Apart from any fair dealing for the purposes of research or private study, or criticism or review, as permitted under

the relevant copyright,

designs and patents acts, this publication may only be reproduced, stored or transmitted, in any form or by any means, with the prior
permission in writing of the publishers.
 Praise for the Winner of the Edwin G. Booz Award
for the
most insightful, innovative management book of the year

"The Living Company is eminently readable. The style is

conversational and jargon-free. While it is reflective and
thoughtful, it is always relevant, meaningful and deals with
issues of central significance to organisations today.

For those interested in the bigger picture, this book provides a

wonderful landscape."

Doron Gunzburg, Monash Business Review

"This is a thoughtful, reflective and philosophical book. It

does not prescribe quick fixes. The author draws on rich
imagery from agriculture, horticulture, psychology and
nature. For example, he explores how the blue tit learned to
pierce aluminium milk bottle tops, while the robin didn’t.
The organizational issue from this is how learning is
distributed and passed on.

The author’s experience and reflection as articulated in this book

provide a valuable resource for further insight and understand-
ing of how organisations survive, learn and flourish."

David Coughlan,
Leadership & Organisation Development Journal
 "Arie de Geus is an international figure who has not only
been a key influence on scenario planning, but is also credited
with originating the concept of the learning organisation.

The world, or at least part of it, is ready to hear the message

that profits are only a symptom of success and not an end in
themselves. Many people want to believe in the wisdom of
power-sharing in organisations. Others long for a move away
from individualism to a greater emphasis on the importance
of community. But the book also sells because it is based on
experience and is incredibly well written by a man whose conversation
and manner are compelling."

Jane Pickard, People Management

"Reading The Living Company is a refreshing experience in a

period when conventional wisdom emphasises short-term
returns on capital; it should be read by all those who think
that there is more to business than that."

Russell Sparkes, RSA Journal

 The Living
Company
Growth, Learning and Longevity
in Business

Arie de Geus
 This paperback edition first published
by Nicholas Brealey Publishing Limited in 1999

36 John Street 1163 E. Ogden Avenue, Suite 705-

229
London Naperville
WC1N 2AT, UK IL 60563-8535, USA
Tel: +44 (0)171 430 0224 Tel: (888) BREALEY
Fax: +44 (0)171 404 8311 Fax: (630) 428 3442
http://www.nbrealey-books.com

First published in hardback in 1997

Reprinted 1997, 1998 (twice)

© Longview Publishing Limited 1997, 1999

The right of Arie de Geus to be identified as the author of this work
has been asserted in accordance with the Copyright, Designs and
Patents Act 1988.

ISBN 1-85788-185-0

British Library Cataloguing in Publication Data

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

All rights reserved. No part of this publication may be reproduced,

stored in a retrieval system, or transmitted, in any form or by any
means, electronic, mechanical, photocopying, recording and/or
otherwise without the prior written permission of the publishers. This
book may not be lent, resold, hired out or otherwise disposed of by
way of trade in any form, binding or cover other than that in which it
is published, without the prior consent of the publishers.

Digital processing by
The Electric Book Company, London, UK, www.elecbook.com

The ideas and views expressed in this book are those of the author and
not of any organisation with which he is associated.
Contents

FOREWORD 1
PROLOGUE: THE LIFESPAN OF A
COMPANY 7
PART I LEARNING 20
1 THE SHIFT FROM CAPITALISM TO A
KNOWLEDGE SOCIETY 21
2 THE MEMORY OF THE FUTURE 30
3 TOOLS FOR FORESIGHT 49
4 DECISION MAKING AS A LEARNING
ACTIVITY 70
PART II PERSONA (IDENTITY) 94
5 ONLY LIVING BEINGS LEARN 95
6 MANAGING FOR PROFIT OR
LONGEVITY: IS THERE A CHOICE? 123
PART III ECOLOGY 157
7 FLOCKING 158
8 THE TOLERANT COMPANY 170
9 THE CORPORATE IMMUNE SYSTEM 190
PART IV EVOLUTION 201
10 CONSERVATISM IN FINANCING 202
11 POWER: NOBODY SHOULD HAVE
TOO MUCH 220
EPILOGUE: THE COMPANY OF THE
FUTURE 235
NOTES AND REFERENCES 239
INDEX 246
ACKNOWLEDGEMENTS 253
Foreword
Peter M. Senge

IT WAS THROUGH ARIE DE GEUS, WHOM I MET OVER 15 YEARS

ago, that I first became seriously acquainted with the concepts
of organizational learning. That meeting began the journey of a
lifetime.
He introduced me to the famous study done at Royal
Dutch/Shell, where he was the coordinator of planning world-
wide, which found that the average life expectancy of Fortune 500
firms, from birth to death, was only 40 to 50 years. The study also
found many companies over 200 years old. Arie convinced me
that most corporations die prematurely — the vast majority before
their 50th birthday. Most large corporations, he said, suffer from
learning disabilities. They are somehow unable to adapt and
evolve as the world around them changes.
More importantly, he got me thinking for the first time
about the connections between low life expectancy and low vital-
ity of firms while they are still operating. Both are symptoms of
the overall health of the enterprise. Like individuals who are
unhealthy and can expect an early demise, most large, apparently
successful corporations are profoundly unhealthy. The members
of these organizations do not experience that their company is
suffering from low life expectancy. They experience most corpo-
rate health as work stress, endless struggles for power and con-
trol, and the cynicism and resignation that result from a work
 THE LIVING COMPANY 2

environment that stifles rather than releases human imagination,

energy and commitment. The day-to-day climate of most organi-
zations is probably more toxic than we care to admit, whether or
not these companies are in the midst of obvious decline.
This is a book of practical philosophy. It has been my expe-
rience that extraordinary practitioners like Arie can make unique
contributions to management thinking, but that their contribu-
tions are rarely acknowledged. Unlike academics who write
about what they have thought, practitioners think about what
they have lived through. Because the source of their thinking is
experience rather than concepts, they show how sometimes the
most profound ideas are the simplest.
At the heart of this book is a simple question with sweep-
ing implications: what if we thought about a company as a living
being?
This raises the obvious question: what is the alternative
view of a company if we do not see it as a living being? The alter-
native view is to see a company as a machine for making money.
The contrast between these two views — machine for mak-
ing money versus living being — illuminates a host of core
assumptions about management and organizations.
I believe that almost all of us adopt the machine assumption
without ever thinking about it. In so doing, we probably mould
the destiny of individual organizations far more than we imagine.
For example, a machine is owned by someone. We are used
to thinking of companies in exactly that way: they are owned by
owners, usually distinct from the company’s members. But what
does it mean to say that a living being is owned by someone?
Most people in the world would regard the idea that one person
owns another as fundamentally immoral. Is it no less problemat-
ic with regard to a company?
A machine exists for a purpose conceived of by its builders.
 FOREWORD 3

Again, that is the conventional view of a company: its purpose is

to make as much money as possible for its owners. But living
beings have their own purpose. This inherent purpose can never
be completely supplanted by the goals of another, even though a
living being might respond to others’ goals. What happens to the
life energy of a living being when it is unable to pursue its
purpose?
To be effective, a machine must be controllable by its opera-
tors. This, of course, is the overarching raison d’être of manage-
ment — to control the enterprise. But living beings are not
controllable in the ways a machine is. (Anyone doubting this
premise might consider their success in controlling their teenagers.)
They are ‘influenceable’, but only through complex interactive
processes which are just as likely to alter the influencer as the influ-
encee. Are struggles over control not the root of most corporate
politics and game playing?
Going further, seeing a company as a machine implies that
it is created by someone outside. This is precisely the way in
which most people see corporate systems and procedures — as
something created by management and imposed on the organiza-
tion. Seeing a company as a living being implies that it creates its
own processes, just as the human body manufactures its own cells,
which in turn compose its own organs and bodily systems. Is this
not exactly how the informal organization of any large company
comes into being? The networks of relationships and communi-
cation channels essential to anyone doing any job are indeed cre-
ated by the people themselves.
Seeing a company as a machine implies that it is fixed, stat-
ic. It can change only if somebody changes it. Seeing a company
as a living being means that it evolves naturally.
Seeing a company as a machine implies that its only sense of
identity is that given to it by its builders. Seeing a company as a
 THE LIVING COMPANY 4

living being means that it has its own sense of identity, its own
personhood.
Seeing a company as a machine implies that its actions are
actually reactions to goals and decisions made by management.
Seeing a company as a living being means that it has its own goals
and its own capacity for autonomous action.
Seeing a company as a machine implies that it will run down,
unless it is rebuilt by management. Seeing a company as a living
being means that it is capable of regenerating itself, of continuity
as an identifiable entity beyond its present members.
Seeing a company as a machine implies that its members are
employees or, worse, ‘human resources’, humans standing in
reserve, waiting to be used. Seeing a company as a living being leads
to seeing its members as human work communities.
Finally, seeing a company as a machine implies that it learns
only as the sum of the learning of its individual employees. Seeing
a company as a living being means that it can learn as an entity,
just as a theatre group, jazz ensemble or championship sports
team can actually learn as an entity. In this book, Arie argues that
only living beings can learn.
It is hard for me to ponder the above list of characteristics
of machines versus living beings and not feel drawn to the view
that Arie puts forward. Why then, I wonder, have I not come to
this view earlier? Why does it seem so difficult for me actually to
think of a company as a living being? Why does this very simple
idea seem not so very easy to internalize?
Is it that we think life starts and ends with us? Surely, sim-
pler organisms are alive. Why then can’t we regard more complex
organisms, like families or societies or companies, as being alive
as well? Is the tide pool, a teeming community of life, any less
alive than the anemones, mussels or hermit crabs that populate it?
Is it that our mental model of ‘company’ is just so fixed in our
 FOREWORD 5

minds that we cannot suspend it? Or are we simply not willing to

suspend it? If, indeed, we have thought of the companies of which
we have been a part as machines, this implies that we are mechan-
ical elements in the machine. A machine does not have living
parts. For many of us, this has undoubtedly fostered a deep
antipathy towards our organizations. At some level, we deeply
resent being made machine like, in order to fit into the machine.
If there is some element of truth in this, it probably says a lot
about just how important Arie’s simple question actually is.
As Arie points out, the machine metaphor is so powerful
that it shapes the character of most organizations. They have
become more like machines than like living beings because their
members think of them that way.
So, perhaps our first mandate is to shift our thinking. As
Einstein said, ‘Problems cannot be solved at the same level of
awareness that created them.’ As we do this, the host of practical
insights which Arie offers for how a company as a living being
might plan, learn, manage and govern itself will prove invaluable
stepping-stones into what for most of us will be a very different
world.
It might also help to reflect that, as odd as Arie’s view might
at first seem to some of us, it is in fact quite old. Apparently cul-
tures around the world have embraced similar notions for a very
long time. In Swedish, the oldest term for ‘business’ is
näringslivet, literally ‘nourishment for life’. The ancient Chinese
characters for ‘business’, at least 3000 years old, are:
 THE LIVING COMPANY 6

