frequently produce unexpected re¬

sults, it is worth having them reviewed critically by at least
a sample of those they are intended to affect, and this
should be done before they are implemented.
52 INCENTIVES

Incentives will not serve their intended purposes unless they also
serve the purposes of those they are intended to affect.

REFLECTIONS
 15
The Resurrection of
A&P

Between 1974 and 1982 the Great Atlantic & Pacific Tea
Company (A&P) closed about 2500 of its stores; 60 of them
were in the Philadelphia area. Blame was placed on labor
costs, which had been about 5 percent higher than the in¬
dustry average.

Local 1357 of the United Food and Commercial Workers, a

major victim of these closings, lost thousands of members.
Wendell Young, the local's president, asked the Busch
Center of The Wharton School to determine whether it
would be possible for some of the workers who had lost
their jobs to buy and operate stores abandoned by A&P. A
study was conducted with positive results. When the local
issued a call for volunteers 600 members offered to put up

53
54 THE RESURRECTION OF A&P

$5000, the amount required of each, and two stores were

purchased from A&P. Helped by the Busch Center, the
worker-owners prepared a detailed design of the opera¬
tion and management of the stores. From the day they
opened the O&O (Owned and Operated) stores showed
better performance than the A&P stores had realized in
the same sites.

O&O's success prompted A&P to reevaluate its failure and

to suggest a study to determine whether their stores could
be redesigned to make them viable. With the help of the
Busch Center a joint effort that involved company man¬
agement, the local, and a number of workers was
launched. A Quality of Work Life Plan was produced, on the
basis of which A&P agreed to reopen 20 stores. By mutual
agreement these stores were to be part of a new subsidiary
of A&P called "Super Fresh." The following conditions
were critical parts of the agreement:

1. The workers accepted shorter vacations and pay cuts

up to $2 an hour.

2. The workers would receive 1 percent of gross sales if

the labor costs were no more than 10 percent of
operating revenues and a higher percentage if they
were below 9 percent.

3. Management committed itself to a Quality-of-Work-

Life program, the foundation of which was em¬
ployee involvement. All employees were to be mem¬
bers of planning boards, which were to be
established for each work unit at every level of the
organization and which would have a significant
THE RESURRECTION OF A&P 55

amount of control over the operation of their depart¬

ments, stores, or regions.

At the stores that have since been reopened workers have

the right to make dozens of decisions that were once ex¬
clusively in management's hands. They participate in de¬
cisions regarding layout, allocation of shelf space, items to
be carried, pricing and promotions, and assignment of
personnel. Workers have changed store hours, replaced
direct-mail advertising with newspaper advertising, and
discontinued slow-moving items. One store's meat de¬
partment employees suggested changing the meat display
to add more soup bones next to the stew beef. Not only
did sales improve, but customers did not "ring the bell" so
often. Prompted by employees, an intercom system was
installed at registers to allow check-out clerks to price an
unmarked item with less delay.

The Wall Street Journal of September 29, 1983, quoted

Gerald Goode, president of Super Fresh, as saying: "We
don't have employees at our stores. We have associates."

Twenty-four Super Fresh stores, employing 2105 workers,

were opened in the last five months of 1982. These stores
reported record sales and profits from the start. The chain
is growing more rapidly than any other in the Delaware
Valley.

. . . while competitors complain. Super Fresh continues to

open stores at such a rapid clip that it is beginning to run
out of prime locations. As of this week, 45 of the 79 area
markets that once were A&P stores have reopened as Su-
56 THE RESURRECTION OF A&P

per Fresh markets and eight to 12 more are expected to be

opened by July 1. {Philadelphia Inquirer, April 3, 1983.)

Little wonder that the Super Fresh format is now being

used to convert existing A&P stores and to reopen others
that had been closed.

The resurrection of A&P as Super Fresh suggests, first,

new collaborative and entrepreneurial roles for unions in
the revitalization of dying or dead businesses; second, that
managers are learning how to make better use of the
knowledge and good will of their workers; and, finally,
that third parties can enable two parties locked in mortal
combat to find creative win-win ways of dissolving their
differences.

REFLECTIONS
 16
Paternalism

A large number of business enterprises are still seeking to

become a “great big happy family." Not surprisingly,
these organizations tend to be managed paternalistically
by an “old man" at the top. Authority is concentrated in
the “Father" who dispenses it downward. Such chief ex¬
ecutives can revoke or revise any decisions made at any
level below them. Therefore, decisions to be made tend to
be pushed up the organizational ladder as far as possible.
They are often made by managers who are far removed
from the area directly affected and who lack the informa¬
tion and knowledge required to make them well.

Patronage is commonplace in paternalistic organizations.

Who one knows is usually more important than what one

57
58 PATERNALISM

knows. This provides those who are patronized with a

convenient way of avoiding responsibility. By obtaining
approval from above before acting they absolve them¬
selves of responsibility for what they eventually do. Their
guiding principle seems to be: Don't make any decision that
you can get someone higher to make for you. As a result, even
routine decisions often require inordinate amounts of
time. In contrast, requests from above are acted on at
once. The higher the source of a request, the less time re¬
quired for a response.

In a paternalistic organization time is allocated to work

through a priority system based on the rank of the person
requesting it. This raises havoc with schedules and ap¬
pointments. Anyone can be bumped from a schedule be¬
cause someone higher wants that time. (The higher au¬
thority is never informed of the "bump.") Appointments
are often canceled at the last moment or not kept. Lateness
is the rule. Waiting is an organizational pastime. The
higher the rank of a manager, the larger the waiting room.

The "Father" is often a workaholic who puts in long days

and weeks and seldom takes a vacation. His work sched¬
ule has a rippling effect on the rest of the organization, for
he expects his subordinates to be there whenever he
wants them. As a result, the family life and health of sub¬
ordinates frequently suffer. Their advancement often de¬
pends more on the amount of time they spend at work
than on the amount of work they do.

The principal shortcoming of paternalistic organizations is

PATERNALISM 59

revealed in an apparently unrelated episode that occurred

in a university's psychological clinic. A four-year-old son
of two university professors was brought in by his parents
because, they said, he hardly ever spoke. He did, how¬
ever, understand all that was said to him. The parents
thought their child might be retarded.

The clinician put the child through a battery of nonverbal

tests and found him to have superior intelligence. These
results, together with information obtained by monitoring
the behavior of the parents in the child's presence, sug¬
gested a treatment. The clinician had observed that when¬
ever he asked the child a question one of the parents im¬
mediately answered for him. The clinician advised the
parents not to talk to or for the child for the next few
weeks. The parents followed this advice. When they re¬
turned to the clinic the child was speaking fluently.

Paternalism breeds paralysis of the tongue, if not the

mind. It ignores the fact that it is often better to do things
for one's self, no matter how badly, than to have them
done by others, no matter how well. People generally
learn more from their own mistakes than they do from the
correct decisions of others. To be sure, in some cases "fa¬
ther knows best," but it should be borne in mind that fa¬
ther is fallible.

Moreover, paternalistic organizations are even more fallible

than their fathers.
60 PATERNALISM

REFLECTIONS
 17
Corruption

Although corruption is widely believed to be evil, it is

widely tolerated. Like the weather, more is said than done
about it.

The damage to the development efforts of organizations

and societies that is caused by corruption is often well hid¬
den. It can do much more than obstruct these efforts; it
can completely subvert them. Consider the case of
CONASUPO, the National Basic Commodity Agency of
Mexico.

One of the purposes for which CONASUPO was created

was to increase the income of peasants (campesinos). It ini¬
tiated a price-support program for such crops as corn.

61
62 CORRUPTION

wheat, and beans. Each year, before these crops were

planted, CONASUPO announced prices at which they
would be purchased by the government if they met mini¬
mal quality standards. These prices were higher than
those paid by local buyers who as a rule were affluent mer¬
chants and often the only source of provisions and sup¬
plies and the credit required to buy them. Loans were
made at exorbitant rates. In addition, these merchants
were the only providers of trucks with which to haul pro¬
duce to markets. Little wonder, then, that they were usu¬
ally the local heads (caciques, or chiefs) of the ruling politi¬
cal party, PRI. Needless to say, their incomes were
seriously threatened by CONASUPO's price-support pro¬
gram, but not for long.

CONASUPO set up a large number of rural buying sta¬

tions to which campesinos could take their produce. Con¬
sistent with the widespread practice of patronage in
Mexico, the casiques were asked to nominate candidates
for the job of tending these stations. Their candidates were
selected more often than not.

When the program began many campesinos took their

small harvests to CONASUPO's buying stations only to
have them rejected by the attendants who claimed that
they did not meet the minimal quality standards. The cam¬
pesinos then had no alternative but to offer their produce
to the casiques, who, having been dutifully informed by
attendants of its rejection, pretended to be reluctant to
buy. Eventually, they "gave in" to the pleas of the campe¬
sinos and bought it, but at a much lower price than they
had paid before CONASUPO's program was initiated and
CORRUPTION 63

not without lecturing the campesinos on the cost of disloy¬

alty.

When the peasants withdrew, poorer and penitent, the

casiques took the produce they had bought to the buying
stations of CONASUPO where it had previously been re¬
jected and sold it to the government at a much higher
price than they had paid. For this the attendants were ap¬
propriately rewarded.

The casiques and the buying-station attendants prospered

for many years until a new administrator of CONASUPO
did something about it.

Even the best intentioned development programs can be

subverted by corruption, and this is as true of business en¬
terprises as it is of societies; for example, one large corpo¬
ration recently made a friendly acquisition of another com¬
pany as part of its diversification strategy. Both companies
would have benefited if the acquisition had been made at
the price the acquiring company had expected to pay for
the acquired company's stock. But a member of the
acquiring company's board told some of his friends about
the intended acquisition and substantial amounts of the
stock of the company about to be acquired were bought.
This drove up the price of that stock. The increased price
of acquisition diluted the acquiring company's earnings
significantly. As a result, the amount that could be in¬
vested in improvements of the acquired company was re¬
duced and the development of both companies was re¬
tarded.
64 CORRUPTION

The only thing more detrimental to organizational development

than corruption is tolerating it.

REFLECTIONS
 18
Alcoholism and Stress

Alcoholism is widely known as a cause of problems in the

workplace. What is not widely known is that alcoholism is
strongly related to stress produced by the workplace.

Trevor Williams, George Calhoun, and P studied 65 adults

at a treatment center for alcoholics and compared them
with a matched sample of 69 nonalcoholics. We looked for
evidence of differences between these groups in fre¬
quency, strength, and duration of stress.

We defined stress as a condition of people who (1) have

certain expectations of themselves or others or are aware
of others' expectations of them, (2) believe that these ex¬
pectations are reasonable, and (3) believe that these expec-

*“Stress, Alcoholism, and Personality, Human Relations," 35 (6), 491-510, 1982.

65
66 ALCOHOLISM AND STRESS

tations are not being met and are not likely to be met in the
future; for example, stress may be caused by failure to
reach our level of aspiration or the level that others have
for us. It can also be caused by our failure or that of others
to do a job as well as we think it should be done.

The questionnaire we used to get at the frequency,

strength, and duration of stress differentiated stress
arising (1) at work, (2) at home, (3) in social life, (4) from
self-imposed expectations, and (5) from expectations of
others.

Alcoholics identified far more unmet expectations from

every source than nonalcoholics. Alcoholics responded to
their failure to meet relevant expectations with more anxi¬
ety than nonalcoholics. The anxiety they experienced was
stronger than that of nonalcoholics. The peak levels of
anxiety provoked by unmet expectations lasted longer in
alcoholics than in nonalcoholics. (Among black alcoholics
94 percent reported persistent or continuous stress, com¬
pared with only 56 percent of black nonalcoholics.)

Of the alcoholics interviewed, 66 percent reported strong

or very strong anxiety at work, whereas only 21 percent of
the nonalcoholics did. The workplace was revealed as a
major source of stress.

We also investigated the ways in which alcoholics and

nonalcoholics cope with stress. We classified coping be¬
havior in five categories:

1. Consumption, including drinking, eating, smoking,

and taking drugs or medication.
ALCOHOLISM AND STRESS 67

2. Aggression, starting a fight or expressing overt hostil¬

ity.

3. Escape, getting away from the source of stress; for ex¬

ample, quitting one's job or leaving home.

4. Self-distraction, engaging in unrelated physical or so¬

cial activities to take one's mind off the problem.

5. Miscellaneous constructive activities, including seeking

advice from others, working through one's problem
on one's own, solving difficulties by cooperation
with others, and praying.

Even with this crude classification of coping behavior we

found striking differences among the ways in which alco¬
holics and nonalcoholics cope with stress. More than half
the alcoholics preferred one of the destructive strategies,
consumption or aggression, whereas more than two-
thirds of the nonalcoholics preferred the two constructive
options (self-distraction and miscellaneous constructive
activities). About one-fifth of each group chose the neutral
strategy of escape.

More alcoholics (88 percent) than nonalcoholics (32 per¬

cent) drank to relieve stress. The contrast was even greater
among whites: more white alcoholics drank (93 percent)
and smoked (83 percent) to reduce stress than
nonalcoholics (31 and 35 percent). More black and white
nonalcoholics (39 percent), however, used eating to cope
with stress than alcoholics (15 percent). More white alco¬
holics (41 percent) used drugs or medication to reduce
stress than white nonalcoholics (0 percent).

Summarizing, we found that in every area measured alco-

68 ALCOHOLISM AND STRESS

holies reported more frequent, more prolonged, and sev¬

erer stress than nonalcoholics. Although work was not the
only source of stress, it was important, particularly for
blacks. Finally, alcoholics tend to cope with stress more
self-destructively than nonalcoholics.

Our work did not show that stress causes alcoholism but it
did show them to be strongly associated. Therefore, we
cannot claim that a reduction in stress will reduce alcohol¬
ism but it does seem likely. Given the strong association of
stress and alcoholism and the very high cost of alcoholism
in the workplace, it is worth some time, money, and effort
to reduce the stress generated. For this reason an
increasing number of companies have instituted employee
assistance programs in which employees who suffer from
alcoholism, drug abuse, and other personal problems can
receive expert counseling and treatment. These programs
have been shown to reduce alcoholism and drug abuse
and have more than justified their cost by increasing job
satisfaction, productivity, and product quality.

REFLECTIONS
 19
Too Much
Communication?

Too much of a good thing, even communication, is a bad

thing. Unconstrained communication within corporations
can improve their performance only when their parts are not
in conflict—when their objectives are compatible and mu¬
tually reinforcing. It is apparent that in war the more op¬
ponents know about one another, the more harm they can
inflict. If each side knew absolutely nothing about the
other, war could not be waged.

It is commonplace for parts of a corporation to be at war

among themselves, or in competition. I have heard attrib¬
uted to Peter Drucker the observation that there is more
competition within corporations than between them and
that internal competition is often waged less ethically than

69
70 TOO MUCH COMMUNICATION?

external competition. When parts of the same organiza¬

tion compete, increased communication can reduce the
performance of the organization as a whole and its parts,
and obstruct growth and development.

This is illustrated by a considerably simplified version of a

real case, but none of its essential properties is omitted. It
involved a department store in which the two most impor¬
tant activities are buying and selling.