The first of these characters translates as ‘life’ or ‘live’. It can also

be translated as ‘survive’ and ‘birth’. The second translates as
‘meaning’.
As we enter the twenty-first century, it is timely, perhaps
even critical, that we recall what human beings have understood
for a very long time — that working together can indeed be a deep
source of life meaning. Anything less is just a job.
Peter M. Senge
December 1996
Prologue:
The Lifespan of a Company
IN THE WORLD OF INSTITUTIONS, COMMERCIAL CORPORATIONS
are newcomers. Their history comprises only 500 years of
activity in the Western world, a tiny fraction of the timespan of
human civilization. In that time, they have had immense success
as producers of material wealth. They have been the major vehi-
cle for sustaining the exploding world population with goods and
services that make civilized life possible. In the years ahead, as
developing countries expand their standards of living, corpora-
tions will be needed more than ever.
Yet, if you look at them in the light of their potential, most
commercial corporations are dramatic failures — or, at best,
underachievers. They exist at a primitive stage of evolution; they
develop and exploit only a fraction of their capability. For proof,
you need only consider their high mortality rate. The average life
expectancy of a multinational company — Fortune 500 or its
equivalent — is between 40 and 50 years. This figure is based on
most surveys of corporate births and deaths. A full one-third of
the companies listed in the 1970 Fortune 500, for instance, had
vanished by 1983 — acquired, merged or broken to pieces.1
Human beings have learned to survive, on average, for 75 years or
more, but very few companies are that old and flourishing.
There are a few. The Stora company, for example, is a major
paper, pulp and chemical manufacturer; it has had the character of
 THE LIVING COMPANY 8

a publicly owned company from its very early beginnings, more

than 700 years ago, as a copper mine in central Sweden.
Sumitomo Group has its origins in a copper casting shop found-
ed by Riemon Soga in the year 1590. Examples like these are
enough to suggest that the natural average lifespan of a corpora-
tion should be as long as two or three centuries.
I didn’t see these astonishing statistics until I had already
spent more than two decades as a professional manager. It took
another decade for their implications to sink in fully. I worked all
my life for a major Anglo-Dutch multinational, the Royal
Dutch/Shell Group of companies. Born and educated in the
Netherlands, I went to work for Shell directly out of university. I
held jobs ranging from accountant to group planning coordinator
(coordinator is the group’s equivalent of a senior vice president),
working on three continents and in Shell operating companies
whose businesses ranged from refining to marketing to exploration
and from oil to chemicals to metals. As it happens, I am a second-
generation Shell man, because my father worked for the same com-
pany. During our two generations, he and I clocked up 64 working
years. So it cannot be a great surprise that, for a long time, I took it
for granted that most companies (including Royal Dutch/Shell)
simply could not die. They would naturally exist forever.
Well, they don’t. Even the big, solid companies, the pillars
of the society we live in, seem to hold out for not much longer
than an average of 40 years. And that 40-year figure, short though
it seems, represents the life expectancy of companies of a consid-
erable size. These companies have already survived their first 10
years, a period of high corporate ‘infant mortality’. In some coun-
tries, 40 per cent of all newly created companies last less than 10
years. A recent study by Ellen de Rooij of the Stratix Group in
Amsterdam indicates that the average life expectancy of all firms
investigated, regardless of size, in Japan and much of Europe, is
PROLOGUE: THE LIFESPAN OF A COMPANY 9

only 12.5 years.2 I know of no reason to believe that the situation

in the United States is materially better.
The implications of these statistics are depressing. Between
the centuries of age of a Stora or a Sumitomo and the average life-
span — whether 12.5 or 40 years — there exists a gap which repre-
sents the wasted potential in otherwise successful companies. The
damage is not merely a matter of shifts in the Fortune 500 roster;
work lives, communities and economies are all affected, even dev-
astated, by premature corporate deaths. Moreover, there is some-
thing unnatural in the high corporate mortality rate; no living
species, for instance, endures such a large gap between its maxi-
mum life expectancy and its average realization. Moreover, few
other types of institutions — churches, armies or universities —
seem to have the abysmal demographics of the corporate life form.
Why then do so many companies die prematurely? There
are many speculations about the reason, and this area undoubted-
ly needs much more research. However, experience is accumulat-
ing that corporations fail because the prevailing thinking and
language of management are too narrowly based on the prevail-
ing thinking and language of economics. To put it another way:
companies die because their managers focus on the economic
activity of producing goods and services, and they forget that
their organization’s true nature is that of a community of humans.
The legal establishment, business educators and the financial
community all join them in this mistake.

Some companies last hundreds of years

These understandings stemmed from a surprising study which we
conducted in 1983, when I was coordinator of planning for the
Royal Dutch/Shell Group. Royal Dutch/Shell, based in Britain
and the Netherlands, is one of the top three corporations in the
 THE LIVING COMPANY 10

world in size — composed of more than 300 companies in more

than 100 countries around the world. All of these companies are
co-owned by an interlinked pair of holding companies, one
Dutch and one British. The history of the Shell Group dates back
to the 1890s. Its British founders began as sellers of oil for the
lamps of the Far East (Shell was named after the fact that seashells
were used as money in the Far East), while the Dutch founders
imported kerosene from Sumatra. From the moment they
merged, in 1906, Shell’s primary business was the worldwide pro-
duction and marketing of oil and petroleum.
That was true at least until the 1970s. Then, feeling the pres-
sure of the energy crisis, Shell’s managers (along with managers of
other oil companies and firms in other industries) were swept up
in the trend of diversification. We entered into metals, nuclear
power and other businesses that were new to us, with varying
degrees of success. By the early 1980s, serious doubts had sur-
faced in the Shell Group about the wisdom of this diversification.
Yet we weren’t sure we could survive with our core oil and petro-
leum business alone. Reserves of reasonably accessible oil were
projected to last three or four decades before they would be
exhausted. Shell executives cannot avoid discussing whether there
is life after oil. What other businesses might Shell reasonably
enter? How might we prepare for switching to another primary
business? And what effect would that switch have on our compa-
ny as a whole?
In the early 1980s, the planners in my department conduct-
ed some research to see what other companies were doing with
their business portfolios. But Lo van Wachem, then chairman of
the Committee of Managing Directors (the most senior board of
Royal Dutch/Shell managers), pointed out that the companies we
had studied were nowhere near the size of the Shell Group. Size,
when you get to the level of turning over $100 billion per year,
PROLOGUE: THE LIFESPAN OF A COMPANY 11

presents its own unique problems. The examples were also too
recent. Other companies’ diversification moves had not yet stood
the test of time. Some of Shell’s diversification moves, like the
opening of the chemicals business, were already at least 30 years
old, and we still didn’t have consensus within the company about
their value.
Van Wachem added that he would be more interested if the
planners could show him some examples of large companies that
were older than Shell and relatively as important in their industry.
Most importantly, he wanted to know about companies which,
during their history, had successfully weathered some fundamen-
tal change in the world around them — such that they still existed
today with their corporate identity intact.
That was an interesting question. Looking for companies
older than Shell would mean going back to the final quarter of the
nineteenth century — or earlier, into the first years of the
Industrial Revolution. Tens of thousands of companies had exist-
ed in those days, in every corner of the world. But which ones
were still alive today with their corporate identity intact?
Some companies exist only as a name, a brand, an office
building or a memory: remnants of a glorious past. But after some
research and reflection, we began building up a list of companies
that met van Wachem’s criteria. In North America, there were
DuPont, the Hudson’s Bay Company, W.R. Grace and Kodak —
all older than Shell. A handful of Japanese companies traced their
origins to the seventeenth and eighteenth centuries but were still
thriving. They included Mitsui, Sumitomo, and the department
store Daimaru. Mitsubishi and Suzuki were younger; they traced
their origins merely to the nineteenth century, having emerged
from the business opportunities which opened up around the
Meiji Restoration (1868). During that period of fundamental
change in Japan, sparked by Admiral Perry’s first visit of 1853,
 THE LIVING COMPANY 12

some ancient Japanese companies had got into serious difficulties;

but Mitsui, Sumitomo and Daimaru had survived with their cor-
porate identities intact.
In present-day Europe, a sizable number of firms were 200
years old or more. In fact, there were so many such firms in the
UK that they had their own trade association, the Tercentenarians
Club, which only accepts member companies over 300 years old.
However, most of these were family firms which did not meet our
size requirements, many of them still under the control of the
founding family dynasty.
We commissioned a study, conducted by two Shell planners
and two outside business school professors, to examine the ques-
tion of corporate longevity. From the very first moment, we were
startled by the small number of companies which met van
Wachem’s criteria of being large and older than Shell. In the end, we
found only 40 corporations, of which we studied 27 in detail, rely-
ing on published case histories and academic reports. We wanted to
find out whether these companies had something in common
which could explain why they were such successful survivors.
After all our detective work, we found four key factors in
common:

✦ Longlived companies were sensitive to their environment.