The store's chief executive decided to engage in "manage¬

ment by objectives." In his negotiation with the
purchasing manager the executive proposed the following
objective: "minimization of stock." The purchasing mana¬
ger pointed out that this could be accomplished by buying
nothing. They eventually settled on an alternative: "mini¬
mization of stock subject to the requirement that the store
be able to meet expected demand."

In a similar session between the executive and the sales

manager the sales objective was formulated as "maximiza¬
tion of gross sales, less selling expenses."

To facilitate pursuit of his objective the sales manager had

his statistical staff derive price-demand curves from data
gathered for each major product category. Each plot
showed three curves (Figure 1):

1. The optimistic curve representing the most they could

expect to sell at each price.

2. The realistic curve representing the average amount

previously sold at each price.
 TOO MUCH COMMUNICATION? 71
Ooantity

3. The pessimistic curve representing the least they

could expect to sell at each price.

Once the sales manager had selected a price (Pj) at which

he intended to offer an item, he used the appropriate chart
to determine how many he would need in stock (Figure 2).
To do so he used the optimistic curve because he wanted
to minimize sales lost because of lack of stock. Lost sales
hurt his performance but excess inventory didn't. The
sales manager then told the purchasing manager that he
would need quantity Qi of the item.

The purchasing manager, who had previously served as

assistant sales manager, also had access to the
price-demand curves. He was told by his former subordi¬
nates in the sales department that the sales manager had
used the optimistic curve; therefore, he adjusted the num-
72 TOO MUCH COMMUNICATION?
Qucanl’vty

ber required downward to Q2 by using the realistic curve

(Figure 3). He wanted only enough stock to meet expected
demand. He let his old boss, the sales manager, know
about this adjustment.

The sales manager then returned to the curves and ad¬

justed the price upward to P2 to maximize gross sales,
given the amount of stock that would be available (Figure
4). The purchasing manager learned about this change
and made a further reduction in the stock to Q3 (Figure 5).

Had this process continued, nothing would have been

bought or sold. It did not continue because the chief exec¬
utive intervened by prohibiting communication about
prices and stock levels between the sales and purchasing
managers. This enabled the store to survive but it did not
 Ps

Price Figure 3.
Ouan^'i+y

Price Figure 4.

73
74 TOO MUCH COMMUNICATION?
Quan^i^y

Price Figure 5.

thrive. It was not until the conflict between purchasing

and selling objectives was removed that unconstrained
communication could be permitted without hurting per¬
formance.

The moral of this fable is this: Don't let enemies within a cor¬
poration communicate with one another, no matter how friendly
they are. Better still, eliminate them or the conflict between
them.

REFLECTIONS
 20
Communicating Up

Communicating up in an organization is widely recog¬

nized as more difficult and less effective than
communicating down. To improve upward communica¬
tion organizational gravity must be overcome. This re¬
quires removing the belief of most managers that their
subordinates, unlike those of other managers, communi¬
cate with them easily and effectively. This is seldom the
case.

One CEO I know learned this the hard way. He spent sev¬
eral weeks of each year making “state-of-the-corporation"
presentations to all his company's hourly paid workers. It
was a shirt-sleeve presentation, which was followed by
time for questions and discussion. These sessions were al¬
ways lively and gratifying to the CEO.

75
76 COMMUNICATING UP

One year, shortly after completion of these sessions, a

company-wide strike was called. The puzzled CEO asked
me why none of the strike issues had been raised in any of
his communication sessions. "Because," I answered, "the
meetings were yours, not theirs." He asked me to explain.
I told him that if he wanted to hear what the workers
wanted to say, he would have to reverse roles by allowing
them to organize communication sessions that he would
attend. Most of what workers want to say cannot be put
into short questions or comments. Even if this were possi¬
ble it would be unlikely when workers must talk up to a
boss standing in a pulpit. They must be in the pulpit and
talk down to him. I asked the CEO if he thought he could
communicate with the workers if he were part of a large
audience listening to a presentation they had prepared.
He got the point.

Subordinates seldom say what is on their minds even

when they organize a session with their superiors. They
often fear the consequences of doing so; but the openness
with which they communicate to superiors can be in¬
creased in several ways.

First, a third party, one who is respected by subordinates

and superiors, can help by making the presentation for the
subordinates. To do this effectively the third party has to
participate in the discussions in which the subordinates
decide what they want to tell their superiors.

Second, sessions run by subordinates should not be used

for bitching but for suggesting what superiors can do to
enable them to do their work better and with more satis¬
faction. These sessions should be devoted to discussing
COMMUNICATING UP 77

what the subordinates want, not what they do not want.

Getting rid of what they do not want does not necessarily
get them what they want; for example, getting rid of an
unwanted supervisor does not necessarily get them one
they do want.

Third, meetings conducted by subordinates can lead to

improvement if their superiors respond positively then
and there to as many of the proposals as they can. They
should commit themselves to responding to the other pro¬
posals by a designated time and stick to that commitment.
Finally, when superiors cannot accept a proposal they
should give a full explanation of their rejection.

Unfortunately, many managers underestimate the intelli¬

gence and goodwill of their subordinates. Those managers
who get over this hurdle can communicate with their sub¬
ordinates at least as well as they do with their peers.

Good one-way communication in organizations is not pos¬

sible. Good communication must be two-way. The “com"
in 'communicate' means together.

REFLECTIONS
 27
Jargon

The jargon commonly used by technicians frequently cre¬

ates a chasm between them and managers.

Oscar Wilde once wrote that the United States and

England are two great countries separated only by a com¬
mon language. Jargon is an uncommon language that can
separate even members of the same nation. Worse yet, it
can separate people from themselves; for example, a while back
I had a brilliant student, now a well-known professor at a
leading university, who wrote a highly technical doctoral
thesis on which he was examined by a committee of five
faculty members. I chaired that session.

It was apparent from the beginning of the examination

78
JARGON 79

that the candidate knew more about the subject of his

thesis than any of his examiners. He answered our ques¬
tions with a display of technical pyrotechnics that left us in
awe.

As chairman I was the last to question him. I asked him to

assume that I was an ordinary corporate manager who
wanted to know what his thesis was about. Would he
please explain it to me briefly?

He went to the blackboard and began to cover it with

mathematical symbols. I stopped him to remind him that 1
was an ordinary manager, not a mathematician. “Oh," he
said, "that kind of manager." He stood thoughtfully for a
moment, then started over, but once again resorted to
mathematical jargon. This time he stopped himself. After
a long pause he said, "Tm sorry, but I can't do it. The
thesis is too technical to be explained in nontechnical lan¬
guage."

"No," I said, "I think there's a different reason." After

some thought he said, "1 guess 1 can't do it because I don't
understand what I've done well enough to explain it in
nontechnical language." This time he was right.

He asked for an adjournment of the examination until he

was able to deal with my question. It was granted. He re¬
turned in a week and did a great job.

Unless people can express themselves well in ordinary

English, they don't know what they are talking about. It is
only in ordinary English that we can communicate effec-
80 JARGON

tively with others and with ourselves. Communicating

with ourselves is the most important kind of communica¬
tion.

Jargon is noise that keeps our brains from understanding what

our mouths are saying.

REFLECTIONS
 22
Computer-Controlled
Managers

An increasing number of computer-based systems are be¬

ing used to assist management. Managers should under¬
stand these systems well enough to know if they are doing
what they are supposed to. If managers don't, they run
the serious risk of being controlled by, rather than
controlling, these systems; for example, a CEO of a large
corporation once asked me to evaluate a document
submitted to him by his computing center. It was a pro¬
posal for updating the computing equipment used primar¬
ily for a large production and inventory control system. I
asked him why he hadn't evaluated the proposal himself.
He said that he had tried but that he didn't understand the
proposal or the system involved. I asked him why he
hadn't had those who prepared the proposal explain it to

81
82 COMPUTER-CONTROLLED MANAGERS

him in ordinary language. He told me that this would

have revealed his ignorance.

We made a bargain: I agreed to make the evaluation if he

would attend a course that would enable him to make
such evaluations for himself in the future.

Arrangements were made for the designers and operators

of the production-inventory-control system to present
their proposal to me orally.

The system they described to me at a subsequent meeting

was supposed to control the production and purchasing of
a large number of parts made or used at multiple loca¬
tions. Nevertheless, its underlying logic was quite simple.
On the basis of the previous use of each part and the time
required to acquire it once it had been ordered, the system
monitored its stock level and determined the level at
which it should be reordered and the quantity required.

I asked the operators of the system if many items had been

overstocked when the system was installed. The answer
was “yes." I asked for a list of 25 of these items. Once it
had been supplied, I prepared a plot for each in which the
reorder level, the reorder quantity, and therefore the max¬
imum permissible stock level were identified. Then I
plotted the end-of-the-week inventories of each of the 25
parts since the initiation of the control system. I found that
almost half of the items whose stock was over the maxi¬
mum level when the program was begun had been reor¬
dered repeatedly when they reached their maximum stock
level. This meant that the average inventory levels of these
items were considerably higher than they had been before
COMPUTER-CONTROLLED MANAGERS 83

the system was installed. The system's operators had been

completely unaware of it.

I went on to discover that the system was doing a number

of other things it was not intended to do. It was out of con¬
trol and in ways that most managers would easily have
discovered had it been hand-operated.

No manager should use a system, computerized or not,

whose operations are not thoroughly understood. The
manager need not know how the computer works but
must know what it does. Those who designed the system
should be required to explain it to the manager in a lan¬
guage that can be understood, so its performance can be
evaluated.

Every manager who uses computer-based systems should

become computer literate. This is not difficult to do, nor
does it take a lot of time. A manager who is familiar with
the computer and the systems that are on it may well see
possibilities for new systems and improvements in the old
ones that the experts may have missed.

No manager has the right to allow a system to take control

rather than serve. Ignorance does not excuse failure to
fulfill a responsibility.

REFLECTIONS
 23
Management
Misinformation
Systems

A large number of the management information systems

in operation have failed to meet the expectations of the
managers they are supposed to serve. For this reason ref¬
erence to them as MlSs seems particularly appropriate.

There are many reasons for the disappointing perform¬

ance of MISs. For one, their design and operation are usu¬
ally based on the assumption that the most critical need of
managers is for more relevant information. This is not true.
They have a greater need for less irrelevant information, and
this is not a mere play on words.

Studies of the amount of reading material managers re¬

ceive have shown that they could not possibly take in all of

84
MANAGEMENT MISINFORMATION SYSTEMS 85

it even if they did nothing else at work. They suffer from

information overload. Other studies have shown that the
greater the overload, the less reading done. Therefore, the
more information provided to overloaded managers, the
less they use.

What most managers need is a filter to eliminate irrelevant

data. Few MISs do any filtering. Moreover, the relevant
information that most managers receive requires more
time to read than it should. It ought to be condensed.

The need for filtration and condensation is illustrated by

an experiment that my colleagues and I conducted many
years ago. We prepared a list of articles from recent issues
of the principal journals in one branch of applied science.
The list was sent to a large number of practitioners with
the request that they indicate which articles they had read
and which of them were above and below average. From
their responses we selected four papers that were unani¬
mously evaluated as above average and four similarly
evaluated as below.

These eight articles were given to two professional science

editors who were asked to reduce them by one-third only
by eliminating words, sentences, or paragraphs and to do
so with minimal loss of content. When they had com¬
pleted this task they were asked to reduce their already re¬
duced versions by half. Finally, they were asked to pre¬
pare short abstracts of each article. This provided us with
four versions: 100, 67, 33, and 5 percent.

While these condensations were being prepared we wrote

86 MANAGEMENT MISINFORMATION SYSTEMS

to the authors of the articles to tell them that our students

were being required to read their articles. Because, we
said, we wanted to be sure our students understood the
articles as intended, we asked them to prepare an objec¬
tively gradable test of the students' comprehension. All
the authors complied.

Using a carefully designed experiment, we had each of a

group of students read a 100 percent version of one article,
a 67 percent version of another, and 33 percent and 5 per¬
cent abstract versions of others. No two students read the
same four versions of the same four articles but each ver¬
sion of each article was read by the same number of stu¬
dents. Each student was then given four author-prepared
examinations on the articles read.

For the above-average articles there were no significant

differences in the scores obtained by those who had read
the 100, 67, and 33 percent versions. Those who had read
only the abstracts obtained a lower average score. This
showed that even good scientific writing could be reduced
by at least two-thirds without much loss of content.

In addition, there were no significant differences in the

scores obtained by those who had read the 100, 67, and 33
percent versions of the below-average articles; but those
who had read only the abstract obtained a higher average
score. This showed that the optimal length of a bad message is
zero.

Computerized systems that filter and condense informa¬

tion are available but seldom incorporated into MlSs.
MANAGEMENT MISINFORMATION SYSTEMS 87

Studies of the information managers receive show that

about two-thirds of it is unsolicited. Nevertheless, virtu¬
ally no MISs handle these externally imposed inputs.

Good secretaries serve as filters and condensers of solic¬

ited and unsolicited information. Little wonder, then, that
few managers are willing to swap their secretaries for
misses.

REFLECTIONS
 24
Ma thema nagemen t

The use of mathematical models to solve management

problems emerged out of World War II and became the
preoccupation of the field called operations research (OR)
or management science (MS). These models have been on
the decline in the last decade. It is important for managers
to know why.

About a decade ago an operations research group, highly

placed in an important government agency in a third-
world country, asked me to review one of its major pro¬
jects. Most of the members of this group had received ad¬
vanced degrees in OR from major universities in the
United States.

88
MATHEMANAGEMENT 89

The project involved the distribution of grains from

government-operated collection points to processing cen¬
ters and from these centers to food producers and whole¬
sale and retail establishments throughout the country. The
researchers were very proud of the number of variables
and constraints included in the linear programming model
they had developed, but they complained about their ina¬
bility to obtain the quality or quantity of data their model
required. To make up for this deficiency they had engaged
in what is called "data enrichment," a euphemism for
"data fabrication."

In evaluating their model the researchers had compared

the costs of the solutions it yielded with the costs of the
solutions obtained by managers without it. They had used
their model to calculate the costs of the solutions being
compared. Because their solutions minimized the sum of
the costs included in their model, they had to come out better
than those of the managers.

The minimization of the sum of costs included in a model

is not the same as minimizing the sum of costs in the real
world unless the model is a perfect representation of reality. It
never is. All models are simplifications of reality. If this
were not the case, they would be much more difficult—
perhaps too difficult—to use. Therefore, it is critical to de¬
termine how well models represent reality before using
them. In this project the researchers had failed to make
this determination.

I also learned that the managers to whom the research

90 MATHEMANAGEMENT

team's solutions were submitted invariably modified them

to take into account variables that were not included in the
model. However, the team had neither identified these
variables nor made any effort to incorporate them in their
model. When I asked why they told me that the variables
added by managers were qualitative and therefore could
not be used.

Finally, I learned that after a few months the managers

had stopped using the solutions provided by the model
because their political environment had changed. The
changes, the researchers told me, could not be included in
their model because they were neither measurable nor
predictable.

Clearly, if the researchers had solved a problem, it was not

the one the managers had.

Unfortunately, most managers are not equipped to evalu¬

ate the mathematical models that technicians apply to
their problems or the solutions these models yield. Too
many managers accept these models and solutions be¬
cause of their blind faith in "quantitative methods." Man¬
agers should never use "solutions" that are extracted from
models they do not understand. Nor should they stand in
awe of mathematics. Rather they should be aware of how
awful its products can be.
MATHEMANAGEMENT

REFLECTIONS
 25
Management
Consultants

No profession seems to have so many professional

consultants available to it as management. These
consultants, like all others, fall into three main categories,
defined by the same criteria used to divide labor in an or¬
ganization. These criteria are input, output, and markets.

Labor organized by input is functionally divided; for ex¬

ample, into purchasing, maintenance, manufacturing,
and marketing. Labor organized by output is divided by
products or services; for example, the Chevrolet, Pontiac,
Oldsmobile, Buick, and Cadillac divisions of General Mo¬
tors. Finally, labor organized by markets is divided by
geography—North, Central, and South America—or by

92
MANAGEMENT CONSULTANTS 93

type of customer—national accounts, wholesalers, retail¬

ers, and direct sales.

Management consultants are oriented correspondingly.

Input-oriented consultants are identified by the tools,
techniques, and methods they use to solve problems; for
example, applied mathematicians, statisticians, computer
programmers, accountants, and operations researchers.

Output-oriented consultants are identified by their prod¬

uct, the problems they solve; for example, designers of
compensation or incentive systems, management infor¬
mation systems, or automated office systems. These pro¬
fessionals usually have available a greater variety of tools,
techniques, and methods than input-oriented profession¬
als but they tend to be less sophisticated technically.

Finally, consultants who are market-oriented are

identified by the class of users they try to serve. They deal
with any type of problem their users may have, employing
whatever tools, techniques, and methods appear to be ap¬
propriate; for example, general practitioners of medicine.
In contrast, medical specialists are output-oriented and
medical technicians (e.g., radiologists) are input-oriented.
General-purpose management consultants, like general
practitioners of medicine, are market-oriented.

Output-oriented consultants frequently consult input-

oriented consultants and market-oriented consultants fre¬
quently consult both. Yet such consultation seldom goes
the other way.
94 MANAGEMENT CONSULTANTS

Professions can evolve in either direction, changing from

one type to another. Operations research (OR) is a case in
point.

During World War I, when OR came into existence, and

perhaps for a decade thereafter, it was a market-oriented
profession. Pure and applied scientists and engineers
worked on whatever problems were given to them by mili¬
tary commanders and, later, by corporate executives and
public administrators. Not only did they apply scientific
techniques and methods, but they also made liberal use of
raw intelligence and common sense.

Immediately after the war corporations in the United

States and Europe were confronted with accumulated de¬
mands for products that had become scarce. To meet these
demands production facilities had to be converted from
military to civilian production, expanded, and used
efficiently. This brought up questions for which OR devel¬
oped effective answers.

By the early 1960s Western industry had overexpanded its

production capacity. As a result competition for satisfying
the demands that existed and creation of new demand
preoccupied corporate management. OR was not well
equipped to deal with these marketing problems; its in¬
struments were better suited to the study of the
machinelike behavior of factories than the purposeful be¬
havior of consumers. Most OR practitioners stayed with
the problems they could handle well and further devel¬
oped the instruments for dealing with them. OR gradually
converted from a market-oriented to an output-oriented
MANAGEMENT CONSULTANTS 95

profession that, together with its familiar routines, was

pushed down to the lower levels of management.

This degradation continued. By the early 1970s the nature

of the problems confronting corporate management had
once again changed. When corporations reached the limits
to growth by manipulating marketing variables, further
growth could be obtained only by acquisitions, mergers,
joint ventures, internal development of new products and
services, or foreign expansion. OR was even less applica¬
ble to these problems than it was to those in marketing.
These growth- and development-related situations re¬
quired strategic planning rather than operational and tac¬
tical problem solving. As a result OR continued to move
down in corporations and, in many cases, out.

During its descent from the corporate executive level to

lower and lower levels of management, OR focused al¬
most exclusively on the development of its tools, tech¬
niques, and methods. It became an input-oriented profes¬
sion. Its journals reflect this; they are almost completely
devoid of case studies or discussion of the problems con¬
fronting corporate management. OR's dissociation from
the world of management is now almost complete.

Input-oriented consultants become obsolete when their

tools do. Output-oriented consultants become obsolete
when the work in which they specialize becomes less im¬
portant or disappears entirely. Therefore, if management
consulting is to be a vibrant and developing profession, it
must focus on its users and change with their needs.
96 MANAGEMENT CONSULTANTS

Those who serve management should focus on those they serve,

not on the services they render or the instruments used in render¬
ing them.

REFLECTIONS
 26
Types of Problems

There is no such thing as a marketing, production,

financial, personnel, or distribution problem. Such
modifiers in front of the word "problem" tell us absolutely
nothing about its nature, but they do tell us something.

A while ago some professors at my university met with

leaders of a self-development effort being made by a
nearby neighborhood. A member of the community broke
into the meeting with bad news. That morning an 83-year-
old woman who lived in the neighborhood and was active
in the development effort had gone to the area's only free
health clinic for her monthly checkup. She had been told
she was fine and left for home, a fourth-floor walkup.

97
98 TYPES OF PR0BLEA4S

While climbing the third flight of stairs she had a heart at¬
tack and died.

The silence that followed this announcement was finally

broken by the professor of community medicine who said,
'T told you we need more doctors at the clinic. If we had
them, we'd be able to make house calls and this sort of
thing wouldn't happen." After another silence the profes¬
sor of economics spoke up: "You know, there are plenty of
doctors in Philadelphia who still make house calls. She
just couldn't afford one. If welfare or medical benefits
were adequate, she could have called one and this
wouldn't have happened." The professor of architecture
then asked why elevators weren't required in all multiple¬
dwelling units of more than three floors.

What kind of problem was this—medical, economic, or ar¬

chitectural? Actually, none of these. It was just a problem.
The adjectives are indicative only of the point of view, the
mindset, of the person looking at the problem.

When trouble is found in one part of an organization, say

marketing, it is usually called a marketing problem. Then
the effort to solve it is confined to manipulating marketing
variables. This, however, is often not the most effective
way of handling it.

Consider an industrial example. The manager of a paper

mill found that its output was decreasing seriously be¬
cause of an increase in the number of different types of pa¬
per the plant had to produce. This reduced the length of
production runs and increased the amount of time re-
TYPES OF PROBLEMS 99

quired for setups. Therefore, the production manager saw

this as a production-scheduling problem.

Because the problem was caused by an increase in the

number of products in the product line, the research team
he called on for help asked why that number had not been
reduced, particularly when many of the products were not
profitable. The production manager replied that the con¬
tent of the product line was not his responsibility; it was
marketing's. Despite his opposition to its doing so, the
team made its proposal to the marketing manager. He re¬
jected it because, he said, most unprofitable products
were bought by consumers of profitable products. He
didn't want to risk losing those customers.

The research team then took another tack; it developed a

profit-based compensation system for salesmen to replace
the existing system which was based on dollar volume.
The new system provided no commission for unprofitable
sales but increased commissions for those that were
profitable. The company implemented this system. As a
result the salesmen sold fewer of the unprofitable prod¬
ucts and more of those that were profitable. Their incomes
and company profits increased and production improved
by more than four times as much as it could have with per¬
fect sales forecasting and production scheduling.

What kind of problem was this—production scheduling,

product-line design, or salesmen compensation? None of
these; it was just a problem.

Wherever problems appear they should be looked at from

100 TYPES OF PROBLEMS

as many different points of view as possible before a way

of attacking them is selected. The best place to solve a problem
is not necessarily where it appears.

REFLECTIONS
 27
Problem Treatments

There are four ways of treating problems: absolution, resolu¬

tion, solution, and dissolution.

To absolve a problem is to ignore it and hope it will go away

or solve itself.

To resolve a problem is to do something that yields an out¬

come that is good enough, that satisfies. Problem resolvers
take a clinical approach to problems; they rely heavily on
experience, trial and error, qualitative judgments, and
common sense. They try to identify the cause of a prob¬
lem, remove or suppress it, and thereby return to a previ¬
ous state.

101
102 PROBLEM TREATMENTS

To solve a problem is to do something that yields the best

possible outcome, that optimizes. Problem solvers take a re¬
search approach to problems. They rely heavily on experi¬
mentation and quantitative analysis.

To dissolve a problem is to eliminate it by redesigning the

system that has it. Problem dissolvers try to idealize, to ap¬
proximate an ideal system and thereby do better in the fu¬
ture than the best that can be done now.

The differences between these approaches is illustrated by

the following case. A large city in Europe uses double-
decker buses for public transportation. Each bus has a
driver and a conductor. The driver is seated in a compart¬
ment separated from the passengers. The closer the driver
keeps to schedule, the more he is paid. The conductor col¬
lects zoned fares from boarding passengers, issues re¬
ceipts, collects these receipts from disembarking passen¬
gers, and checks them to see that the correct fare has been
paid. He also signals the driver when the bus is ready to
move on after stopping to receive or discharge passengers.
Undercover inspectors ride the buses periodically to deter¬
mine whether conductors collect all the fares and check all
the receipts. The fewer misses they observe the more the
conductors are paid.

To avoid delays during rush hours, conductors usually let

passengers board without collecting their fares and try to
collect them between stops. Because of crowded condi¬
tions on the bus they cannot always return to the entrance
in time to signal the driver to move on. This causes delays
that are costly to the driver. As a result hostility has grown
PROBLEM TREATMENTS 103

between drivers and conductors which has resulted in a

number of violent episodes.

Management of the system first tried to ignore the prob¬

lem, hoping that if it were left alone it would absolve itself.
This effort at absolution did not work; the situation got
worse.

Management then tried to resolve the problem by

proposing a return to an earlier state by eliminating incen¬
tive payments and accepting less on-schedule perform¬
ance. The drivers and the conductors rejected this pro¬
posal because it would have reduced their earnings.

Next management tried to solve the problem by having

the drivers and conductors on each bus share equally the
sum of the incentive payments due each. This proposal
was also rejected by drivers and conductors; they were op¬
posed to cooperating in any way.

Finally, a problem dissolver was employed by manage¬

ment to deal with the situation. Instead of trying to com¬
promise the conflicting interests of the drivers and con¬
ductors, he decided to take a broader view of the system.
He found that during rush hours there were more buses in
operation than there were stops in the system. Therefore,
at his suggestion, conductors were moved off the buses at
peak hours and placed at the stops. This reduced the num¬
ber of conductors required at peak hours and made it pos¬
sible to improve the distribution of their working hours.
Under the new system conductors collected fares during
peak hours from people waiting for buses and were al-
104 PROBLEM TREATMENTS

ways at the rear entrance to signal drivers to move on. At

off-peak hours, when the number of buses in operation
was fewer than the number of stops, conductors returned
to the buses.

The problem was dissolved.

To problem dissolvers problems are opportunities, not

threats. By redesigning the systems with the problems, a
better performance than the best currently possible can be
obtained.

REFLECTIONS
 Creativity

Everyone would like to be creative, but what is creativity?

I believe it is the ability to identify self-imposed constraints, re¬
move them, and explore the consequences of their removal. This
is the same ability required to solve puzzles. Puzzles are
problems that are difficult to solve precisely because of
self-imposed constraints; for example, consider the nine-
dot puzzle that many of us tried to solve as children. The
nine dots form a square. The instructions are to place a
pen or pencil on any one of the dots and, without raising
the pen or pencil from the paper, to cover all the dots with
four straight lines.

If we try the most obvious "solution" to determine what

105
106 CREATIVITY

4^ V2

Figure 1.

the problem is (Figure 6), we leave one dot uncovered. If

we try two diagonals (Figure 7), we leave two dots uncov¬
ered. This is hardly progress. The most common
solution—there are many—consists of drawing lines that
go outside the perimeter of the square (Figure 8). Many solu¬
tions can be obtained by folding the paper. Neither going
outside the square nor folding the paper is precluded by
the instructions but each is often assumed to be disal¬
lowed. Such assumptions are self-imposed constraints.

Unfortunately, knowing what creativity is does not help

much in any effort to capture it. The principal difficulty
lies in identifying self-imposed constraints; we are gener-

4. Az
Figure 2.
CREATIVITY 107

them to consciousness or avoiding them even without

raising them to consciousness. Among them are lateral
thinking, brain-storming, synectics, TKJ, conceptual
block-busting, and idealized redesign, the last of which, I
believe, is the most effective.

An idealized redesign is one prepared on the assumption

that the system was destroyed last night but that its envi¬
ronment remains intact. It is this assumption that removes
most self-imposed constraints.

The product of idealized redesign is a design of that sys¬

tem with which the designers would now replace the sys¬
tem assumed to have been destroyed; that is, if they were
completely free to do so. The only constraints placed on the
design are (1) that it be technologically feasible, to preclude
science fiction, and (2) that it be operationally viable, capable
108 CREATIVITY

of surviving in the current environment if it were brought

into existence. However, the design need not be capable of be¬
ing brought into existence. Nevertheless, the designers are
almost always surprised at how closely their design can be
approximated. The reason is that the idealized design
process clearly reveals that many constraints thought to be
externally imposed are actually self-imposed; for example,
in 1973 the Fourth District Federal Reserve Bank was faced
with an exponentially increasing number of checks to
clear. It did not see how it could continue to deal with this
increase for more than a few years. There was not enough
space to house all the check clearers that would be re¬
quired. No way out of the difficulty was apparent.

The Bank's management decided to prepare an idealized

redesign of the money-exchange system. It began by ob¬
serving that a check is nothing but a way of transmitting
information between banks. Was this the best way? Once
this question was asked, the designers were able to re¬
move a self-imposed constraint: the assumption that the
transfer of funds between bank accounts requires the use
of checks. On reflection it became apparent that this trans¬
fer could be done electronically. As a result an electronic
funds transfer system, a version of which is now widely
used, was designed. This system has significantly reduced
the rate of growth of the number of checks that require
processing, thereby eliminating what was once an im¬
pending crisis.

To idealize is to think without constraints. To think without

constraints is to think creatively.
CREATIVITY 109

REFLECTIONS
Beating the System

If I could add only one subject to business-school curricula

it would be on how to beat the system. Beating the system
means making a well-designed system work poorly or a
poorly designed system work well.

Recall that British workers found an effective way to keep

the organizations that employed them from functioning
well: working to rule. What is not so well known is that if
managers went strictly “according to the book/' they too
would keep their organizations from functioning well.
Breaking some of the rules some of the time and getting
away with it—beating the system—is essential to the effec¬
tiveness, if not the survival, of even the best organiza¬
tions.

110
BEATING THE SYSTEM 111

Does this mean that something is inherently wrong with

all the rules imposed by organizations on their employees?
No, not all, but many. Those who make the rules often as¬
sume that those subjected to them are not capable of
exercising good judgment. Therefore, most rules are made
without allowing for exceptions and there are justifiable
exceptions to every rule; for example, the Federal Aviation
Authority has imposed the rule that seat backs in an air¬
plane must be in a forward position on takeoff and land¬
ing. This maximizes the distance between the backs of
seats and the faces of the persons sitting behind them, and
it permits passengers to get out of their seats and into the
aisle more easily in an emergency. But if there were no one
seated in the row behind a passenger, the rule would
make no sense. If a plane is going to hit something, it is
best to be as close as possible to a reclining position, with
feet forward and resting against something solid. Never¬
theless, when flight attendants work to rule they create a
potential risk to some passengers.