Whether they had built their fortunes on knowledge (such as
DuPont’s technological innovations) or on natural resources
(such as the Hudson’s Bay Company’s access to the furs of
Canadian forests), they remained in harmony with the world
around them. As wars, depressions, technologies and political
changes surged and ebbed around them, they always seemed to
excel at keeping their feelers out, tuned to whatever was going on
around them. They did this, it seemed, despite the fact that in the
past there was little data available, let alone the communications
PROLOGUE: THE LIFESPAN OF A COMPANY 13

facilities to give them a global view of the business environment.

They sometimes had to rely for information on packets carried
over vast distances by portage and ship. Moreover, societal con-
siderations were rarely given prominence in the deliberations of
company boards. Yet they managed to react in timely fashion to
the conditions of society around them.

✦ Longlived companies were cohesive, with a strong sense of

identity. No matter how widely diversified they were, their
employees (and even their suppliers, at times) felt they were all
part of one entity. One company, Unilever, saw itself as a fleet of
ships, each ship independent, yet the whole fleet stronger than the
sum of its parts. This sense of belonging to an organization and
being able to identify with its achievements can easily be dis-
missed as a ‘soft’ or abstract feature of change. But case histories
repeatedly showed that strong employee links were essential for
survival amid change. This cohesion around the idea of ‘commu-
nity’ meant that managers were typically chosen for advancement
from within; they succeeded through the generational flow of
members and considered themselves stewards of the longstanding
enterprise. Each management generation was only a link in a long
chain. Except during conditions of crisis, the management’s top
priority and concern was the health of the institution as a whole.

✦ Longlived companies were tolerant. At first, when we wrote our

Shell report, we called this point ‘decentralization’. Longlived
companies, as we pointed out, generally avoided exercising any
centralized control over attempts to diversify the company. Later,
when I considered our research again, I realized that seventeenth-,
eighteenth- and nineteenth-century managers would never have
used the word decentralized; it was a twentieth-century inven-
tion. In what terms, then, would they have thought about their
 THE LIVING COMPANY 14

own company policies? As I studied the histories, I kept return-

ing to the idea of ‘tolerance’. These companies were particularly
tolerant of activities in the margin: outliers, experiments and
eccentricities within the boundaries of the cohesive firm, which
kept stretching their understanding of possibilities.

✦ Longlived companies were conservative in financing. They

were frugal and did not risk their capital gratuitously. They
understood the meaning of money in an old-fashioned way; they
knew the usefulness of having spare cash in the kitty. Having
money in hand gave them flexibility and independence of action.
They could pursue options that their competitors could not.
They could grasp opportunities without first having to convince
third-party financiers of their attractiveness.
It did not take us long to notice the factors which did not appear
on the list. The ability to return investment to shareholders
seemed to have nothing to do with longevity. The profitability of
a company was a symptom of corporate health, but not a predic-
tor or determinant of corporate health. Certainly, a manager in a
longlived company needed all the accounting figures that he or
she could lay hands on. But those companies seemed to recognize
that figures, even when accurate, describe the past. They do not
indicate the underlying conditions that will lead to deteriorating
health in the future. The financial reports at General Motors,
Philips Electronics and IBM during the mid-1970s gave no clue of
the trouble that lay in store for those companies within a decade.
Once the problems cropped up on the balance sheet, it was too
late to prevent the trouble.
Nor did longevity seem to have anything to do with a com-
pany’s material assets, its particular industry or product line or its
country of origin. Indeed, the 40-50-year life expectancy seems
PROLOGUE: THE LIFESPAN OF A COMPANY 15

to be equally valid in countries as wide apart as the US, Japan and

in Europe, and in industries ranging from manufacturing to
retailing to financial services to agriculture to energy.
At the time, we chose not to make the Shell study available
to the general public and it remains unpublished. The reasons had
to do with the lack of scientific reliability for our conclusions.
Our sample of 30 companies was too small. Our documentation
was not always complete. And, as the management thinker
Russell Ackoff once pointed out to me, our four key factors rep-
resented a statistical correlation; our results should therefore be
treated with suspicion. Finally, as the authors of the study noted
in their introduction: ‘Analysis, so far completed, raises consider-
able doubts about whether it is realistic to expect business histo-
ry to give much guidance for business futures, given the extent of
business environmental changes which have occurred during the
present century.’3
Nonetheless, our conclusions have recently received cor-
roboration from a source with a great deal of academic
respectability. Between 1988 and 1994, Stanford University pro-
fessors James Collins and Jerry Porras asked 700 chief executives
of US companies — large and small, private and public, industrial
and service — to name the firms they most admired. From the
responses, they culled a list of 18 ‘visionary’ companies. They
didn’t set out to find longlived companies but, as it happened,
most of the firms that the CEOs chose had existed for 60 years or
longer. (The only exceptions were Sony and Wal-Mart.) Collins
and Porras paired these companies up with key competitors
(Ford with General Motors, Procter & Gamble with Colgate,
Motorola with Zenith) and began to look at the differences. The
visionary companies put a lower priority on maximizing share-
holder wealth or profits. Just as we had discovered, Collins and
Porras also found that their most admired companies combined
 THE LIVING COMPANY 16

sensitivity to their environment with a strong sense of identity:

‘Visionary companies display a powerful drive for progress that
enables them to change and adapt without compromising their
cherished core ideals.’4
At Shell we had not conducted a study of similar diligence.
Nonetheless, the Shell study remained uppermost in my mind for
years. In our unscientific way, we had found four characteristics
that seemed, when put together, to give us a description of a high-
ly successful type of company — a company that could survive for
very long periods in an ever-changing world, because its man-
agers were good at the management of change.

Defining the living company

Over time, the same four factors that we developed in our study
of longlived companies at Shell have continued to resonate in my
mind. Gradually, they began to change my thinking about the real
nature of companies — and about what this means for the way that
we, managers at all levels, run those companies. I now see these
four components in this way:

✦ Sensitivity to the environment represents a company’s ability

to learn and adapt.
✦ Cohesion and identity, it is now clear, are aspects of a compa-
ny’s innate ability to build a community and a persona for itself.
✦ Tolerance and its corollary, decentralization, are both symp-
toms of a company’s awareness of ecology: its ability to build con-
structive relationships with other entities, within and outside
itself.
✦ And I now think of conservative financing as one element in a
very critical corporate attribute: the ability to govern its own
growth and evolution effectively.
PROLOGUE: THE LIFESPAN OF A COMPANY 17

Moreover, the question remains: why would these same charac-

teristics occur again and again in companies that had managed to
outlive others? I am convinced that the four characteristics of a
longlived company are not answers. They represent the start of a
fundamental enquiry into the nature and success of commercial
organizations and their role in the human community.
Not coincidentally, these four basic components have also
provided the framework for this book. Put together, they give
clues to the real nature of companies and they form a set of orga-
nizing principles of managerial behaviour — critical aspects of the
work of any manager who wants his or her company to survive
and thrive for the long term.
The Shell study also reinforced a concept which I have devel-
oped since my student days: considering and talking about a com-
pany as a living entity. I am not alone in this. Many people
naturally think and speak about a company as if they were speak-
ing about an organic, living creature with a mind and character of
its own. This common use of the language is not surprising. All
companies exhibit the behaviour and certain characteristics of liv-
ing entities. All companies learn. All companies, whether explicitly
or not, have an identity that determines their coherence. All com-
panies build relationships with other entities, and all companies
grow and develop until they die. To manage a ‘living company’ is
to manage with more or less consistent, more or less explicit appre-
ciation for these facts of corporate life, instead of ignoring them.
It probably doesn’t matter very much whether a company
is actually alive in a strict biological sense, or whether the ‘living
company’ is simply a useful metaphor. As we will see throughout
this book, to regard a company as a living entity is a first step
towards increasing its life expectancy.
This book is about the idea of the living company, its philo-
sophical underpinnings, its application in practice, and the power
 THE LIVING COMPANY 18

and capability that seem to come from adopting it.

The idea of a living company is not just a semantic or acad-
emic issue. It has enormous practical, day-to-day implications for
managers. It means that, in a world which changes massively,
many times, during the course of your career, you need to involve
people in the continued development of the company. The
amount that people care, trust and engage themselves at work has
not only a direct effect on the bottom line, but the most direct
effect, of any factor, on your company’s expected lifespan. The
fact that many managers ignore this imperative is one of the great
tragedies of our times.
What then does managing a living company mean on a day-
to-day basis? The path to the answer starts with another question,
that of corporate purpose: what are corporations for?
Financial analysts, shareholders and many executives tell us
that corporations exist primarily to provide a financial return.
Some economists offer a rather broader sense of purpose.
Companies, they say, exist to provide products and services, and
therefore to make human life more comfortable and desirable.
‘Customer orientation’ and other management fashions have trans-
lated this imperative into the idea that corporations exist to serve
customers. Politicians, meanwhile, seem to believe that corpora-
tions exist to provide for the public good: to create jobs and ensure
a stable economic platform for all the ‘stakeholders’ of society.
But from the point of view of the organization itself — the
point of view which allows organizations to survive and thrive —
all of these purposes are secondary.
Like all organisms, the living company exists primarily for its
own survival and improvement: to fulfil its potential and to become
as great as it can be. It does not exist solely to provide customers
with goods, or to return investment to shareholders; any more than
you, the reader, exist solely for the sake of your job or your career.
PROLOGUE: THE LIFESPAN OF A COMPANY 19

After all, you too are a living entity. You exist to survive and thrive;
working at your job is a means to that end. Similarly, returning
investment to shareholders and serving customers are means to a
similar end for IBM, Royal Dutch/Shell, Exxon, Procter &
Gamble, General Motors and every other company.
If the real purpose of a living company is to survive and
thrive in the long run, then the priorities in managing such a com-
pany are very different from the values set forth in most of the
modern academic business literature. Such a purpose also contra-
dicts the views held by many managers and shareholders. To be
sure, many management fashions resonate with the idea of a learn-
ing company — for example, the concepts of the ‘learning organi-
zation’ and ‘knowledge as a strategic asset’. But there are serious
doubts that even the most enthusiastic managers and shareholders
have fully explored the ramifications of those concepts.
The result: in today’s increasingly volatile business environ-
ment, without the priorities of the living company, most man-
agers will find that their companies do not have the right habits
to accomplish what they hope to achieve. On the other hand,
exploring the ramifications of managing an entity which is alive,
with the intent of handing it over to your successors in better
health than when you received it, is deeply gratifying. The own-
ers of the firms in London’s Tercentenarian Club and the man-
agers of the Shell study survivors are usually exponents of a
deeply felt corporate pride.
Part I
Learning
1
The Shift from Capitalism to a
Knowledge Society
IF, AS A MANAGER, YOU TAKE THE FINDINGS OF THE ROYAL
Dutch/Shell study or of Collins and Porras seriously, then you
are faced with a seemingly insoluble dilemma. In the language of
economics, companies are expected to operate with profits as
their primary goal. On the other hand, as suggested by the stud-
ies, adopting this goal could well conflict with companies’
longevity and life expectancy. Faced with this dilemma, managers
often throw up their hands and choose the path of highest imme-
diate return on investment, feeling that the survival of the com-
pany — and their jobs — will depend on following this path.
Indeed, many managers at Shell and other companies have asked
me why I would expect them to manage for the long run, with the
risk of being dead in the short term.
My answer is that the dichotomy between profits and
longevity is false. It is no longer necessary to choose between the
two. Corporate success and longevity are fundamentally inter-
woven, in a way that, nowadays, is qualitatively different from
the relationship between success and longevity in the economic
environment of five decades ago. The twin policies of managing
 LEARNING 22

for profit and maximizing shareholder value, at the expense of all

other goals, are vestigial management traditions. They no longer
reflect the imperatives of the world we live in today. They are
suboptimal, even destructive — not just to the rest of society, but
to the companies which adopt them.
In short, the conventional management wisdom remains
focused on the idea of conserving and maximizing capital. But dur-
ing the past 50 years, the world of business has shifted from one
dominated by capital to one dominated by knowledge. This shift
explains the interest in organizational learning that has emerged in
the last few years. Managers recognize that, unless their companies
can accelerate the rate at which they learn, their primary asset will
stagnate and their competitors will outpace them.
Basic economic theory tells us that there have always been
three key sources of wealth: land and natural resources, capital (the
accumulation and reinvestment of possessions) and labour. The
combination of those three creates the products and services that
society needs for its material well-being. During most of human
history, the critical factor of economic success was land: those who
could dominate and possess the land were guaranteed the control-
ling role in creating wealth. Thus owners of land, at least in Western
society, were rich and people who had no land were poor.
Then, as historians such as Fernand Braudel and Henri
Pirenne have described in graphic detail, a dramatic shift took
place between the late Middle Ages and the beginning of the
twentieth century — a shift from land to capital as the primary fac-
tor in generating wealth.1 The addition of more capital to the
process of creating material wealth led to considerable increases
in the effectiveness and efficiency of technological and commercial
activity. Ships became bigger, voyages became longer, machines
became more capable. By the end of the Middle Ages, much more
money was available for such ends, at least in Western Europe.
 THE SHIFT FROM CAPITALISM TO A KNOWLEDGE SOCIETY 23

These savings were converted into the assets of the growing com-
mercial ventures, which evolved into mining companies, shipping
and trade enterprises, the first textile ateliers, and eventually mod-
ern corporations. The modern company, in short, developed when
capital became available for the wealth-creating processes of the
medieval trader.
In the age of capital, wealth passed from those who con-
trolled the land to those who controlled access to capital. The rich
were no longer the landowners; they were the owners of capital.
The ability to finance endeavours became the scarcest commodi-
ty of production.
Moreover, with the breakup of the old craft guilds and their
evolution into companies, the owners of capital were able to con-
trol the human production factor. In the language of economic
theory, capital was worth far more, and was far more scarce, than
labour. Labour moved from being part of the everyday life of
human beings, an integral aspect of the human community, to
being a commodity, offered for sale on the market. As Braudel
put it in his book The Wheels of Commerce:

All [a worker] had to offer was his arm or hand, his ‘labor’ in
other words. And of course his intelligence or skill. The phe-
nomenon can be seen with unusual clarity in the case of the
miners of Central Europe. Having long been independent arti-
sans, working in small groups, they were obliged in the fif-
teenth and sixteenth century to put themselves under the
control of the merchants who alone could provide the consid-
erable investment required for equipment to mine deep below
the surface. And they became wage-earners.2

Over the course of the following centuries, a new element

emerged in management thinking. If a company got into trouble,
 LEARNING 24

jobs were cut first, because the capital assets (and the investors’
goodwill) were far more scarce and valuable and managers saw
the optimization of capital as their first priority. During the Great
Depression of the 1930s, for example, it had been considered
good banking practice to liquidate and destroy client institutions,
and all the jobs associated with them, if this would help recover
even a scrap of invested capital. (This may have seemed harsh, but
it was necessary. Capital-supplying institutions were a lot less
robust than they are nowadays, and they were fighting for their
own survival.3) In their attitude towards capital, companies were
very different from their older siblings, the church and the mili-
tary. Even in the grim retreat at Dunkirk during the Second
World War, the British expeditionary forces scuttled their war
machines to save their soldiers. Capital assets were not as impor-
tant as people.
Sometime over the course of the twentieth century, how-
ever, the Western nations moved out of the age of capital and into
the age of knowledge. Few managers recognized it at the time, but
capital was losing its scarcity. After the Second World War, an
enormous capital accumulation began. Individuals, banks and
companies became much more resilient. Technology also began
to change, thanks to telecommunications, television, computers
and commercial air travel, with the effect of making capital far
more fungible and resilient, easier to move around — and conse-
quently less scarce.
With capital easily available, the critical production factor
shifted to people. But it did not shift to simple labour. Instead,
knowledge displaced capital as the scarce production factor — the
key to corporate success. Those who had knowledge and knew
how to apply it would henceforth be the wealthiest members of
society: the technological specialists, investment bankers, creative
artists and facilitators of new understanding. This was not mere-
 THE SHIFT FROM CAPITALISM TO A KNOWLEDGE SOCIETY 25

ly a function of the need for people to supply technical skills,

under the direction of their bosses. The growing complexity of
work created a need for people to be a source of inventiveness,
and to become distributors and evaluators of inventions and
knowledge, through the whole work community. Judgement, on
behalf of the company as a whole, could no longer be the exclu-
sive prerogative of a few people at the top.4
Had we known where to look, even back in the 1950s we
could have seen the shift of value from capital to knowledge. It
was becoming visible in the rise of asset-poor, brain-rich compa-
nies and partnerships — international auditor firms, management
consultancies, and advertising and media businesses. Within a
decade or two, even these would be eclipsed by the explosively
growing software and information technology companies. All of
these brain-rich companies cannot be managed in the old asset-
oriented style. Their managers have had to shift their priorities,
from running companies to optimize capital to running companies
to optimize people. In these companies, people are the carriers of
knowledge and therefore the source of competitive advantage.

Economic success versus learning

In the early 1950s, during the beginning of the shift from capital
to knowledge, I was a business school student at Erasmus
University in Rotterdam. I can distinctly remember the definition
of business that they taught us there. This ‘economic company’,
as I now call it, was a very comforting entity. It was rational, cal-
culable and controllable:

Companies produce goods and services, for which other peo-

ple are prepared to pay a price, by trying to find the optimum
combination of the three production factors — labour, capital
 LEARNING 26

and land. These three are substitutable. Labour can be replaced

by capital, for example. The optimum combination of the pro-
duction factors is the one at which the company is producing
the goods and services at minimum costs to be sold at maxi-
mum price for the maximization of profits.5

This definition has great clarity. Many people will recognize it as

the way they have been told to think about companies. The defi-
nition is also clear in distinguishing between successful and
unsuccessful companies. Success equals maximum profits at the
optimum combination of the production factors. The definition
makes it not only easy to measure success, but quick. You do not
have to wait 50 years to find out whether you have been success-
ful. At the end of each quarter of each year, you can measure suc-
cess from the quarterly results. And your company might win the
corporate equivalent of an Oscar when Fortune magazine and the
Financial Times publish their lists of the 100 or 500 Chosen Ones.
To have the highest return on capital employed, the highest
turnover, the highest market capitalization — those are the criteria
of success under the economic definition.
We students of the 1950s, fledgling examples of homo eco-
nomicus, learned this definition enthusiastically. We were not
aware that, already, it was an inaccurate description of true cor-
porate success. I now realize that I had my first inkling of this
inaccuracy after leaving Erasmus, when I entered my first place of
work: the Shell refinery near Rotterdam. Soon after I walked in
the door I felt a slight level of discomfort. The theories back at
business school had mentioned labour, but there had been no talk
of people. Yet the real world, the refinery, seemed to be full of
them. And because the workplace was full of people, it looked
suspiciously as if companies were not always rational, calculable
and controllable.
 THE SHIFT FROM CAPITALISM TO A KNOWLEDGE SOCIETY 27

Today, I recognize that the economic company is an

abstraction with little to do with the reality of corporate life. Not
only does labour not equate to people, but the emphasis on prof-
its and on the maximization of shareholder value ignores the two
most significant forces acting on companies today: the shift to
knowledge as the critical production factor and the changing
world environment.
Companies could act according to the economic definition
of success when managers felt that they were in control of their
world. But rare is the manager who feels in control of today’s tur-
bulent environment. Therefore, to cope with a changing world,
any entity must develop the capability to shift and change, to
develop new skills and attitudes: in short, the capability to learn.
As we will see in this book, the essence of learning is the ability to
manage change by changing yourself — as much for people when
they grow up as for companies when they live through turmoil.
The pioneering learning theorist Jean Piaget called this form of
change ‘learning through accommodation’.6 Its essence, he said,
was to change one’s internal structure to remain in harmony with
a changed environment.
This gives us an entirely different imperative for corporate
success: a successful company is one which can learn effectively.