To cite a corporate example, an R&D organization works

mostly on government contracts but one of its depart¬
ments deals almost exclusively with corporate clients,
whom the organization charges on a cost-plus basis. Until
recently it applied overhead only to direct salaries of pro¬
fessional personnel; other expenses were charged without
overhead. The organization's executive then changed its
practice by applying overhead to all costs. This created a
serious problem for the corporately oriented department
which had more travel-related expenses than any other
department because most of its work is done at corporate
sites. If it had to apply overhead to these costs, it would
112 BEATING THE SYSTEM

lose most of its clients. To get around the new practice the
head of the department had each client open accounts
with the organization's travel agent and the hotels at
which his department's personnel stayed when visiting
their clients. The clients are now billed directly for most
travel expenses, thereby avoiding the overhead. If the
manager of this department had not beaten his company's
charging system, he would not have the clients he has and
he would not be able to serve them or his parent organiza¬
tion nearly as well as he does.

My course on beating the system would teach people how

to break obstructive rules. One of the experts I would
bring in as a guest lecturer is the neighbor of a friend. She
recently told me that for years a nearby department store
had cashed her checks. This had been a great convenience
because the store was much closer to her home than her
bank. But, she told me, this service had recently been dis¬
continued. Nevertheless, she had figured out a way to
beat its system. If she needed cash, she bought something
in the store for approximately the amount she needed and
paid for it by check, which was acceptable to the store.
Then she immediately took her purchase to the return
desk, where, in accordance with store policy, she was
given a cash refund.

Would that all managers had her ability. Fortunately,

many of them have at least some. If all managers took
their budgets as literally as flight attendants take FAA reg¬
ulations, most corporations would grind to a halt. Mana¬
gers worth their salt know how to beat the budget and get
BEATING THE SYSTEM 113

what they need. When this is done, organizations worth

their salt look the other way.

REFLECTIONS
 30
Defending Against
New Ideas

Not all new ideas presented to management are good. De¬

fenses against the bad are necessary, but they tend to be
applied to the good as well. This makes it difficult to inno¬
vate in many organizations.

Having spent many years in attacking these defenses, I am

a self-proclaimed expert on the subject. Let me identify
some of the most commonly used defenses.

114
DEFENDING AGAINST NEW IDEAS 115

1. HAS THIS IDEA EVER BEEN APPLIED

SUCCESSFULLY?

If the answer is "No, it is too new to have been applied,"

the defender will dismiss it because it has not been tested.
He will not try anything that has not been done success¬
fully by others. He is repelled by the prospect of being the
first.

If the answer is "Yes," the defender will go on to the sec¬

ond defense.

2. THE IDEA IS A GOOD ONE BUT IT DOESN'T

APPLY TO OUR KIND OF BUSINESS OR IN OUR
KIND OF ENVIRONMENT

The naive presenter usually responds to this defense by

citing a case of a business or an environment similar to the
defender's in which the idea has been applied success¬
fully. The defender then points out the inevitable differ¬
ences between any two businesses or environments.

This sequence could be continued indefinitely, but even¬

tually the presenter or the defender will give up without
changing his mind. To paraphrase Ambrose Bierce: there
is always an infinite number of reasons for not doing
something, but only one for doing it: it is right. But this is
hard to prove to someone who prefers to do nothing.
116 DEFENDING AGAINST NEW IDEAS

Even when the “offender" overwhelms the defender with

successful cases the battle is not won. The third line of de¬
fense is called into play.

3. HAVE ANY APPLICATIONS OF YOUR IDEA

FAILED?

If the answer is “Yes," the response is “Aha! I thought

so/' and the idea is dismissed. If the answer is “No," the
defender clearly doesn't believe it and dismisses the idea.

Defenses 1, 2, and 3 are used by those who accept the nov¬

elty of the idea presented. If its novelty is not accepted,
different defenses are used.

4. THIS IS NOTHING BUT. . .

In the early 1950s, when operations research was new, I

collected 14 assertions that it was “nothing but . ..." It
was said to be identical to such diverse disciplines as in¬
dustrial engineering, economics, statistics, applied mathe¬
matics, and cybernetics. Therefore, I formulated the fol¬
lowing definition; “Operations research is a field that is
identical to each of 14 fields, each of which is different
from all the others." This definition had no effect on the
“this-is-nothing-but-ers." Nothing does.

The “old-stuffers" are equally intransigent.

DEFENDING AGAINST NEW IDEAS 117

5. WE TRIED IT A EONG TIME AGO AND IT DIDN'T

WORK THEN. WHY SHOUED IT NOW?

I once presented a newly developed mathematical tech¬

nique that could be used to solve a serious problem that
had long faced the managers I addressed. One of them
told me he had tried the technique 20 years earlier and it
hadn't worked then. I naively pointed out how recently it
had been invented and that he could not have used the
same technique that long ago. He told me that the in¬
ventor had obviously rediscovered the wheel. I gave up.

If an "old-stuffer" can't refer to his own imagined experi¬

ence with an idea, he refers to that of others.

6. (SUCH AND SUCH A) COMPANY TRIED THIS IDEA

AND IT DIDN'T WORK THERE. WHY SHOULD IT
HERE?

More than a decade ago Volvo built its radically new as¬
sembly plant in Kalmar, Sweden. Representatives of the
American automotive industry and business journalists
went to examine it. Almost without exception they re¬
ported that it was not working or would not work in the
United States. A decade later, when the American auto¬
motive industry was in dire straights, it reevaluated
Kalmar and has since imitated it increasingly.

Alas! There is nothing new under the sun or in the way

managers defend themselves against new ideas.
118 DEFENDING AGAINST NEW IDEAS

REFLECTIONS
 31
The Obvious

When the word “obvious" is applied to an assertion it

does not mean that the truth of the statement is so appar¬
ent that it requires no supporting evidence or argument. I
learned this as a college freshman from an eminent profes¬
sor of mathematics who, because of his eminence, had ex¬
clusive use of a classroom connected to his office.

One day he was going through a proof in geometry, writ¬

ing the steps in one column and the supporting reasons in
another. After writing the third step on the board he drew
a squiggly line in the "reasons" column and told us that
this step was obvious. Halfway through the next step he
stopped, stepped back, and looked at the blackboard quiz-

119
120 THE OBVIOUS

zically. He then turned to the class and said, "Excuse me.

ril be right back." He disappeared into his office and was
gone for what seemed to be a very long'time. When he re¬
turned he had a self-satisfied grin on his face. "I was
right," he said, "that step was obvious."

Obviousness is a property not of statements that require

no proof but of statements made by those who are unwill¬
ing to have them questioned.

Statements said to be obvious are often untrue. Therefore,

their acceptance without questioning often obstructs effec¬
tive problem solving. The following case is an illustration:

Back in the days when stewardesses had to be unmarried I

was part of a team that worked for a major airline on im¬
proving the scheduling of its school in which these young
women were trained. While engaged in this effort we
found that on the average stewardesses flew fewer hours
per month than the maximum allowed. We looked into
this and eventually developed a way of scheduling stew¬
ardesses that would increase their average flying time con¬
siderably.

We presented our findings to the airline's senior mana¬

gers, all of whom but the vice-president of personnel
showed great enthusiasm for our proposal. This vice-
president argued that an increase in flying time would in¬
crease the stewardesses' attrition rate and would cost the
airline more in recruiting and training than it would save
in operating costs. This had not occurred to us but we re¬
acted in typical academic fashion. We asked, "How do
THE OBVIOUS 121

you know?" He replied that it was obvious to anyone who

had worked with stewardesses, and obviously we hadn't.
When we pressed for more convincing proof he refused to
discuss it further. Fortunately, the remaining managers
were willing to discuss it further. That discussion eventu¬
ally led to their authorizing our looking into the matter.

Because monthly flight assignments were selected by

stewardesses in the order of their seniority, we compared
the average hours of the most senior with those of the
most junior. Contrary to what the vice-president of per¬
sonnel had told us, we found that on the average, senior
stewardesses flew more hours than their juniors. Further
investigation revealed the reason: the senior stewardesses
preferred schedules with regular days off and most free
evenings at their home bases. We learned that schedules
with these characteristics made it easier for them to organ¬
ize their social lives and were more likely to be found in
assignments with greater flying time.

We then modified our initially proposed scheduling proce¬

dure to provide all stewardesses with regularity of days off
during a month and at least 20 percent more evenings at
home. Despite these modifications, we were able to retain
most of the increased flying time made possible by our
earlier scheduling procedure. These results were pre¬
sented to the stewardesses' union which supported our
proposal enthusiastically and joined in submitting it to
management. It was accepted and implemented. The ex¬
pected reduction in the airline's costs was obtained.

This improvement in performance was almost lost because

122 THE OBVIOUS

one executive, presumed to be an authority, had said that

more flying time would not be acceptable to the steward¬
esses and that this was obvious.

The great American wit Ambrose Bierce perceived all this

when he defined "self-evident"—a synonym of
"obvious"—as "evident to oneself and no one else."

REFLECTIONS
 32
Objectivity

Objectivity is a scientific ideal particularly sought by man¬

agement scientists. Although its meaning is not clear, ob¬
jectivity is generally believed to be what Winnie the Pooh
called a “GOOD THING." It is also believed to require the
exclusion of ethical and moral judgments from inquiry and
decision making. Objectivity so conceived is not possible.

Most, if not all, scientific inquiry involves testing

hypotheses or estimating the values of variables. These
procedures necessarily entail balancing two types of error.
In testing hypotheses these errors are rejecting
hypotheses when they are true and accepting them when
they are false. Naturally we would like to minimize the
probabilities of making them but unfortunately

123
124 OBJECTIVITY

minimizing one maximizes the other. Therefore, setting

these probabilities requires a judgment of the relative seri¬
ousness, hence value, of the two types of error. Research¬
ers seldom make this judgment consciously; they usually
set the probabilities at levels dictated by scientific conven¬
tion. This attests not to their objectivity but to their igno¬
rance.

The choice of a way of estimating the value of a variable

requires the evaluation of the relative importance, hence
values, of underestimates and overestimates of the vari¬
able. Each estimating procedure contains a (usually im¬
plicit) judgment of the seriousness of the two possible
types of error. Therefore, estimates cannot be made with¬
out a value judgment, however concealed it may be.

The most commonly used estimating procedures are said

to be “unbiased." The estimates they yield, however, are
best only when errors of equal magnitude but of opposite
sign are equally serious. This is a condition that I have vir¬
tually never found in the real world.

In testing hypotheses and estimating the values of vari¬

ables, science unconsciously equates objectivity with un¬
consciousness of the value judgments.

The prevailing concept of objectivity is based on a distinc¬

tion between ethical-moral man—who is believed to be
emotional, involved, and biased—and scientific man—
who is believed to be unemotional, uninvolved, and unbi¬
ased. Objective decision makers are expected to take their
heads—not their hearts—into the workplace. To assume
OBJECTIVITY 125

that the heart and head can be separated is like assuming

that the head and tail of a coin can be separated because
they can be discussed or looked at separately.

Objectivity does not consist of making only value-free

judgments in conducting inquiries and making decisions.
It consists of making only value-/w// judgments; the more
extensive the values, the more objective the results. A de¬
termination is objective only if it holds for any values that
those who can use it may have. For this reason objectivity
is an ideal that can never be attained but can be continu¬
ously approached.

Objectivity cannot be approximated by an individual in¬

vestigator or decision maker; it can be approached only by
groups of individuals with diverse values. It is a property
that cannot be approximated by individual scientists but
can be by science taken as a system.

All this has an important implication for management.

The values of all those affected by a decision, its stakehold¬
ers, should be taken into account in making that decision
but this cannot be done without involving them in the
decision-making process. To deprive them of opportuni¬
ties to participate in making decisions that affect them is to
devalue them, and this, it seems to me, is immoral. Mana¬
gers have a moral obligation to all who can be affected by
their decisions, not merely to those who pay for their ser¬
vices.
126 OBJECTIVITY

REFLECTIONS
 33
Infallibility

Technology is all but worshiped in a large part of the

world. As a result, scientists and engineers, like the
clergy, are often believed to be infallible. When confronted
by those who believe or tend to believe this, I usually tell
them three stories.

The first is about research conducted many years ago by

Dr. Tibor Fabian at the University of California, Los An¬
geles. Fabian had developed one of the first computerized
management games. It required those who played it to
manage a production process and to try to minimize the
sum of production and inventory costs. A model in the
computer generated the situations that required decisions

127
128 INFALLIBILITY

as well as their consequences. It was a simple model, one

with which most students of management are familiar.
However, the players were not aware of this.

At a national conference of economists held in Los An¬

geles, Fabian offered those attending an opportunity to
play the game, then a great novelty. In return, he asked
the players to collect data and make an effort to derive the
model that generated the decision situations and the re¬
sults of their decisions. Not one player obtained the correct
model.

The second experiment was conducted by Alex Bavelas at

MIT, also many years ago. His subjects were taken into a
room furnished with a screen on which slides were pro¬
jected. The slides had been produced by waving a
flashlight in a dark room over unexposed film. The sub¬
jects sat at desks on which there were two buttons. They
were told to press one of the buttons after each slide. If
they pressed the right one, they would be paid; if they
pressed the wrong one, they would get nothing.

The subjects naturally asked what principle was involved

in determining the right button. Bavelas told them this
was what they had to find out by trial and error.

After a few slides most subjects began to formulate theo¬

ries to explain the rewards they had received, and soon
they were quite sure of the correct explanation. They
played accordingly. If they were not rewarded as ex-
INFALLIBILITY 129

pected, they blamed it on their observations, not on their

theories.

When the experiment was completed the subjects were

asked to reveal their theories. All of them did so willingly.
Bavelas then told them that he had actually rewarded
them purely at random. There had been no relationship
between the buttons they pressed and the rewards. Most
of the subjects were surprised but insisted that their theo¬
ries were at least approximately correct. They would not
abandon their theories.

The third piece of research was conducted by Professors

C. West Churchman and Philburn Ratoosh at the Univer¬
sity of California, Berkeley. They too developed a manage¬
ment game, but theirs required play by a team of four.
One acted as the CEO, the others as managers of manufac¬
turing, marketing, and finance. The teams were asked to
maximize the performance of the simulated firm. As in the
Fabian experiment, this simulation was generated by a
well-known mathematical model from which an optimal
solution could easily be derived. Furthermore, all the
graduate students who were used as subjects had at¬
tended a class on quantitative methods in which this
model and its solution had been presented.

In each team the student who served as the financial man¬

ager was informed beforehand of the nature of the model
and its solution. These managers, however, were told not
to reveal this information to their teammates until they re-
130 INFALLIBILITY

ceived a signal from the experimenter. They were then to

pretend to have made the discovery on their own.

Only a small percentage of the teams adopted the optimal

solution when it was proposed to them.

Churchman and Ratoosh described this experiment and

its results at a large number of professional meetings, at
each of which explanations of the failure of the teams to
implement the optimal solution were proposed and cor¬
rective actions suggested. Churchman and Ratoosh re¬
corded these suggestions and subsequently tested them in
the same experimental situation they had already used but
with new subjects. The probability of implementation was not
significantly increased.

The morals suggested by these stories are the following, in

turn:

1. When there is regularity in nature, scientists may not find

it.

2. When there is no regularity in nature, they may insist that

there is and that they know what it is.

3. When there is regularity in nature and it is revealed to sci¬

entists, they may be disinclined to use it.

Those who do not know but think they do are more dan¬
gerous advisers than those who do not know—but know
it.
INFALLIBILITY

REFLECTIONS
 34
Who's Irrational?

We often say of people whose behavior we do not expect

and cannot explain that they are irrational. In this way we
absolve ourselves of any responsibility for their behavior.

I have never seen a problem believed to be caused by the

behavior of others that could be solved by assuming that
they were irrational. On the other hand, by assuming that
we are irrational, solutions to these problems can often be
found; for example, during a working visit to India in 1957
I met a number of family planners from the United States
who had made no progress in their extended efforts to re¬
duce India's birthrate. Most of them attributed their failure
to the irrationality of the Indians.

132
WHO'S IRRATIONAL? 133

After hearing this a number of times I suggested to one of

the family planners that it could be they who were irra¬
tional. 1 pointed out that a Brazilian woman had recently
given birth to her forty-second child. Assume, 1 said, that
the average woman can produce only about half this num¬
ber, say 20. The difference between 20 and 4.6, the aver¬
age number of children per Indian family, is much greater
than the difference between 4.6 and 0. This, I suggested,
was an indication that the size of Indian families was not
due to lack of birth control.

The family planner I had addressed thought this argument

was ridiculous. He left, terminating our conversation. For¬
tunately, a distinguished Indian demographer, T. K.
Balakrishnan, approached me with apologies for having
overheard our conversation. He suggested that we collab¬
orate in research on the possible rationality of Indian re¬
productive behavior.

We began at the Indian Statistical Institute, but I had to

return to the States before our research was finished.
Luckily, Glen Camp, one of my colleagues, replaced me in
India and helped Balakrishnan finish the work. Briefly,
this is what they found.

The average Indian male could expect a number of years of

unemployment when he got older. India had no social se¬
curity program and the typical worker did not earn
enough to save for these unemployed years. His only
hope, then, was to be provided for by his children. It took
an average of 1.1 wage earners to support one unem-
134 WHO'S IRRATIONAL?

ployed adult at the minimal subsistence level, but, be¬

cause it takes two to produce a child, each family needed
at least 2.2 wage-earning children. Because half the chil¬
dren born were female, and females were essentially un¬
employable in India at that time, 4.4 children were re¬
quired. To cover infant and child mortality, this number
had to be adjusted upward to 4.6 children.

This result could have been obtained by pure chance, but

its validity was easy to determine. If family size could be
explained even in part by the desire for insurance against
old-age unemployment, then families whose first three
offspring were male should have few if any additional
children. Those families whose first three offspring were
female should just be getting started. These inferences
were found to be correct.

Those family planners who had attributed irrationality to

the Indians had unknowingly expected them to commit
delayed suicide by limiting the size of their families.

Consider another example. At one time producers of gaso¬

line advertised heavily in an effort to convince consumers
of the superiority of their respective products. There were
no significant performance differences between the differ¬
ent brands, but their producers assumed that they could
persuade consumers to the contrary and thereby induce
them to behave irrationally.

Subsequent research showed that the hundreds of mil¬

lions of dollars spent on gasoline advertising were wasted;
it had virtually no effect on consumers. Behaving ration-
WHO'S IRRATIONAL? 135

ally, they had no brand preferences. They bought gasoline

at those service stations at which they believed the time
required to get service was minimized. The oil company
that sponsored this research was able to use these findings
to increase its market share by locating, designing, and
operating its stations to minimize service time. It also
made reduced service time the theme of its advertising.

Corporate managers who think that consumers, employ¬

ees, suppliers, competitors, or government officials are
behaving irrationally should think twice and they should
think differently the second time.

REFLECTIONS
 35
Advertising: A
Wonder or a Waste?

I've done a great deal of research on advertising that has

convinced me that many agencies do a better job of selling
themselves to their clients than of selling their clients'
products or services to potential customers.

I find it difficult to understand why companies are willing

to spend billions of dollars each year on advertising with¬
out any measure of its effectiveness. Advertising expendi¬
tures are based more on faith than fact. Agencies argue
that the impact of advertising on sales cannot be measured
because sales are affected by a large number of interacting
variables and their effects can't be separated. This is non¬
sense. Modern experimental designs have been applied to

136
ADVERTISING: A WONDER OR A WASTE? 137

“multivariate" situations for a long time and they have

been used effectively to measure the results of advertising
by several companies, including Anheuser-Busch.

Anheuser-Busch has learned by experiment that a number

of assumptions commonly made about advertising are
false.

First, increases in advertising do not always produce an

increase in sales and, believe it or not, sometimes result in
a decrease. The response to advertising is like the re¬
sponse to sales calls and most other types of stimulus. It
takes a certain amount of advertising or sales calls to pro¬
duce sales (Figure 9). This amount can be thought of as a
threshold, beyond which sales increase with more advertis¬
ing or calls up to a saturation point at which the customers
cannot, or will not, buy more. Then, even with increased
advertising or calls, sales remain relatively flat up to a su¬
persaturation point. Beyond that customers are so annoyed

Figure 9.
 138 ADVERTISING: A WONDER OR A WASTE?

by advertising or salesmen that they stop buying the prod¬

ucts or services offered.

Companies that advertise beyond the supersaturation point, and

there are many, can increase sales by reducing their advertis¬
ing. This ought to be expected. Imagine what would hap¬
pen to a product if all TV advertising were devoted to it or
if a salesman decided to take up residence in a customer's
establishment.

Although different populations respond differently to ad¬

vertising, the shape of their response curves is the same.
They have different heights and different locations on the
horizontal axis (Figure 10); for example, heavy users of a
product are generally the most responsive to advertising;
moderate users, less so; and occasional users, least so.
Then, if any population can be segmented into use-
categories, a composite response curve can be obtained
Sales

Figure 10.
ADVERTISING: A WONDER OR A WASTE? 139

Figure 11.

(Figure 11). This composite curve can be used to maximize

return from the investment in advertising.

It is also generally assumed that to be effective advertising

must be continuous. It is clear, however, that if it is
stopped for a day, no one notices. How long does adver¬
tising have to be discontinued before it is noticed and sales
begin to fall? Most companies cannot answer this ques¬
tion. Anheuser-Busch can; it found that the length of time
that one can be off a medium with no effect on sales is sev¬
eral times longer than most advertisers believe or that
most agencies will allow them to believe. This finding
made it possible for Anheuser-Busch to pulse advertising,
to get the same effect on sales obtained by relatively con¬
tinuous advertising but with a much smaller expenditure.

Furthermore, if a company uses several media, it can

140 ADVERTISING: A WONDER OR A WASTE?

phase its pulses on each to have advertising running in

one medium at all times.

How generalizable are the findings of Anheuser-Busch?

We don't know because only a few companies conduct
such research, and many that do don't publish it for obvi¬
ous reasons. The results obtained from advertising re¬
search by Anheuser-Busch should not be generalized, but
the value obtained from it should be.”^

REFLECTIONS

*For a more detailed description of the research done at Anheuser-Busch see

Russell L. Ackoff, The Art of Problem Solving, Wiley, New York, 1978, Chapters 10
and 11.
 36
Managing Interactions

A company is obviously a system. What is not so obvious

is what this implies.

Many managers believe that if every part of a company,

considered separately, is made to perform as well as possi¬
ble, the company as a whole will perform as well as possi¬
ble. Nothing could be further from the truth; for example,
consider an automobile, a mechanical system. Suppose we
ask a group of top automotive engineers to determine
which make and model of car has the best carburetor and
they choose one from a Buick. Then we ask them to do the
same for the transmission and they select one from a
Mercedes. Suppose we continue until we identify the best
of each part needed to make an automobile. Then we ask

141
142 MANAGING INTERACTIONS

the engineers to create an automobile by assembling those

parts identified as the best. They would not succeed be¬
cause the parts would not fit together. Even if they did,
they wouldn't work well together.

The performance of any system, therefore of a company,

is never equal to the sum of the performances of its parts
taken separately. It is the product of their interactions.

For this reason effective managers do not focus on actions

but on interactions. They coordinate and integrate the in¬
teractions of their subordinate units and individuals and
those of their units as a whole with other parts of the or¬
ganization. Effective managers do not manage the actions
of the units and individuals reporting to them; they man¬
age their interactions and allow units and individuals to
manage their own actions. Employees at any level who
cannot do their jobs without supervision are not fit for
their jobs.

The management of interactions involves the coordination

and integration of plans, policies, programs, projects, prac¬
tices, and courses of action. This coordination and integra¬
tion can be facilitated by providing each manager with a
group—call it a board, committee, or whatever—that
consists of (a) the manager, (b) the immediate superior,
and (c) the immediate subordinates (see Figure 12). All
managers except the one at the top and those at the bot¬
tom would serve on their own boards, their bosses', and
their immediate subordinates'. In an organization with
five or more levels, middle managers would have face-to-
face interactions with five levels of management: their
MANAGING INTERACTIONS 143

Figure 12.

own and two higher levels on their bosses' boards and two
lower levels on their subordinates' boards.

Each board should have the responsibility for coordinating

the actions of the level below it. Because subordinates are
in the majority on the board coordinating them, they
coordinate themselves with the help of two higher level
managers. Except at the top and bottom of the organiza¬
tion, all managers participate in the coordination of (a)
their subordinate units (on their own boards), (b) the units
subordinate to their subordinate units (on their subordi-
144 MANAGING INTERACTIONS

nates' boards), and (c) their units with others at the same
level (on their superiors' boards).

Because all boards except the one at the top and those at
the bottom consist of members who participate in two
higher and two lower level boards, these boards can also
integrate the plans, policies, programs, projects, practices,
and actions of as many as five different levels of the organ¬
ization.

It should be emphasized that boards manage the interac¬

tions of units, not their actions. They affect the actions of
units or individuals only when the actions affect other
units or individuals.

A number of organizations favor this system; for example,

Anheuser-Busch, Clark Equipment Company, Tremec (in
Mexico), the U.S. Naval Education and Training Com¬
mand, and lECSA (in Argentina).

Managers in these organizations are frequently asked how

they find the time to get their work done when they sit on
as many as 10 boards. These boards meet no more than
twice a month and seldom for longer than two hours.
Therefore, even if a manager is active on 10 boards, no
more than 40 hours a month are spent on them, or 25 per¬
cent of the time, if only a 40-hour week is worked. This
leaves sufficient time for other activities, most of which are
considerably less important than coordinating and
integrating the units reporting to them and these units
with others in the organization. Therefore, a more appro-
MANAGING INTERACTIONS 145

priate question would be: What do they do with the rest of

their time?

REFLECTIONS
 37
The Aesthetics of Work

Ancient Greek philosophers identified three ideals—truth,

the good, and beauty—the pursuit of which they believed
was necessary for progress and development. Modern
man has added a fourth—plenty or abundance.

Science is dedicated to the pursuit of truth, and technology,

to its application; ethics and morality, to the pursuit of the
good; aesthetics, to beauty; and economics, to plenty.

These pursuits are relevant to management. Management

science, business ethics and morality, and managerial eco¬
nomics are subjects that are familiar to managers. But
what about the aesthetics of work or management? What in

146
THE AESTHETICS OF WORK 147

the world does that mean? It is not surprising that the an¬
swer to this question is not apparent because the aestheti-
cian has been the "odd man out" for a very long time.

Although most people believe that we have made a great

deal of scientific and economic progress and some believe
that at least some ethical and moral progress has been
made, few believe that there has been any aesthetic prog¬
ress. We seem neither to produce more beauty nor to ap¬
preciate beauty more than preceding generations.

Aesthetics is the least understood aspect of progress and

development. Little wonder, then, that most managers
have no idea of its relevance to their work and that of oth¬
ers.

Aesthetics is related to two things: recreation and creation.

Recreation is activity that refreshes one's mind and body,
activity from which immediate satisfaction is derived, re¬
gardless of its outcome or consequences. It is intrinsically
valuable; this means that its value lies in the fun and enjoy¬
ment we get out of it. To the extent that managing is fun
and enjoyable, it has aesthetic value.

The creative aspect of aesthetics is reflected in the sense

one can have of getting somewhere, of developing. It is
this sense of progress that endows human activity with ex¬
trinsic value and makes it meaningful. Beauty inspires, pro¬
duces visions of possible progress, and encourages the
pursuit of these visions, whatever short-run sacrifices are
required. Therefore, it motivates us to pursue develop¬
ment, to pursue progress. Recreation provides refreshing
148 THE AESTHETICS OT WORK

pauses in the pursuit of progress and makes the pursuit

itself a satisfaction.

The currently growing concern with quality of life, in gen¬

eral and work life in particular is a matter of aesthetics. To
improve the quality of life or work life is to increase the
(recreational) satisfaction derived from what we do, what¬
ever we do it for, and the (creative) satisfaction derived
from making progress toward our ideals.

A few years ago the CEO of a very successful corporation

asked me to look around his organization for any serious
problems that were being overlooked. I spent several
months traveling on reconnaissance. When I reported
back to the CEO I told him there was one overriding prob¬
lem that required attention: many of the company's em¬
ployees, especially its managers, were not enjoying their
work, thought it was unimportant, and had little sense of
personal progress. As a result, their efficiency and effec¬
tiveness were deteriorating. The company was aesthetic¬
ally deficient.

After some discussion of what might be done about it the

CEO authorized an effort to improve the aesthetics of
work. A participatively designed quality-of-work-life pro¬
gram for all employees was initiated and eventually suc¬
ceeded.

Work that is neither fun nor meaningful is not worth doing well,
no matter how much one is paid to do it.
THE AESTHETICS OF WORK 149

REFLECTIONS
 38
Friendship

Two of my academic friends once conducted an experi¬

ment built around a management game.”^ The game was
played by teams of four graduate students selected at ran¬
dom, each assigned to a different role. One was made the
chief executive officer and the others were assigned the re¬
sponsibility for manufacturing, marketing, and finance.
The game consisted of managing a simulated company in
an effort to maximize a specified measure of performance.
Before each team began its play the experimenters told its
financial manager which decision rule would yield the
best performance. However, the instructions were not to

*This is the same research described in Chapter 33 on "Fallibility."

150
FRIENDSHIP 151

reveal this rule to the other team members until they had
played a while and a signal had been received from the ex¬
perimenters. The financial manager was to pretend to
have discovered the rule alone.

Very few of the teams that played this game accepted the
optimal decision rule when the financial manager pre¬
tended to have discovered it. My friends, the experimen¬
ters, were surprised at this result. I suggested that they
might increase acceptance of the solution by selecting stu¬
dents who were close friends to act as chief executive
officer and financial manager.

The experiment was rerun according to my suggestion.