When success is based on learning

Under this definition, people are much more critical to a compa-
ny. Knowledge, after all, is carried in the heads of people. This is
not to say that capital assets are unimportant. They are vital.
Without capital, mankind would never have reached the econom-
ic output we have attained. But the current way of thinking about
companies frames the issue as an either/or question: if you are
promoting the interests of people, then you are necessarily short-
 LEARNING 28

changing the interests of capital. From a learning orientation, the

nurturing of people and the nurturing of capital reinforce each
other.
During the 1980s, I heard a debate occur with great regu-
larity inside the Shell Group. How could we best improve our
return on investment? The usual response, at senior executive lev-
els, was: ‘Overheads must come down.’ In Shell, as in all large
corporations, 80 per cent of the overhead costs relate to the
expenses of people. The outcome of these discussions was there-
fore inevitable: a number of people would have to be let go. The
rest of our managerial attention then went into figuring out how
this reduction would be achieved. Which functions or regions
would be cut? Would we cut through staff transfers, voluntary
redundancy or enforced dismissal?
During one such debate, I remember the lone voice of Ian
McCutcheon, the group controller, asking: ‘But what about the
loss of human potential, experience and loyalty?’ I also remember
how this remark disappeared in the wind. No serious attempt was
made to consider the potential for increasing future proceeds;
everyone focused attention on the immediate prospect of reduc-
ing costs.
The same debate takes place in many companies today, and
in many of these companies the concerns about human capital are
lost in much the same manner. That is why we need a new way of
thinking about the measurement of success in our companies. By
outsiders, we are judged and measured in economic terms: return
on investment and capital assets. But within the company, our
success depends on our skill with human beings: building and
developing the consistent knowledge base of our enterprise.
Personally, I’m inclined to believe that the sharp difference
between these two definitions — the economic company definition
and the learning company definition — lies at the core of the crisis
 THE SHIFT FROM CAPITALISM TO A KNOWLEDGE SOCIETY 29

which managers face today. The tension between them is almost

certainly one of the key reasons behind the surprisingly low aver-
age life expectancy of companies in the northern hemisphere.
On one hand, all the experts, academics and measures of
managerial success line up on the side of the economic definition.
They suggest that the company’s heart, the core of its nature, is the
economic activity it pursues to stay alive. On the other hand is the
evidence of managers’ own eyes, ears and feelings: that the core of
their company’s nature, its heart, is its existence as a continuous
work community — in short, as a living, learning company.
To make the case for a non-economic company credible, we
thus need to look more closely at learning in organizations. What
links can we draw between the idea of a ‘living company’, acting in
its own self-interest as an entity unto itself, and the idea of a ‘learn-
ing company’, with sensitivity to its environment?
2
The Memory of the Future
LEARNING BEGINS WITH PERCEPTION. NEITHER AN INDIVIDUAL
nor a company will even begin to learn without having seen
something of interest in the environment. That is why surviving
and thriving in a volatile world require, first of all, management
which is sensitive to its company’s environment. At least a few of
the company’s leaders should be attentive and responsive to the
world in which they live, even to the extent of playing an active
role in that outside world. Navel gazers are necessary in every
company, but they see little of the forces which will affect the
future of that company.
By contrast, an open and extroverted management will per-
ceive whatever is happening outside much earlier. Only after see-
ing that something is about to change (or has already begun to
change) outside the company will management be ready to deal
with the effects of that change. Many of these effects lie in the
future and are uncertain. In the desire to ‘know’ and reduce that
uncertainty, most managers spend far too much time on a rela-
tively useless question: what will happen to us?
But managers who perceive change early should spend
more time on a far more useful question: what will we do if such-
and-such happens? Only this latter question can lead managers to
 THE MEMORY OF THE FUTURE 31

make changes inside the company which will allow it to survive

and thrive in the new world. Indeed, as experience shows, funda-
mental and painful change may be necessary, possibly even
including the abolition of a company’s core business.
In fact, at Royal Dutch/Shell in 1983, our study of longlived
large companies provided many examples of the fundamental
nature of the internal changes needed for companies to survive in a
turbulent environment. The study was initially triggered by the
ever-present disquiet in any natural resource company: what will
happen when the oil runs out? To our relief, we found that no com-
pany we studied, in all of our literature research, had ever failed
because its key natural resource was depleted. Yet many companies
had switched away from their original natural resource base or
their original business. DuPont is a classic, well-known example.
Its business portfolio had moved, over time, from gunpowder to
chemicals via its majority shareholding in General Motors.
After the Shell study team reported this sort of finding, I
found myself asking another question: had any of our large, long-
standing companies changed in fundamental ways — not because
they were forced to, but because they anticipated the need for
change?
Indeed they had. Most of the longlived companies, accord-
ing to the team, had anticipated the need for change at least once
during their lives. More often than not, this opportunity was
born of a crisis, amid a new threat from the business environ-
ment — a new competitor, a trade restriction, a shift in the market-
place or a rival technology. The longlived companies seemed to
have an innate ability to exploit these crises and turn them into
new business. Frequently, someone within the enterprise would
identify the crisis ahead of time, but not as a crisis; it was a new
opportunity, an alternative avenue for company growth and
profitability. Here again, DuPont was an example. Its leaders,
 LEARNING 32

mostly family members, had astutely navigated the technological,

political and social cross-currents of nineteenth- and twentieth-
century America. The company took an early lead with dynamite
production in America, and later with cellophane packaging and
plastics research. Some members of the du Pont family, which
owned and managed the company, had become prominent politi-
cians — including US Congressman and Governor Pierre S. du
Pont IV. Ethel du Pont had married Theodore Roosevelt, Jr. And
there was evidence that this sensitivity had permitted the family
decision makers to move rapidly from one business to another,
consolidating their position in General Motors and the chemical
industries in time for the industrial expansion of the Second
World War.1
As we put it in our report:

Identifying the opportunity or the threat was one matter;

stimulating the change necessary to take advantage of the
opportunity was another. There is a considerable difference
between companies that stared blindly at threat and opportu-
nity and those that reacted and changed.2

What gave the companies the ability to accomplish this? We will

return to that question throughout this book, because it depend-
ed on all four distinctions of a living company: its adaptiveness to
the outside world (learning), its character and identity (persona),
its relationships with people and institutions inside and around
itself (ecology) and the way it developed over time (evolution).
We could see that the most accessible of these capabilities, and the
one which often came first, was learning. The longlived compa-
nies were sensitive to their community and their environment.
This sensitivity was not soft or driven by social responsibility. It
was driven by the living company’s self-interest.
 THE MEMORY OF THE FUTURE 33

Behind all displays of sensitivity to the community [we wrote

in our report] there generally lay a hard headed approach and
a recognition that ... alertness and responsiveness ... helped
create the climate in which business growth could take place.3

Consider, for example, the case of the Swedish company Stora. If

you feel overwhelmed by the turbulence in your business envi-
ronment today, then think of the shifting forces with which Stora
had to cope. The first written mention of the company dates from
1288. In those days it was a copper mine based in Dalecarlia, a
province of central Sweden.
When it was a mere 270 years old, during the fifteenth cen-
tury, the company had to fight the king of Sweden to maintain its
independence and identity. Kings throughout Europe, enmeshed
in the struggle to establish centralized nation-states, were grasp-
ing for every penny they could lay their hands on, and their
demands threatened the existence of enterprises like Stora. Thus
Stora took on a political role within Sweden, drawing not just on
its leaders’ financial resources, but also on significant support
from peasant workers in its home base, the province of Dalecarlia.
Ultimately, the master miners of Stora found an appropriate
answer to external turbulence in the manner in which they orga-
nized themselves. As one historian wrote:

A Guild was established at the Mountain ... adopting an inde-

pendent and militaristic profile. For the members, loyalty to
the Guild superseded the law of the land, and the word of the
Master of the Guild weighed heavier than that of a judge.4

During that period of unrest, it would have been catastrophic for

the company to concentrate on its business in an introverted fash-
ion, oblivious to politics. Instead, the company reshaped its goals
 LEARNING 34

and methods to match the demands of the world outside. It did

the same thing again and again, throughout the centuries, from
the Middle Ages through the Reformation, into the wars of the
1600s, the Industrial Revolution, and two world wars in the
twentieth century. To appreciate the difficulties of adaptation,
consider how little data was available to the Swedish managers of
Stora. Instead of telephones, aeroplanes and electronic networks,
they had to depend on runners, horsemen and ships to carry mes-
sages. They barely had the facilities for a global view of their busi-
ness, let alone a view of the global business environment. Nor,
apparently, did their boards have the time to spend deliberating
the needs and demands of society. Yet timely reaction to the con-
ditions in society was necessary for the survival of the company
and sometimes even of its individual members.
Over the next several centuries, while it coped with shifting
social and political forces, the company continually shifted its
business, moving from copper to forest exploitation, to iron
smelting to hydro power and eventually to paper, wood pulp and
chemicals. Its production technologies also shifted — from steam
to internal combustion, then to electricity and ultimately to the
microchip.
Each one of these changes, in hindsight, seems herculean,
but for the people running enterprises in those days they may
well have been gradual and almost imperceptible at the outset.
Certainly, some of these changes must have involved a crisis at the
root of the company, but Stora — and every other successful com-
pany we studied — managed to effect its dramatic changes without
sacrificing its corporate identity or corporate life in the process.
This can mean only one thing: these companies reacted early
rather than later, by foresight rather than by catchup.
 THE MEMORY OF THE FUTURE 35

Changing to match the outside world

Today, businesspeople ignore public attitudes on such issues as
national sovereignty, colonialism and imperialism, pollution, con-
servation, exploitation, the decline of the middle class and even
free trade at their peril. Social changes — such as the changing
position of women in society, the growth in leisure, shifts in
transport and the evolution of consumer taste — continually cre-
ate new employment opportunities and new markets, while old
markets falter. Economic indicators rise and fall wildly, including
international currency rates, inflation rates, interest rates and
product lifecycles (led by the electronics industry). Shareholder
attitudes change from docile to demanding and sometimes back
again. With the fall of the Berlin Wall, we have seen political
changes which have undone the hegemony of 70 years of com-
munism in some countries. All of these attitudes are key aspects
of a company’s business environment.
When I say environment, I don’t use the word as an ecolo-
gist might, to refer to natural surroundings. Rather, I use it to
mean the sum total of all forces that affect a company’s actions. In
the last 20 years, that business environment, within which all
companies must operate, has shown oscillations of increasing fre-
quency and amplitude. These, in turn, reorient the corporate
sense of purpose. In the heat of restructuring and reengineering,
it’s often easy to lose sight of the purpose of the change: to meet
the changing pressures from the outside world.
There are times when a company’s knowhow, product
range and labour relations are in harmony with the world around
it. The situations are familiar; the company is well organized,
trained and prepared. The managers do not need to develop and
implement new concepts. During those times, the essence of
 LEARNING 36

management is to allocate resources to promote growth and

development. This is a very gratifying type of work. It means
channelling capital and human talent to those parts of the organi-
zation which are best placed to benefit from the converging har-
monious environment — and these parts of the organization
return the compliment by becoming larger, better established and
more powerful within the firm.
But then comes the inevitable moment, just as the company
has neatly organized itself to cope with the previous situation,
when the current business environment diverges. It becomes
disharmonious with the way management had structured the
company — particularly with the large, well-established, powerful
components that benefited from the previous structure. If this
disharmony is of a sufficiently fundamental nature, then funda-
mental changes are required in the company as well.
Continuous, fundamental changes in the external world — a
turbulent business environment — require continuous manage-
ment for change in the company. This means making continuous
fundamental changes in the internal structures of the company.
For many psychologists, this principle represents one important
aspect of learning.
This imperative also has important consequences for the
way we run our companies. The company must be able, when
necessary, to alter its marketing, its product range, where and
how it does its manufacturing, and its organizational form — to
stay in harmony with the surrounding world. A fundamental
revision in finance regulations can lead a bank to consider new
markets and new products, dramatically stretching its existing
capabilities. An increase in oil prices (which an oil marketing
company might treat as routine) can force an airline into a funda-
mental revision of its costs, its price structure, its flight schedules
or the composition of its fleet (to more fuel-efficient aircraft).
 THE MEMORY OF THE FUTURE 37

Once these new solutions have been implemented, whether they

involve a new technology, marketing policy, project portfolio or
service schedule, the company is no longer the same. It has moved
into a new phase of its life. This is the essence of learning.
To accomplish this type of learning, the company must see
clearly what is happening in its environment. Once again, learn-
ing begins with perception. How else can managers know when
significant change is necessary, or how to act effectively to achieve
a new sort of harmony? Yet corporate managers, enmeshed in the
details of their change efforts, often think about their outside
pressures in only the vaguest terms. They do not see; they do not
develop a careful sensitivity to the signals of pressures outside the
firm and how those pressures are changing.
Why is it so difficult for managers to maintain their sensi-
tivity? Why do companies fail to see the signals of change ahead
of time? In my last assignment in the Shell Group, as coordinator
of corporate planning, it was important to find an answer to that
question. If companies could see early and manage internal
change by foresight, a great deal of capital destruction and social
misery would be prevented — not only in our company, but in any
company. Why then are so many companies seemingly so blind
and deaf to what is happening around them?
Over time, five different answers to that question emerged.
The first two or three of them have probably occurred to you;
they represent the unspoken, but prevalent, myths about why
managers fail to perceive effectively.

Theory 1: Managers are stupid

Business commentators and academics, with the benefit of hind-
sight, imply or suggest that managers are idiots — either blind,
deaf or plain stupid. Otherwise, why did the great US railway
 LEARNING 38

companies not see the highways for motorcars being built along-
side their tracks in the beginning of this century? Why did the
makers of Western consumer electronics fail to foresee the
Japanese and Korean competitors who would engulf them?
Clearly, businesspeople are not intellectually equipped to cope
with the changing nature of their environment.
I have never liked this explanation. The great majority of
people I meet in business circles are neither deaf, blind nor stupid.
In any case, the problem managers face is not acting intelligently
in isolation, but tapping all of the company’s intelligence to fore-
see problems together.

Theory 2: We can only see when a crisis opens our eyes

I asked a number of psychologists why managers fail to exercise
foresight. They explained that there is a human resistance to
change — a resistance which is basically good, for both the individ-
ual and society. One should not change for change’s sake. However,
said these psychologists, in effect, when change is demanded for
survival, this resistance must be overcome and the only way for this
to happen is through pain — deep, prolonged pain!
The corporate equivalent of pain is a crisis. In the heat of a
lengthy crisis, according to this theory, people in the organization
will feel the pain and be convinced that something must be done.
Indeed, you will often hear managers say, ‘What this company
needs is a nice little crisis. Then we’ll be able to get some change
around here.’
There is no denying that many a fundamental change has
had a crisis at its roots. Whenever I talk to managers of a compa-
ny which went through structural adaptation, they always
remember clearly how painful a time they had in the period pre-
 THE MEMORY OF THE FUTURE 39

ceding the changes. And this is not only true for companies.
Think of OPEC in 1986, many trade unions during the last
decade, and the former Soviet Union as it struggles through its
current transition. In each case, the world around them changed.
Warnings were ignored until, finally, the disharmony began to
show up (for companies) in the quarterly results (when it is
almost too late anyway). Even then, the old objectives were
doggedly pursued. Belts were tightened. Jobs were lost. People
struggled under the stress. Ultimately, the institution’s survival
was at stake. And finally, grudgingly, people began to adopt an
orientation of learning.
Despite the stress involved, many managers enjoy a crisis.
At last, it is possible to do things. Because time is of the essence,
at last the company can move without consultations and lengthy
deliberations. Autocratic, heroic management thrives. Decisions
must be made quickly. Power is centralized and concentrated —
pulled inwards and upwards into a few heads: the company ‘goes
for broke’.
But does that mean that crises are the only avenue for learn-
ing? Or that crises necessarily produce learning? One can think of
a great many crises in which little institutional adaptation took
place at all: companies wilt under the pressure of a hostile
takeover, a fierce new competitor or an unexpected lawsuit. These
crises tend to follow the same generic pattern:

✦ At some point, as prospects worsen, the damage or danger

becomes evident and a consensus, grudging or not, develops
about the inevitability of change.
✦ When that happens, there is little time left.
✦ Because there is little time, few options remain open. They are
not necessarily the best options; they are limited to those which
require little time to implement.
 LEARNING 40

✦ Almost by definition, these tend to be the tough options, dev-

astating to morale and difficult to pull off with corporate identi-
ty intact: improve cash flow drastically, cut costs, cut capital
expenditure, cut staff.
✦ The crisis is a self-reinforcing cycle. The more deeply you
become enmeshed, the more options you must forgo and the
more you run out of time — which cuts your possibilities further
and enmeshes you more deeply in the crisis. To act by foresight
would surely be superior.
But can fundamental change be brought about by foresight? In
practice, this can happen only if the company’s managers can see
the signals for change in time — before the situation has deterio-
rated to the point where the company is losing options. In short,
to act with foresight, the company must act on signals, rather than
on pain.
In the end, the psychologist’s view — that the only way to
obtain a fundamental change is by way of a crisis — is a pessimistic
one which I have difficulty accepting. It means that, faced with a
disharmonious environment, there is nothing that a manager can
do on his or her own. Events follow their inescapable road to dis-
aster. Business life is a gamble or, rather, a Greek tragedy. We suf-
fer and cope. There is no hope for improvement.

Theory 3: We can only see what we have

already experienced
With the emergence of cognitive psychology and the study of
mental maps, some psychologists began to claim that people can
only ‘see’ what they have experienced before — at least in some
respect. To receive a signal from the outside world, it must match
 THE MEMORY OF THE FUTURE 41

some matrix already in the mind, placed there by previous events.

Consider, for example, the story of the tribal chief who was
brought to Singapore by a group of British explorers at the begin-
ning of the twentieth century. The explorers had found him deep
in the high mountains of the Malaysian peninsula, in an isolated
valley. His tribe was literally still in the Stone Age. Its people had
not even invented the wheel. Nonetheless, the chief was highly
intelligent and a delightful man to talk to (once they made them-
selves understood). He seemed to have a deep, meaningful per-
ception of his own world.
So, as an experiment, they decided to convey him to
Singapore. It was at the time already a sophisticated seaport, with
multistorey buildings and a harbour with big ships.
Economically, it had a market economy with traders and profes-
sional specializations. Socially, it had many more layers than the
society from which the tribal chief came. They marched the chief
through this world for 24 hours, submitting him to thousands of
signals of potential change for his own society. Then they brought
him back to his mountain valley and started to debrief him.
Of all the wonders he had seen, only one seemed important
to him: he had seen a man carrying more bananas than he had ever
thought one man could carry. What the mind has not experienced
before, it cannot see. The tribal chief could not relate to multi-
storey buildings or giant ships; but when he saw a market vendor
pushing a cart laden with bananas, he could make sense of it. All
the other signs of potential change were so far outside his previ-
ous life experiences that his mind could not grasp what his eyes
were telling him.
There is some truth to this explanation. But it cannot be the
only explanation for companies failing to see signals of change in
their environment. It would mean that old companies with rich
histories would always prevail over young ones, at least when it
 LEARNING 42

came to being flexible, because the old companies would have

built up a much greater store of experience from which to draw
meaningful links with new perceptions.
Actually, I believe that this dynamic does take place. An
older company with a good institutional memory will probably
see more than a young company will. Nevertheless, old and expe-
rienced companies consistently miss the signals. As the statistics
convincingly show, they are just as prone to crisis as newer com-
panies. Some other factor must be at work.