The frequency of acceptance of the proposed solution in¬
creased significantly.

Friends are persons we know, like, and trust; we believe

that they will act in ways that they perceive are in our best
interests. Therefore, we are likely to take their advice more
seriously than that of others.

When managers must make decisions quickly they are un¬

likely to accept the advice of anyone but a friend. They
might accept the same advice from others if they had time
to verify it for themselves. Without such verification they
usually prefer to act on their own intuition than on
nonfriendly or unfriendly advice.

It is for this reason that the most effective management

152 FRIENDSHIP

teams are those in which each member considers every

other member to be a friend, regardless of rank.

The effectiveness of external consultants also depends crit¬

ically on the personal relationships they establish with the
managers to whom they report. In hundreds of research
engagements at the university-based research center of
which I am part, I have observed that the most successful
projects are those in which the project leader and the re¬
sponsible manager have become close friends. They inter¬
act socially as well as in the business context.

Friendship is most rapidly established by demonstrating

an awareness and consideration of the personal welfare of
others. Some people obviously have a greater capacity for
friendship, but no one is completely lacking in it. It can be
developed if a conscious effort is made to do so.

In most companies it is easier to form friendships among

peers than across ranks. A great deal can be done, how¬
ever, to facilitate the formation of "vertical" friendships.
The fewer prerequisites and privileges used to differenti¬
ate ranks, the more task-oriented groups with members of
different rank, and the more opportunity for informal
nonbusiness interaction across ranks, the smaller the gap
friendship has to bridge.

Any manager who cannot count an immediate superior,

peers, immediate subordinates, and consultants as friends
operates under a handicap that no amount of competence
is likely to overcome.
FRIENDSHIP

REFLECTIONS
 39
Respect

In my opinion a principal difference between excellent ex¬

ecutives and others is the way that they use and are used
by their subordinates. They respect their subordinates and are
respected by them. They give their subordinates their heads
but protect them from abuse or misuse by any higher au¬
thority or external source.

I once heard a CEO say something like this to a subordi¬

nate for whom he had a very high regard: “You consult
me less than anyone else around here. That's OK as long
as you continue to perform well. If you don't. I'll have
your ass. Until then. I'll see to it you're free to do what you
want and I'll give you any help you ask for and I can give.
I could probably help you more if you kept me better in-

154
RESPECT 155

formed. If you do, I promise not to intervene any more

than I do now.” The subordinate smiled, promised to
keep him better informed, and thanked him. Mutual re¬
spect filled the room.

Another CEO once told me: ”1 can't do anything without

good people. No organizational structure or culture, no
leadership, no management style, and no amount of re¬
sources can make up for incompetent and uncommitted
subordinates. The most important job of a CEO is to en¬
sure the presence of competent and committed subordi¬
nates and to create an environment in which they can
show their stuff.”

Another executive, when approached by a subordinate for

help in making a difficult decision, said, "Tm not going to
make that decision for you. I won't even tell you what de¬
cision I would make if I were in your shoes. If I didn't
think you could make the decision better than I can, you
wouldn't be where you are. Now, if you want to discuss
your problem because you think discussing it will help you
make a better decision, fine. But if you want me to make it
for you, forget it.”

A first-class executive frequently seeks the advice of a sub¬

ordinate. It is not always accepted, but when it isn't the
rejection is always explained. When possible the subordi¬
nate is given a chance to rebut the rejection before action is
taken.

Respect between superior and subordinate should be sym¬

metrical. A highly valued subordinate was once confron-
156 RESPECT

ted by his boss with a rumor that the subordinate was

negotiating a move. The subordinate said, 'TTs not true. If
I ever consider the possibility of moving, you'll be the first
to know. On the other hand, if you ever think I ought to
leave. I'd like to be the first to know." His boss agreed.

An internal candidate for a senior position in a company

was asked by his prospective boss: "How do you think
you can help me do my job better?" The candidate an¬
swered, "I can't answer that, but if you give me the job
you'll be able to answer it after a while and I'll be able to
tell you what you can do to help me to do my job better. It
is my responsibility to find out how to use my subordi¬
nates, not how to be used by my boss. Few are the best
judges of how they can serve others, but many are the best
judges of how they can be served." He got the job and was
first-rate.

A CEO once asked me to sit in on an interview of an exter¬

nal candidate for a vacant vice-presidency. After the cand¬
idate had entered and had been put at ease the CEO said,
"It's a lot more important that this company be the right
place for you than that you be the right person for this
company. Therefore, I'd like you to interview me." The
surprised candidate caught his breath and then did just
that, with poise and intelligence. The CEO was incredibly
frank in his replies to the candidate's questions,
unashamed to reveal what he considered to be the weak¬
nesses of the company and the detractions of the job.
After the candidate had left, the CEO told me that both he
and the candidate could learn a great deal more about the
RESPECT 157

candidate's value to the company in such an interview

than in the kind normally conducted. I agreed completely.

A good executive is a leader, not a commander, except in

emergency situations. Above all else, an effective leader
must have the respect of his or her subordinates and must
respect them. Loyalty is not a substitute for respect, but a
leader who is thought of as a friend by subordinates and in
turn, thinks of them as friends, receives both.

REFLECTIONS
 40
Consensus

When consensus—that is, complete agreement—is not

reached by groups of managers faced with the need to
make a decision, they often default or resort to majority
rule.

The problem with less-than-unanimous majority decisions

is that there is always a minority. Those in the minority are
seldom as committed to implementation of the decision as
are those in the majority. For this reason complete agree¬
ment is clearly desirable. But how can a consensus be
reached when there is a difference of opinion? The answer
lies in the fact that differences of opinion are more likely to
rest on beliefs that involve matters of fact than on attitudes

158
CONSENSUS 159

that involve matters of value; for example, supporters and

opponents of capital punishment differ more on the effect
they believe capital punishment has on the number of cap¬
ital crimes committed than on the value they place on hu¬
man life. Similarly, the current debate on abortion rests
largely on a difference of belief as to when life begins but,
again, not on the value of human life.

It is generally easier to dissolve differences of belief than

differences of attitude. This can often be done by
investigating the relevant facts in a way that is thought to
be fair by all those concerned. Then, when the results are
in, consensus can often be reached, even if it requires the
majority to change its opinion; for example, in a large
metal-producing company, executives could not decide
whether to place maintenance under engineering or man¬
ufacturing. Their disagreement was based on differing be¬
liefs about the impact of the alternatives on the effec¬
tiveness of maintenance. Because the company operated a
number of similar plants, the executives reached a consen¬
sus on a test of the alternatives in different plants. When
the results were in they reached complete agreement on
the location of the function.

In a company in the food and beverage industry, agree¬

ment could not be reached on the appropriate amount to
spend on advertising. However, the managers did agree
on the design of an experiment in which different levels of
advertising were used in a number of marketing areas.
After several months this experiment enabled manage¬
ment to determine the effects of advertising on sales and
to agree on the amount to be spent on it.
160 CONSENSUS

In some cases when consensus cannot be reached within

the time available, no decision is made; current behavior is
continued. This type of response can paralyze an organi¬
zation. There is an alternative.

I once attended a corporate meeting at which nine differ¬

ent solutions to an organizational problem were pre¬
sented, each prepared by a different team of managers. All
the managers were present, but no agreement could be
reached on which proposal was best and no proposal at¬
tracted majority support. They appeared to have reached
an impasse. The chairman asked if I could suggest a way
out of it. I proposed that the group make the following
choice: either I would select one of the nine solutions at
random or the organization would be left as it was. My
suggestion was accepted and a vote was then taken. It was
unanimous in favor of my making a random choice.

I didn't have to make the choice because the outcome of

the vote suggested to the managers that they select one
member from each team to form a group that would
rework the solutions until they reached one on which all
agreed. This was done and it succeeded.

It is frequently easier to get managers to agree that consen¬

sus is desirable than it is to obtain it. The additional time
and effort required to reach consensus, however, is usu¬
ally more than compensated for by the reduction in time
and effort required to implement the decision it produces
and the increase in the effectiveness of that implementa¬
tion.
CONSENSUS

REFLECTIONS
 41
Thinking, Reading,
and Talking

Over the last 30 years I have spent as much time in corpo¬

rate quarters as I have in the halls of ivy. The differences
are striking. Although only a few aspects of academic life
are worthy of corporate emulation, these few would do
corporations a lot of good.

I have seldom seen a corporate manager sitting in his

office deep in thought, reading something other than a
business document, or discussing an idea rather than the
problem at hand. These are activities that occupy much of
an academic's time and a great deal of this time is fruitful.

Corporate cultures almost universally require managers to

look busy, to appear to be doing something that is clearly

162
THINKING, READING, AND TALKING 163

business related. Thinking, reading books or journals, and

discussing the ideas extracted from them is not generally
considered to be “doing something." If managers engage
in these activities, they must do so on "their own time."
This is "not what they are paid to do."

Occasionally, of course, managers take part in manage¬

ment development programs in which they are exposed to
and discuss ideas with their peers and purported experts;
but it is not for this purpose that they are sent to these
meetings. The expectation is that they will pick up some
useful information, tools, or techniques—not ideas. If they
were sent for ideas, they would be asked to share them
with others when they returned. I have never heard of
such a request being made.

These reflections were brought to mind a short time ago

when I was having dinner with the CEO of a major corpo¬
ration. During the meal, in which we did talk about ideas,
he suddenly asked me the name of the last book I had
read. I told him and we discussed it for a while. Then he
asked me what book I had read before the last and we dis¬
cussed that one also. When he made a third request I
asked him why these questions. He told me that he did a
great deal of reading "on his own time" but had no one to
discuss it with at work. He said that his colleagues read
little that was not immediately relevant to their work and
that discussion of his reading at work would make him feel
guilty; it would be perceived as unproductive.

Later I discussed this conversation with a number of his

associates and learned that, in fact, they also did a good
164 THINKING, READING, AND TALKING

deal of reading "on their own time" and felt it inappropri¬

ate to discuss at work. But they all said that they would
very much like to. 1 suggested that they organize periodic
"brown-bag lunches"—lunch-ins at which they hold such
discussions. They thought this was a good idea but no one
was willing to take the initiative.

When corporate executives visit me at the University I try

to have them take part in a bull session with students and
faculty. They invariably enjoy these sessions and tell me
how much they would like to participate in more of them.
My suggestion that they hold similar sessions at their
companies almost always evokes a shrug of the shoulders
and an expression of futility.

What a pity that so many managers indulge in such self¬

deprivation and maintain a state of intellectual
undernourishment.

The anti-intellectualism that pervades most corporations is

costly. It makes it difficult for managers to keep up to date
with relevant ideas and to use them in their work. It also
deprives them of the even more important opportunity to
convert apparently irrelevant ideas into relevant ones.
Moreover, it denies them the fun and stimulation that can
be derived from "kicking ideas around," an activity that
enhances the quality as well as the productivity of work
life.

The search for and development of good ideas should be a

continuing preoccupation of managers. Such a preoccupa¬
tion is not possible without thinking, reading, and
THINKING, READING, AND TALKING 165

discussion. Without thought, there is little learning. With¬

out reading, there is little to think about. Without
discussion, it is difficult to distinguish between good and
bad thinking.

REFLECTIONS
 42
Learning

As everyone knows, we learn from our mistakes. But we

have to know that we have made one before we can learn
from it. Unfortunately, in the gap between making a deci¬
sion and becoming aware of its results we often fail to rec¬
ognize a mistake because our memory plays tricks on us;
for example, at a meeting of corporate executives in which
an inventory-control study was initiated, I suggested a
pool on the change in the value of the inventory that
would be brought about by the research. The suggestion
was enthusiastically accepted. Each manager recorded his
prediction, expressed as a percent change, on a 3 x 5
card, signed it, inserted it in an envelope with a 10 dollar
bill, and sealed the envelope. The president gave the en¬
velopes to the corporation's secretary for safekeeping.

166
LEARNING 167

The managers were kept well informed during the course

of the study. When the results were reported to them it
was decided to settle the pool. The president called the
secretary and asked him to bring in the envelopes. The
secretary did not appear. The president called again and
told him to come at once. A very upset secretary appeared
quickly and told the group that the envelopes could not be
found. He could not explain their disappearance.

The managers were let down considerably. I suggested

that all was not lost. I asked if they remembered their orig¬
inal estimates. They said they did. Then I asked each of
them to record the original prediction on another card,
sign it, and put up another 10 dollars.

When they had done so and 1 had collected their cards and
money, I withdrew the original envelopes from my
briefcase and proceeded to compare the first predictions
with the "remembered" versions. There was significant
variation. Needless to say, the ones remembered were
much more accurate than the originals.

Our memories are far from passive. They modify their

content to make us appear better to ourselves than we ac¬
tually are. Therefore, when we make important decisions
whose effects will not be known for some time, we should
make a record of them that would contain

1. Their expected effects and when we expect them.

2. The assumptions on which our expectations are

based.
168 LEARNING

3. The information used in making the decisions.

4. A description of how the decision was made and

who participated in making it.

Decisions recorded in this way can be monitored not only

to reveal our mistakes and the reasons for them but also to
let us know, even before the results are in, when an un¬
derlying assumption is false. This often enables us to take
action to prevent a mistake from being made.

A study made a number of years ago at the General Elec¬

tric Company showed that managers who recorded the ef¬
fects they expected of their decisions learned significantly
more, and more rapidly, than those who did not.

The only ones who are incapable of learning are those who never
make a mistake or are unaware of the mistakes they make.

REFLECTIONS
 43
Understanding

Information, knowledge, and understanding are very dif¬

ferent things. Information is contained in descriptions: an¬
swers to questions that begin with such words as "who,"
"when," "where," "which," and "how many." Knowl¬
edge is contained in instructions: answers to "how-to"
questions. Understanding is contained in explanations: an¬
swers to "why" questions.

A wise man once said that an ounce of knowledge is worth

at least a pound of information and an ounce of under¬
standing is worth at least a pound of knowledge. Never¬
theless, the time spent by most managers and manage¬
ment educators in acquiring and transmitting information.

169
170 UNDERSTANDING

knowledge, and understanding is inversely related to

their values.

Many explanations are nothing but restatements in differ¬

ent words of the facts to be explained; for example, in a
study carried out for a multinational candy company my
colleagues and I learned that per capita consumption of
sugar in England was greater than it was in the United
States. In an effort to find out why, I called a friend who
was the marketing vice-president of a sugar company. I
asked him if he was aware of the different consumption
rates. He said he was. Then I asked if he could explain it.
"Of course," he said. "The English like sugar more than
we do." Because this was not what I was looking for, I
asked, "How do you know they like it more than we do?"
He replied, "They eat more of it, don't they?" This is like
telling someone that a sleeping pill puts them to sleep be¬
cause it's a soporific.

Misunderstanding, or even a lack of understanding, can

get us into serious trouble. Consider a company that em¬
ployed a large number of women as salaried inspectors of
small items produced in the millions each year. The pro¬
ductivity of these women had decreased over the prece¬
ding five years and the plant manager decided to take cor¬
rective action. He thought he knew what made the
women tick—money. As a result, he designed a
piecework-compensation system that would require the
women who wanted to maintain their earnings to inspect
more items per day than they were doing at the time but
fewer than they had five years earlier. If they handled as
UNDERSTANDING 171

many as they once had, their earnings would increase

significantly.