Theory 4: We cannot see what is

emotionally difficult to see

In the oil industry, we saw the value of sensitivity at first hand dur-
ing three major crises of the 1970s and 1980s: the OPEC oil-
supply crisis of 1973, the overthrow of the shah of Iran in 1979 and
the collapse of the oil price in 1986. In every large company, at
least a few people anticipated these crises and warned about them
in advance. Nonetheless, most of the companies failed to imple-
ment the required internal changes in time. By the mid-1980s this
lack of foresight had taken its toll, with the result that two of the
‘Seven Sisters’ — the major oil companies that had seemed impreg-
nable and unstoppable in the early 1970s — had been weakened or
had merged. Many smaller oil companies also died.
What happened to them? One key factor was the rise and
fall of exploration and production (E&P) — the high-tech compo-
nent of oil companies which drills for oil, pumps it from the
ground and sells it as crude oil. In the 1973 supply crisis (known
in America as the OPEC oil embargo), the price of crude oil
soared to hitherto unknown heights. It remained high for 13
more years. Although this represented a real crisis for just about
everybody, both inside and outside the oil companies, it was not
 THE MEMORY OF THE FUTURE 43

a crisis at all for E&P. Its product had suddenly gone up in price
from $2 per barrel to $30 per barrel — a very harmonious syn-
chronization with the outside world for them.
This harmonious situation resulted in a distinct shift in
resource allocation. Once, perhaps 30 per cent of an oil compa-
ny’s overall budget might have gone to E&P. Now this depart-
ment would receive 50 per cent or, in some companies, 70 per
cent. Careers of existing E&P managers were buoyed by success
and new, capable people went into E&P because that was where
the success was. As a result, at major oil companies, top execu-
tives increasingly had an E&P background.
Then came 1986. Suddenly the oil price fell back to about
$10 per barrel. For E&P in particular, this was a very disharmo-
nious crash. Worse still, the oil companies had now reorganized
themselves around the leadership of E&P. They had successfully
responded to harmony, but they found disharmony much more
difficult. The thinking of top leaders was coloured by their E&P
background, and E&P continued to take the lion’s share of
resources. They could not provide the same level of return on
those resources, but nor could they change their methods and
approaches. At many companies, they reinforced their resistance
to change with wishful thinking: ‘Things will turn around and the
oil price will become "normal" again.’
To make fundamental change was emotionally difficult. For
one thing, E&P managers at many oil companies had to come to
terms with the fact that their priorities and knowledge were no
longer the determining sources of triumph. Moreover, the com-
panies discovered that managing for fundamental internal change
is inherently far less gratifying than managing for growth. In a
non-growth environment, you no longer manage by allocating an
ever-expanding pool of resources. Now you have to cut costs;
you learn about frugality; and you seek new business with much
 LEARNING 44

greater risks, much less room for error and much more uncertain
rewards.
Understandably enough, those who have become used to
the high emotions of the previous phase will resist correction in
this phase. Switching direction is never fun for the people who
were at the front of the previous charge. Consequently, the news-
papers are full of examples of companies which, under conditions
of a shifting environment, postpone changing their (previously
successful) policies for far too long — until they slide into a deep-
ening crisis.
You cannot do anything to remove this emotional pain. It is
a necessary complement of fundamental change. And, indeed, it is
powerful. But it cannot be overwhelming, for otherwise compa-
nies would never change. As we can see from the example of
Stora, some do. There is, contrary to the pessimistic views of
these theories, enormous hope for improvement.
In addition, human beings seem to have developed a much
better capability for foresight than companies have. Thus, after
considering the first four theories during my enquiry at Shell, we
began to look into the cognitive nature of human adaptability.
What did psychological researchers believe was the factor which
gave human beings their ability to anticipate the future? And
could companies emulate this aspect of human behaviour?

Theory 5: We can only see what is relevant

to our view of the future

At Shell, this enquiry eventually led us to the work of David

Ingvar, the head of the neurobiology department at the
University of Lund (Sweden).5 The results of his research, pub-
lished in 1985, show that the human brain is constantly attempt-
 THE MEMORY OF THE FUTURE 45

ing to make sense of the future. Every moment of our lives, we

instinctively create action plans and programmes for the future —
anticipating the moment at hand, the next minutes, the emerging
hours, the following days, the ongoing weeks and the anticipated
years to come — in one part of our mind. This brain activity takes
place throughout the daytime, independent of whatever else we
are doing; it occurs in an even more concentrated form at night,
during sleep.
You have probably created a dozen or more such time paths
during the time you have been reading this chapter: ‘If I continue
reading for another hour, it will be too late to telephone Margaret.
Then I’ll have to leave to see Andrew before dinner. If Andrew
offers me a drink, I might be too late to call Marcia before she
goes to the theatre. It would be better to call her tomorrow morn-
ing anyway. On the other hand, if I stop reading now, I could still
go to town by train. If the train arrives late, I’ll phone from the
station and then take a cab, rather than the subway...’
These plans are sequentially organized, as series of potential
actions: ‘If this happens, I will do that.’ These are not predictions.
They do not pretend to tell what will happen. They are time paths
into an anticipated future. Each combines a future hypothetical
condition of the environment (‘if the train arrives late’) with an
option for action (‘I’ll take a cab’).
Not only does the brain make those time paths in the pre-
frontal lobes, it stores them. We visit these futures and remember
our visits. We have, in other words, a ‘memory of the future’, as
Ingvar calls it, continually being formed and optimized in our
imagination and revisited time and time again. The memory of the
future is an internal process within the brain, related to humans’
language ability and to perception. It apparently helps us sort
through the plethora of images and sensations coming into the
brain, by assigning relevance to them. We perceive something as
 LEARNING 46

meaningful if it fits meaningfully with a memory that we have

made of an anticipated future.
Ingvar remarks that among ‘normal’ people about 60 per cent
of these anticipated futures are favourable: good things happen in
them. And 40 per cent are dire. If the balance is disturbed, you get
perennial optimists or incorrigible pessimists, depending on
whether their prevailing memories of the future are positive or neg-
ative. In any case, the healthier the brain, the more alternative time
paths it makes, striking a reasonable balance between favourable
conditions and unfavourable ones. We make and store a great many
options for the future, far more than we will ever fulfil.
In his research, David Ingvar addresses the question of what
function this sort of memory of the future might serve. Why
would it have evolved? An obvious reason would be to prepare us
for action once one of the visited futures materializes. But he sug-
gests another purpose: as a filter, to help deal with the information
overload to which every human being is constantly subjected.
The human body, notes Ingvar, has a plethora of sensory
channels: the eyes, ears, nose, taste buds and every part of the
skin. Each one of those sensors sends to the brain a continuous
stream of signals about the surrounding world. So much random
information reaches the brain that the vast majority of it must be
ignored. The brain could not function properly if it gave equal
priority to all the information it receives. Ingvar hypothesizes
that our memories of the future provide a subconscious guide to
help us determine which incoming information is relevant. The
stored time paths serve as templates against which the incoming
signals are measured. If the incoming information fits one of the
alternative time paths, the input is understood. Information
becomes knowledge and the signal acquires meaning.
The message from this research is clear. We will not perceive
a signal from the outside world unless it is relevant to an option
 THE MEMORY OF THE FUTURE 47

for the future which we have already worked out in our imagina-
tion. The more memories of the future we develop, the more open
and receptive we will be to signals from the outside world.
If learning begins with perception, then Ingvar’s theory has
important implications for managers who are trying to guide
their company through a turbulent environment. Ingvar is, in
fact, saying that the act of ‘perception’ is not simply a matter of
collecting information — of looking at an object and noting all
sorts of observations and data about it. Perception, to a human
being, is an active engagement with the world. And, in a compa-
ny, it is similarly active. Perception requires a deliberate effort by
management groups within the company to ‘visit their future’
and develop time paths and options. Otherwise, the observations
and data that are collected will have no meaning.
Making this effort is easier for an individual human being
than for a company, because the brain is hardwired to perform
this sort of active engagement. Imagine that a man who lives in
France has taken his car to London by ferry on a business visit.
At eight o’clock the next morning, he drives his car from the hotel
to the office. While subjected to the considerable information
overload that stems from trying to find his way through rush-
hour traffic in a strange town, he switches on the radio to hear the
news. Even more signals reach the brain. At the end of the news-
cast, the announcer briefly mentions that a strike has been
announced for the ferry port of Dover.
Most of us would not even hear this piece of information.
But our French resident has a time path stored in his mind. He
has a memory of the future, seeing himself taking his car that very
night to Dover to catch a ferry. He hears the signal, because it is
relevant to that memory. Information has become knowledge.
A company is not hardwired to produce this sort of mem-
ory of the future. Managers must take specific action to produce
 LEARNING 48

one. That is why David Ingvar’s theory is so significant in point-

ing towards a means of improving a company’s powers of per-
ception. His theory explains, to my satisfaction at least, why
managers do not recognize external events in time to avoid a cri-
sis. And Ingvar’s theory also suggests that corporations can devel-
op the sensitivity they need, by finding ways to build up an
organizational memory of the future.
3
Tools for Foresight
FEW COMPANIES ARE AS CAPABLE AS HUMAN BEINGS AT DEALING
with the future. Managers may see signals of a potential
future and even talk about it together. But the managers and the
companies still do not respond in timely fashion to that future,
even after it has occurred.
Could it be that this difficulty exists because companies are
trying to deal with the future by predicting it? Predicting the
future is very different from creating the sorts of alternative time
paths into the future that David Ingvar’s work suggests. Back in
Chapter I, we noted that some management teams ask, ‘What will
happen to us?’ They are engaged in prediction. Other manage-
ment teams ask, ‘What will we do, if such-and-such happens?’
They are engaged in alternative time paths, particularly when that
question is put on the agenda of a meeting of a corporate board
with decision-making power.
By reverting to predictions as a standard way of thinking
about the future, the corporate powers of perception remain
greatly reduced. That leads them to the conundrum faced by the
mayor of Rotterdam, in a parable which we found very meaning-
ful in Group Planning at Royal Dutch/Shell.1
Imagine that it is 1920 and you have somehow been granted
 LEARNING 50