The plant manager proposed this scheme to the leaders of

the union that represented the women. They rejected it
summarily. The manager was furious and threatened to
impose his scheme anyway. The union countered with the
threat of a strike if he tried.

In desperation the manager turned to his research staff for

help. Their investigation revealed that the women hated
the plant manager because several years earlier he had re¬
fused to change their working hours. Most of them
wanted to get out of work early enough to be home when
their children returned from school. Moreover, they did
not want to increase their earnings because this would put
them in competition with their husbands for the role of
family breadwinner. They saw their earnings as providing
the "extras," not the necessities.

Armed with this understanding, the researchers designed

a new compensation system in which a "fair day's work"
was defined as the high average output of five years
earlier. It was proposed that when the women had in¬
spected this number of items they would be free to leave
the plant. The women accepted this proposal with enthu¬
siasm. Their rate of inspection increased dramatically and
so did its quality.

The most important question a manager can ask when

confronted with the unexpected or the undesirable begins
with why.
172 UNDERSTANDING

REFLECTIONS
 44
Managment Education

In my opinion, management education deserves all the

criticism to which it has been subjected. In general it fails
to prepare students adequately for the practice of manage¬
ment. But I think that most of the many changes proposed
for improving it point in the wrong direction—to curricu¬
lar changes. The major deficiencies in management educa¬
tion are not in what is taught but in how it is taught. The
medium, the educational process, is more at fault than the
messages transmitted by it; for example, a major part of
management education is devoted to trying to solve prob¬
lems given to students by teachers. As a result, students
unconsciously come to believe that it is natural for prob¬
lems to be given to them. In the real world, however, prob-

173
174 MANAGEMENT EDUCATION

lems are seldom given; they must be taken. Nevertheless,

students are neither taught nor learn how to take prob¬
lems.

In management, problems usually have to be extracted

from complex, unstructured, and messy situations. This
can be learned only by practice, preferably under the guid¬
ance of someone who knows how. In learning to take
problems, like learning to drive an automobile, instruction
has little value without demonstration and practice.

Classroom work on case studies is intended to provide

this practice but it cannot succeed. Cases are distilled de¬
scriptions of the real world. They reduce the complexity of
reality and include only what those who prepare them
consider to be relevant. Deciding what is relevant is a ma¬
jor part of taking problems.

Case studies do not present problems; they present exer¬

cises. The difference between problems and exercises is not
widely appreciated, especially by educators; for example,
a statistician once gave me the following so-called prob¬
lem: “You reach into a bowl containing small black and
white balls and pull out a handful. X percent of them are
black; the others are white. Now, if you randomly draw an
additional ball from the bowl, what is the probability it will
be black?"

Instead of answering, I asked: "How do you know all the

balls in the bowl are black or white?" He told me I was to
assume it; it was given. But, I argued, I wanted to know
MANAGEMENT EDUCATION 175

how it was taken; if I knew this I could probably answer the

question easily. He argued that such knowledge would
spoil the problem. No, I said, it was already spoiled. It was
an exercise, not a problem. An exercise is a problem from
which some of the information required to formulate it has been
removed.

Doesn't solving exercises help one to learn how to solve

problems? It may, but solving problems is not the point;
taking problems is, and one does not learn how by solving
exercises. Moreover, learning how to solve exercises
doesn't help much in learning how to solve problems.
Teaching people how to box or play baseball with one arm
tied behind their backs is not an effective way of teaching
them how to box or play baseball with both arms free.

A student can best learn to identify, formulate, and solve

problems by being exposed to and dealing with raw real¬
ity. Unfortunately, raw reality can't be brought into the
classroom; but the classroom can be taken into the real
world. Management education should require faculty and
students to work together with responsible managers to
identify, formulate, and solve problems that these mana¬
gers actually face and this ought to be done in the environ¬
ment in which they are faced. This would not only educate
the students, it would also educate the faculty.

Faculty members need to learn how to take problems at

least as much as their students. Teachers cannot teach
what they do not know and especially what they do not
know that they do not know.
276 MANAGEMENT EDUCATION

REFLECTIONS
 45
Town and Gown

Business schools do a much better job of educating their

faculty than their students. Faculty members more than
students are exposed to more business enterprises and
more enterprises are exposed to them. Moreover, faculty
members have many opportunities to absorb the learning
of others, to reflect on, discuss, and try to understand
their own learning and that of others, and to organize
what they have learned to transmit to others. Managers
have few of these opportunities. Much of faculty learning
takes place in or is stimulated by discussion groups, or¬
ganized or spontaneous. These groups are rare in business
environments.

On the other hand, most managers learn more of rele-

177
178 TOWN AND GOWN

vance to their work on the job than they did at school. At

work they absorb a great deal of information and some
knowledge (skills) but gain little understanding. One can
know a great deal about how to run an automobile or a
business without understanding why they behave as they
do. In contrast, business school faculties generally have
more understanding and less knowledge of management
and business than practicing managers.

Finally, students in business schools acquire little relevant

knowledge or skills. To be sure, they learn all kinds of
techniques, but most are not applicable to the problems
they subsequently face on the job. They absorb little of the
small amount of understanding that faculty members try
to transmit because, as students, they generally lack the
experiential base required for understanding. What busi¬
ness school students usually learn well are (1) a vocabu¬
lary that enables them to speak with authority about sub¬
jects they don't understand and (2) a set of principles that
have demonstrated their ability to withstand large
amounts of disconfirming evidence. Because of this, grad¬
uated students have to go through extensive wnlearning,
as well as learning, before they become useful to their first
employers.

How can the advantages of on-the-job education of mana¬

gers and on-campus education of faculty members be
combined and applied to managers, management educa¬
tors, and students of management?

It might be done by organizing discussion groups at the

workplace, by bringing together managers, educators.
TOWN AND GOWN 179

and students of management. These groups could be de¬

voted to presentations and discussion of problems cur¬
rently faced by managers, faculty members, and students
and their efforts to solve them. Each would try to teach the
others. There are no more effective ways of learning than teach¬
ing and trying to solve problems with those who have them.

Faculty members and students would benefit by sharing

and reflecting on the managers' experiences. They would
learn why many of the tools and techniques presented to
students are not useful and they would become aware of
the types of problem for which there is no relevant or ef¬
fective body of knowledge. They might even succeed in
getting some of their ideas tried and tested in practice.

Students could exercise their natural curiosity, skepticism,

and cynicism by asking critical questions. (Even when
they don't know the right answer to questions asked by
others, they often know the right questions to ask them.)
They would learn how to take problems, not merely solve
those given to them. They would also learn how much
more important it is to formulate a problem correctly than
it is to find the best solution to one incorrectly formulated.
A less than optimal solution to the right problem is more
useful than the optimal solution to the wrong one.

And what might managers get out of these sessions? Some

useful information, knowledge, and understanding, of
course. But of greater importance, some ideas and the op¬
portunity to think about and discuss them. They might
even be moved to make thinking about and discussion of
ideas fashionable in the work environment.
180 TOWN AND GOWN

REFLECTIONS
 46
The Fallacy of
Forecasting

A great deal of the present is being wasted with efforts to

forecast the future. In an environment that is rapidly
changing and becoming more complex, our ability to pre¬
dict the future necessarily decreases. Preparing for an in¬
accurately forecasted future is often worse than doing
nothing. This is reflected in the old saying: ''He who lives
by the crystal ball ends up by eating glass."

Nevertheless, those who conceive of planning as

preparing for a predicted future argue correctly that we
benefit from forecasting the weather and preparing for it,
although both are done imperfectly. True, but there is a
significant difference between our relationship with the

181
182 THE FALLACY OF FORECASTING

weather and a corporation's with what it forecasts in its

planning.

Some believe that carrying an umbrella prevents rain and

washing a car causes it. Nevertheless, our preparations for
the weather have absolutely no effect on it. On the other hand,
corporate planners forecast such things as the behavior of
consumers, suppliers, competitors, and governments; and
these things are affected by what corporations do. In fact,
the principal purpose of planning is to affect them. There¬
fore, once a corporate plan that is based on a forecast has
been prepared, the effects of that plan on what has been
forecast should be taken into account by revising the initial
forecast. But revising that forecast requires revision of the
plan, which in turn requires another revision of the fore¬
cast, and so on ad infinitum. If all this were done, planning
would take a course like that of the gilly-galoo bird, which
flies in ever decreasing concentric circles until it disap¬
pears up its own anatomy.

Now, of course, this is not done. Predict-and-prepare

planners treat the environment like the weather; they act
as though it will be unaffected by their plans. Therefore,
they try to control the effects of the environment on the
organization planned for. The assumption that the envi¬
ronment is unaffected by what corporations do is
sufficient to invalidate the forecasts used by planners and
to make their preparations less effective than they desire.

Most forecasting is based on projections of the past into

the future. Such extrapolations assume that the future is
completely determined by the past. This assumption is
THE FALLACY OF FORECASTING 183

sometimes approximately true for the very near future; but

the more distant the future forecast, the more it depends
on what will happen between now and then. This is even
true for the weather.

Put another way: the more distant the future, the more it
depends on decisions still to be made; therefore, the more
subject it is to control. For this reason corporate planning
should be directed toward trying to control the future, not
the effects on a corporation of a future assumed to be out
of its control. This is exactly what we have done with the
weather.

Buildings are built to bring the weather under control.

They eliminate the need to forecast the weather where we
work and live. Even if we had perfect forecasts we would
be better off working and living indoors than out.

Corporate planning should not consist of predicting and

preparing for an uncontrolled future but of designing a de¬
sirable future and finding or inventing ways of approximating it
as closely as possible.

REFLECTIONS
 47
Obstructions to
Progress

The principal obstructions between an organization and

what it most wants to be lie within the organization itself.
Unfortunately, most managers and planners assume that
these obstructions are imposed from without. Therefore,
even if they succeed in removing or evading externally
imposed obstructions, they often fail to get what they
want because they neither remove nor evade the self-
imposed constraints that are even more obstructive.

The principal obstruction to an organization's progress is

usually the opinion of its managers and planners as to
what is feasible; for example, in 1976 one of Mexico City's
principal planners asked me to review and help him select
the best of six alternative plans he had prepared to reduce

184
OBSTRUCTIONS TO PROGRESS 185

traffic congestion in his city. After reviewing his plans I

said that, in my opinion, none of them would work. He
was shocked and wanted to know why. I explained that all
of his plans were directed at increasing the supply of
transportation enough to meet currently unsatisfied de¬
mand. He had neglected the fact that a new supply always
creates new demand and this often exceeds the demand
that was previously unsatisfied.

“Then you're saying there's no way to solve urban conges¬

tion problems," he said. "No," I replied, "they can be
solved by reducing demand." This, he told me, was nei¬
ther desirable nor possible in a democracy. I disagreed and
he challenged me to tell him how it could be done.

I said "by moving a significant part of the federal govern¬

ment out of Mexico City." I pointed out that a large num¬
ber of people in the city were employed by the federal gov¬
ernment and that their emigration would reduce
congestion by more than all six of his plans combined. Be¬
sides, there were good economic, ecological, and social
reasons for moving the capital. "Of course," he said, "but
you can't just up and move a nation's capital." I pointed
out that the United States had done so twice. "But," he
countered, "that was two hundred years ago." "What
about Tanzania?" I asked. "It is contemplating a move of
its capital right now." "But that," he said, "is in Africa."

He asked if I was aware that Mexico City had been the

capital of the Aztec empire. I said I was, but that did not
explain why the capital couldn't be moved. He then con-
186 OBSTRUCTIONS TO PROGRESS

eluded that I probably never would understand because I

was not a Mexican. There was no debating that point.

After an awkward pause in our conversation he asked if I

could suggest another way by which demand could be re¬
duced. I suggested that it be done by cutting the siesta, the
two- to three-hour midday break, to no more than one
hour. Midday trips between work and home were a
significant part of the total number of trips taken and
could be drastically curtailed. Moreover, I pointed out that
today the siesta was seldom used for sleeping, although
the bedroom was frequently involved. He then informed
me that the siesta was an important part of Mexican cul¬
ture and could not be changed. When I challenged him he
said once again that I would not understand because I was
not a Mexican.

Very shortly thereafter Lopez Portillo assumed the presi¬

dency of Mexico. In his inaugural address he announced
his intention to disperse the federal government and de¬
creed that all new federal buildings be built outside the
city. He also announced a reduction in the midday break.
In his first official act he had done two things that the city
planner had told me were impossible.

The impossibility was obviously in the planner's mind,

not his environment. This locus of obstructions to prog¬
ress was recognized by the great American philosopher
Pogo. On returning from a trip into the woods which he
made "to hunt the enemy," he was asked by a friend if he
had found them. He said he had. His friend asked who
they were. Pogo replied, "The]/ is us."
OBSTRUCTIONS TO PROGRESS 187

REFLECTIONS
 48
Planning Backward

Most corporations plan forward, from now to then, a

point in the future. Such prospective planning consists,
first, of predicting the future to be prepared for and then
preparing for it. The future prepared for is assumed to be
out of the corporation's control, but its effects on the cor¬
poration are assumed to be at least partially controllable.

There is another type of planning—the introspective—that

moves backward from what we want to be to what we are.
It enables us to get much closer to what we want to be
than planning forward from what we are. It does so be¬
cause it expands our concept of what is feasible, our reper¬
toire of potential courses of action.

188
PLANNING BACKWARD 189

In planning forward the feasibility of each possible course

of action is usually evaluated separately. It is assumed that
a course of action that appears to be infeasible when evalu¬
ated separately will continue to appear so when evaluated
as part of a plan. This is not true. A plan is a set of inter¬
acting courses of action, a system of actions. A system al¬
ways has properties that none of its parts has; for exam¬
ple, an airplane can fly us from one place to another but
none of its parts can. Therefore, a plan can be feasible
even though none of its parts considered separately ap¬
pears to be so. Actions that appear to be infeasible when
considered separately can appear feasible when consid¬
ered as part of a system of actions. Planning backward en¬
ables us to consider each course of action as part of a sys¬
tem.

Introspective planning begins by assuming that the organ¬

ization planned for disappeared last night. It no longer ex¬
ists but its environment remains unchanged. The planners
then design the organization with which they would most
like to replace the one that “disappeared." This is called
an idealized redesign because the only constraints imposed
on it are, first, that it be technologically feasible—no sci¬
ence fiction—and, second, that the organization designed
be capable of surviving in the current environment. How¬
ever, the design need not be one that can be realized now. There¬
fore, practicality is irrelevant in making idealized design
decisions. Yet, despite this, when such designs have been
completed—and many have been—the responses to them
have almost always been: “My God, we could realize most
of it if we really tried."
190 PLANNING BACKWARD

An idealized redesign of a system reveals that the princi¬

pal obstructions between an organization and what it most
wants to be lie within the organization and that these bar¬
riers can be removed. This greatly expands our concept of
what is feasible; for example, how feasible do the follow¬
ing proposals, considered separately, appear to be: that
the capital of France be moved from Paris and that Paris be
converted to an open self-governing city no longer subject
to the government of France? Notwithstanding the appar¬
ent infeasibility of these proposals, the cabinet of France
accepted them in the mid-1970s and has since taken steps
toward their realization. How come?

In the early 1970s an idealized redesign of Paris was pre¬

pared by a large number of people both in and out of gov¬
ernment. Although many political differences lay between
them, they agreed that Paris ought to be the capital of the
world. They didn't mean the capital of a world government
because they didn't believe that such a government would
come about in the foreseeable future. What they did mean
was the principal location of the growing number of
organizatons that deal with international affairs. Once this
mission had been adopted, the redesigners saw the move¬
ment of France's capital from Paris and its conversion into
an open city as not only feasible but necessary.

Planning backward consists of trying continuously to close

the gap between what we ideally want to be now and what
we are. We can get much closer to our ideals by working
backward from them than by working forward from
where we are. What follows is a case in point.
PLANNING BACKWARD 191

REFLECTIONS
 49
The Clark-Volvo Joint
Venture

On Wednesday, January 28, 1985, the Clark Equipment

Corporation and AB Volvo of Sweden announced their in¬
tention to merge the Clark Michigan and Volvo BM divi¬
sions. This has since been done. Clark and Volvo have cre¬
ated one of the world's largest construction and mining
equipment corporations. It is called VME, which are the
initials of Volvo, Michigan, and Euclid. The story behind
this joint venture merits a book. A brief commentary can
hardly do it justice, but it is worth a try.

In 1983 Clark Michigan, which had not performed well

during the recession, initiated a major effort to plan itself
out of its mess. It began by formulating that mess: the fu-

192
THE CLARK-VOLVO JOINT VENTURE 193

ture that it would face if it continued its then current prac¬

tices and policies and if its environment changed only in
expected ways. The future that was revealed by this analy¬
sis spelled doom for the division. In the time it had availa¬
ble it could neither generate nor acquire the resources re¬
quired to redirect its future.

Rather than give up in despair, Clark Michigan's manage¬

ment designed without any constraints an ideal competi¬
tor of the two giants that dominate its industry—
Caterpillar and Komatsu. Again without constraints,
Clark Michigan's management selected those companies
in its industry that, with itself, would most closely approx¬
imate the ideal company it had designed. Three were
identified. These organizations, together with Clark
Michigan, resided in four different countries on three dif¬
ferent continents.

Armed with this concept and design, Clark approached

Daimler-Benz, which owned the Euclid Truck Company,
and in almost record time arranged to acquire that com¬
pany. Clark had already begun discussing a joint venture
with Volvo BM, the construction equipment division of
AB Volvo. Volvo displayed interest in the concept but had
a number of reservations. Clark responded by proposing
that a team be formed to address these reservations by
preparing a detailed design of the proposed joint venture
and evaluating its financial prospects.

Volvo agreed. A 10-man team, which consisted of the gen¬

eral managers of Clark Michigan and Volvo BM and their
194 THE CLARK-VOLVO JOINT VENTURE

production, marketing, financial, and engineering mana¬

gers, was formed. This team began by producing a de¬
tailed idealized design of the Clark-Volvo joint venture. It
drew up as nearly feasible an approximation of this design
as it could. Finally, it made a financial comparison of the
feasible design and the sum of two companies taken sepa¬
rately. This revealed that the proposal had potential
financial advantages. The design team also felt that it had
taken care of all of Volvo's reservations.

The feasible design was presented to the executives and

boards of the two parent companies in the autumn of
1984. The design was approved and at the end of that year
a commitment was made to implement it.

In addition to producing a creative and exciting design,

the team members welded themselves into a homogene¬
ous management group that overcame all the major cul¬
tural differences that were initially of great concern to both
parties. By working closely for six months they even
reached agreement on who should occupy each executive
position in the new company.

With this cooperative design process the team anticipated

and dealt with almost every problem that can arise in a
joint venture. The process as well as its product enabled
the two parent corporations to make their commitments
with a degree of confidence that seldom characterizes such
efforts.
THE CLARK-VOLVO JOINT VENTURE 195

REFLECTIONS
 50
Responsiveness

Managers can reduce their dependence on forecasts by

increasing their ability to respond rapidly and effectively
to unexpected events; for example, when we take an auto¬
mobile trip we don't arm ourselves with forecasts of road
conditions, the number of drivers, how they will drive,
and so on. Nevertheless, the likelihood of our reaching
our destination depends critically on these things. We
don't forecast them because we can respond quickly and
effectively to most of what we might encounter. This
means that we can avoid or minimize the undesirable ef¬
fects.

We can increase our response capabilities by preparing for

possible but unexpected events. When using an automo-

196
RESPONSIVENESS 197

bile, for example, we do not ordinarily expect a flat tire,

but because we know one is possible we carry a spare. Our
preparation is based on the assumption that a flat can occur,
not on a forecast that one zvill occur. On the other hand, we
don't carry a spare engine even though the one in our car
might fail. If it did, we couldn't replace it by ourselves. We
can prepare for such a failure better by joining an automo¬
bile club.

It would be foolish, of course, to plan for everything that

can occur. We prepare only for those things that have a
significant chance of occurring and with which we can
deal without great cost or inconvenience.

Preparation for possibilities is called contingency planning,

which consists of measures that can reduce the effects of
conditions and events that would be costly or inconven¬
ient if we did not anticipate them.

Because unexpected conditions and events occur fre¬

quently, corporate plans never work out exactly as ex¬
pected. Therefore, they should be supplemented by con¬
tingency plans, preparation of which requires, first,
identifying the possible events that can mess up normal
plans; second, estimating the likelihood of each event and
the cost and inconvenience of preparing for it; and, finally,
selecting those events that justify attention and dealing
with them.

Many of the unexpected events that spoil corporate plans

are the result of unanticipated behavior of parties outside
the corporation; for example, customers, suppliers, com-
198 RESPONSIVENESS

petitors, special interest groups, and government. By

using countermeasure groups we can improve our anticipa¬
tion of, and preparation for, the unexpected behavior of
others.

A countermeasure group is a research team set up to act as

though it were employed by a troublesome party; for ex¬
ample, a competitor, a supplier, or a customer. The task
assigned to the team is to determine how that party, the
"bad guy," can maximize obstructiveness. To facilitate
this maximization the countermeasure team is provided
with complete information about what the sponsoring cor¬
poration, the "good guy," intends to do.

Once the countermeasure team has determined how to

obstruct the corporation, the corporation's planners take
these obstructions into account and modify their initial
plan appropriately. The countermeasure group then tack¬
les the revised plan and tries to find ways of obstructing it.
This process continues until the corporate planners be¬
lieve they have covered all possible or likely contingencies
adequately.

One corporation recently engaged in just such an exercise.

It wanted to buy a plant put up for sale by one of its com¬
petitors. The firm interested in making the purchase was
reasonably sure that another competitor would try to pre¬
vent the sale because it would be to that competitor's dis¬
advantage. A countermeasure group, set up to represent
the potentially obstructive competitor, determined what
that competitor's first move should be if an offer to buy the
plant were made by the sponsoring corporation. This cor-
RESPONSIVENESS 199

poration's management then prepared its response to that

obstructive move. The countermeasure group returned to
work. Four countermeasure cycles were completed before
the countermeasure group could find no further way of
blocking the purchase.

When the offer to buy was made, the obstructive competi¬

tor acted as though it were following the script prepared
by the countermeasure team. Each of its moves had been
correctly anticipated and was met with an effective re¬
sponse. The plant was successfully acquired.

The cost of preparing for critical events that do not occur is gener¬
ally very small in comparison to the cost of being unprepared for
those that do.

REFLECTIONS
 51
Comprehensive,
Coordinated, and
Participative Planning

No group of corporate executives can prepare a compre¬

hensive corporate plan with or without the aid of a plan¬
ning staff. The most it can provide is an executive plan. The
difference between these plans is best revealed by an ex¬
ample.

A Mexican brewer decided to give all of his employees an

opportunity to participate in corporate planning. Those
interested—and most employees were—received relevant
instruction. They were then organized into small, homo¬
geneous planning groups.

The top executives of the company formed one planning

group but so did the janitors in the brewhouse. Obviously

200
PARTICIPATIVE PLANNING 201

these two groups did not do the same thing. The execu¬
tives considered such issues as diversification, acquisition,
joint ventures, entry into new markets, facility require¬
ments, and raising necessary capital. On the other hand,
the janitors concerned themselves with redesigning the
lavatories in the brewhouse, care of which occupied most
of their time.

I have described this division of planning labor to many

groups of executives. Invariably they smile. When they do
I say, "This is cute, isn't it?" They nod in agreement. Then
I ask: "But what has this to do with corporate planning?"
Again they nod in agreement. I follow with an appeal for
an honest answer to the following question: "What would
happen to your corporation if the top executives were sent
away for three months incommunicado and their offices
were locked?" After some embarrassed tittering, the an¬
swer invariably given is: "Not much, if anything." (Some
even suggest the possibility of improved corporate per¬
formance.) Then I follow with: "What would happen if the
janitors were sent away for three months incommunicado
and the lavatories were locked?" They get the point.

The point is not that the janitors are more important to the
corporation than its executives but that the lavatories are
as important a part of the corporation as its debt, product
line, and markets. Therefore, a company's lavatories
should be taken into account in its planning. If they are
not considered, the planning is not comprehensive, how¬
ever good it may be.

"But," some respond, "the lavatories will be taken care of

202 PARTICIPATIVE PLANNING

even if not planned for. So why include them in the plan?"

The answer lies in the way they will be taken care of and in
how those who take care of them will feel about their
work. This is shown in what happened to the initial plan
prepared by the janitors in the brewhouse. The people
who did the brewing also prepared a plan. It included a
redesign of the brewhouse. When they had completed
their revision they and the janitors met to study the two
plans. It was immediately apparent to both groups that
they did not mesh: the brewhouse workers had neglected
the lavatories and the changes suggested by the janitors
interfered with the brewing process. Therefore, a com¬
bined group was formed to produce a coordinated plan for
producing beer and for getting rid of it.

A similar process took place on the packaging line. When

an integrated plan for this line was completed the brew¬
house and packaging groups were brought together and
shown both plans. Again they did not mesh. A joint effort
subsequently succeeded.

This process continued across the lowest level of the or¬

ganization and was duplicated at every other level. At the
same time the planning staff of the corporation reviewed
each plan as it was produced and called it to the attention
of any other group, at whatever level of the organizaton,
to which it was relevant. This led to meshing plans pro¬
duced at different as well as at the same levels.

As a result of this process employees throughout the cor¬

poration came to understand how their activities affected
and were affected by other corporate activities and, most
PARTICIPATIVE PLANNING 203

important, how they affected the overall performance of the cor¬

poration. This understanding led to a radical transforma¬
tion: each unit began to plan and operate to optimize the
performance of the corporation as a whole rather than just
its own. The improvement in corporate performance
brought about by the reorientation of its parts is the princi¬
pal benefit to be derived from corporate planning. This be¬
nefit cannot be realized unless the planning is comprehensive,
coordinated, and participative.

REFLECTIONS
 52
Continuous Planning

The planning pie can be sliced in many different ways,

one of which consists of dividing it into six parts or pha¬
ses.

1. Setting up: formulating the situation the organization

is in by identifying the threats and opportunities it
faces, tne ways in which they interact, and the ob¬
structions that constrain what can be done about
them. This, I believe, is best done by projecting the
future the organization would have if it were to con¬
tinue on its present course and if its environment
were to change only in expected ways. This reference
projection is a picture of the “future that the organiza-

204
CONTINUOUS PLANNING 205

tion is in." It is not a forecast of the organization's fu¬

ture because it is based on two assumptions ("ifs")
known to be false. If the organization intended to
continue on its present course, it would not be plan¬
ning, and no organization expects its environment to
change only in expected ways. Nevertheless, a refer¬
ence projection is useful because it reveals how and
why the organization's performance would deterio¬
rate if it were to continue on its current path.

2. Ends planning: designing an ideally desirable present

and determining the differences between this pres¬
ent and the future revealed by the reference projec¬
tion. These differences constitute the gaps that the
remainder of the planning process should attempt to
close or reduce.

3. Means planning: selecting or inventing the means by

which efforts will be made to close or reduce the
gaps. Means assume such varied forms as courses of
action, practices, projects, programs, and policies.

4. Resource planning: estimating how much of each type

of resource will be required by the means selected,
when they will be required, and how much will be
available; determining how the deficiencies revealed
will be removed by acquisition or generation of addi¬
tional resources and how the excesses revealed will
be used or disposed of. Five types of resource should
be taken into account: people; facilities and equip¬
ment; materials, energy, and services; money;-and
information.
206 CONTINUOUS PLANNING

5. Design of implementation: scheduling and assigning re¬

sponsibility for the things to be done.

6. Design of controls: setting up procedures for (a)

determining whether they are being done as ex¬
pected and, once done, whether they are producing
the intended effects, and (b) taking corrective action
when implementation or its effects deviate from ex¬
pectations.

Resource planning and controlling the implementation

and effects of a plan make it necessary for planning to be
continuous, not an off-again-on-again process.

The resources required by the means selected in the plan¬

ning process seldom, if ever, are equal to the amount
needed. In turn, demands are for sequential revisions of
ends, means, and resource plans; for example, in a plan
produced by a company in the consumer products indus¬
try the planners proposed building a new plant that would
be about twice the size of the largest plant of its type that
had ever been built. Because of the large investment in¬
volved, board approval was required. The board accepted
the need for the large plant but was unwilling to incur in¬
creased debt to build it. The planners were instructed to
try to generate the required capital internally.

Fortunately the planners found a way to reallocate pro¬

duction to the company's existing plants that would re¬
duce costs enough to provide the required capital and
more. When they returned to the board with this result
CONTINUOUS PLANNING 207

construction of the new plant was approved. However,

they were asked how they intended to use the excess capi¬
tal that would be generated. The planners suggested
diversification. The board agreed but asked what kind of
diversification. The planners replied that plans had not yet
been prepared. The board asked them to do so and submit
their results.

Three possible diversification plans were developed, each

based on a different type of product or service. When
these plans were presented to the board it decided it
wanted to pursue all three. Because enough capital would
be available to implement only one of the plans, the plan¬
ners were sent back to find ways of generating the capital
required to implement the others.

This continuous planning cycle began about 20 years ago.

It is still going on.

Furthermore, because plans are never implemented ex¬

actly as expected and the effects of implemented plans al¬
ways deviate from expectations, effective control requires
that plans be modified almost continuously.

To be effective, planning cannot be a sometime thing, off-

again-on-again. It must go on all the time. The output of
effective planning is no more a discrete plan than the out¬
put of producing a motion picture is a still photograph
clipped from it. It is an ever-changing plan, one that
reflects the continuous learning and adaptation of those
who prepare it.
208 CONTINUOUS PLANNING

REFLECTIONS
 DATE DUE

FE 1 0 *88

CATLORO 1 ARINTCO IN U.« A.

 tlHRYGROUE COLLEGE LIBRARY
Management in small doses
658 ftc5

3 1127 DDDllEEfl 3

658
Ac5 Ackoff, R. L.
Management in small
doses

658
Ac 5