absolute power to predict the future. You happen to visit the

mayor of Rotterdam and, during that time, you describe in vivid
detail what is going to happen to his town over the next 25 years.
Thus, in an otherwise perfectly normal working day, the mayor
hears about the advent of the Weimar Republic, hyperinflation,
the 1929 stock exchange crash, the Great Depression that fol-
lowed, the rise of Nazism in Germany with its (for Rotterdam)
damaging economic policy of autarchy, the outbreak of the
Second World War, the carpet-bombing of the town’s city centre
and, finally, the systematic destruction of the town’s port installa-
tions during the calamitous winter of 1945.
The mayor listens to this information placidly. He gives
every sign of believing you. And then he asks, ‘If you were in my
shoes, hearing all this, amid all the other opinions and facts that
reach me during the course of my day, what would you reason-
ably expect me to do about this information?’
What is it reasonable to expect the mayor to do?
When I ask this question in discussion groups, we always
reach the same answer: there is nothing the mayor can be expect-
ed to do. Even if he gives your prediction a higher degree of cred-
ibility than most of the other information which reached him, he
would have neither the courage nor the powers of persuasion to
take the far-reaching decisions required by such a prediction.
The future cannot be predicted. But, even if it could, we
would not dare to act on the prediction.
Most people accept this thesis in a cool, academic debate.
Nevertheless, in real life there is an insatiable demand for predic-
tions. The yearning for some certainty about the future is so
strong that most of us will at times act against our better judge-
ment and demand some precise prediction of the future. That is
why there are worldwide industries which supply information
about the future — from fortune tellers and astrologers to consul-
 TOOLS FOR FORESIGHT 51

tants, academics and economists. These are industries rich in

euphemisms, in which the product is often wrapped in sheets of
fine print and jargon which the customer does not normally read.
But that doesn’t matter anyway, because few people with real
responsibilities dare take decisions based on the information,
even though they eagerly asked (and paid) for it in the first place.
Perhaps we can understand the corporate demand for pre-
diction better by looking at the individual demand. There is a
huge market, for example, for astrological forecasts — in which
some highly successful suppliers try to satisfy an almost insatiable
demand for a feeling of reduced uncertainty. One of those highly
successful suppliers has been the British astrologer Patrick Walker,
syndicated worldwide through Rupert Murdoch’s News
International. The Murdoch syndication department in New York
estimates Walker’s audience to consist of at least a billion people.2
From the perspective of decision making, interviews with
Walker himself are fascinating. He describes his brand of popular
newspaper astrology as a sort of weather forecast and finds it hard
to explain why people believe that he is such a good astrologer.
‘Perhaps it is because I try not to make rigid predictions . . . astrol-
ogy has got nothing to do with your marriage breaking up. You
are the one that breaks up the marriage . . . All I can do is highlight
the circumstances. How you behave is up to you.’
One would think that an astrologer’s methods deserve lit-
tle attention when contemplating the serious subject of interpret-
ing the future for business decisions. Yet it is interesting to
contrast Walker’s view with the approach implicitly taken by
many financial forecasting systems. Walker leaves room for free
will; he tells his audience that they, at least in part, will shape their
own future. Corporate and government planners, by contrast,
imply (or pretend) that the future is a fatalistic given. No action
by management will make any difference to the way the future
 LEARNING 52

will unfold. The job of the planner is to divine the ‘right’ future
as closely as possible. And the managers, when faced with their
own bad decisions, use the excuse that they were given the wrong
prediction!
This is abdication of managerial responsibility: dealing with
the future can never be delegated. It is an uncomfortable compo-
nent of the manager’s job. It is one of the reasons for senior exec-
utives being paid such high salaries.
Daniel Yergin, author of the Pulitzer Prize-winning history
of oil The Prize,3 found exactly this sort of resistance in the oil
industry during the mid-1980s. In 1985, he once told me, there
was much apprehension in oil company boardrooms that the
price of crude oil could crash dramatically. But he did not know
of any company in which the senior leaders had addressed the
question of what they would do if the price did fall.
It is always easy for managers to try to predict. It is always
tempting to address at great length the question: what will hap-
pen to us? Will the oil price fall? Will our competitors expand
into our business? Will the Soviet empire disband or the South
African government change? Will the technology we use become
obsolete? Like a great many advisers to an imaginary mayor of
Rotterdam, managers tend to spend many hours comfortably
debating the likelihood of one future or another, without arriving
at any conclusion.
Imagine, instead, if managers asked themselves what they
would do if such-and-such happened. Suppose the oil price fell (or
rose)? How would we react? Suppose our competitors expanded?
Suppose there were a change in government or a shift in our tech-
nologies? What would be our response? Answering these ques-
tions, or questions like them, would allow managers to work out
one or more of David Ingvar’s mental time paths. It would allow
us to build ourselves a series of memories of the future — anticipa-
 TOOLS FOR FORESIGHT 53

tions of events that might or might not take place.

Thereafter, we would be prepared. We would have thought
about our course and played it out in our imagination. We would
not have to try to predict the future, because we could rely on our
memory of the many futures that we have already visited.

Planning and the illusion of certainty

Back in the 1930s, the corporate world made its first serious
attempt to deal systematically with the future. A series of ‘tools for
foresight’ were developed, under the generic name of ‘planning’.
Over the next three decades, every line manager learned to incor-
porate the output from these planning tools into his or her deci-
sion making. Yet, as we have just seen, planning does not usually
mean learning to anticipate possible futures, build memories of
them and prepare ourselves for them. Instead, planning is typical-
ly seen as the work of reducing uncertainty through prediction.
When I was first exposed to planning, as a student shortly
after the Second World War, the task of reducing uncertainty was
organized so that trained specialists would handle it. By about
1940, many companies had begun to hand over the task of worry-
ing about the future to the ‘backroom boys’ in the planning
department. The more pragmatic characters on the line could thus
‘get on with the job’, without having to waste valuable time imag-
ining possible events. Management was thus split between those
who do and those who plan.
In most companies, this specialized planning activity found
a home in the finance function — or, more precisely, in accounting.
In retrospect, this was the worst possible environment for devel-
oping memories of the future, but by the standards of corporate
practice, it was a perfectly logical place for the new profession of
planners to land. Financial-based planning promises to provide a
 LEARNING 54

great deal of factual, quantified information about the future, in a

form (figures and money) which is highly beguiling and sugges-
tive of precision.
When given something new to do, most people begin by
doing what they already know how to do well. Thus, when asked
to satisfy the company’s yearning for knowledge about the future,
the accountants of corporate finance began with their strengths:
their ability to make balance sheets and profit and loss accounts.
They set out to predict next year’s budgets, balance sheets and
P&L accounts based on estimates of next year’s sales and operat-
ing costs. They calculated whether the company would be short of
money or have a surplus one year into the future. Elegantly col-
lated in a binder, the final estimate provided the company’s ‘doers’
with a quantified report — useless, but compelling — on how
tomorrow, next month or next year would look.
There was one useful aspect to this exercise, however. In the
process of compiling their information, the accountants had to
think quite hard about what was happening in the outside world.
People who have to estimate next year’s sales will find themselves
thinking: what will economic activity be like? Will GNP grow or
will there be a recession? To answer such questions seriously, one
must come up with some idea about inflation over the next 12
months and its influence on wage and price levels. In short, look-
ing from the inside out, the planners of accounting saw a great
deal of the outside world. And some of this awareness percolated
through to the rest of the company.
But then, in the 1960s, financial planning underwent anoth-
er refinement. Rather than relying on an educated head-office
guess at next year’s sales, many companies developed ‘bottom-up’
planning. The planners went out to the people close to the action,
managers throughout the organization, and asked (for instance)
each district manager what he thought he would sell next year,
 TOOLS FOR FORESIGHT 55

two years from now and even five years hence. The planners
added up all the figures (changing some if they did not like the
totals) and compiled the results into the ‘budget’ or ‘forecast’, or
whatever it might be called.
From here it was only a small step to ‘management by objec-
tives’. If we were going to take the trouble of going all the way to
the field and asking the district manager what he would sell next
year, we might as well make him think very hard about the reply.
Thus, if we agreed with his estimates, we would declare them to be
his ‘target’ figures. His performance would be checked and his
rewards calculated on how well he chose and met his targets.
Now the forecast was an internal contract. Little outside
information could break in, and all decisions emerged from the
same introverted process. Planning, which had once been cast off
from the ‘decision making’ of the pragmatic line manager and rel-
egated to a specialized intellectual realm, was now one of the
principal vehicles for making decisions!
All through the 1950s and 1960s, the Shell Group faithfully
took part in these developments. We finally reached our epitome
of specialized planning with the launch in 1967 of the Unified
Planning Machinery — the ‘planning system to end all planning
systems.’ The UPM procedures, set out in a thick manual that all
managers were supposed to follow, contained all the elements of
a state-of-the-art financial prediction system. There were target-
setting and performance-control procedures, to be administered
both up and down the hierarchy and applied inside each of the
more than 100 countries in which the group operated. The esti-
mates and predictions would gradually percolate from the local
offices around the globe up into the coordinating units which
formed the matrix organization in the two central offices in
London and the Hague. These meetings would culminate in a
presentation to the Committee of Managing Directors and to the
 LEARNING 56

board, who would formally approve the capital budget and next
year’s operating plans.
This process continues today. Sometimes the time horizon
looks out two years; at other times, five. (The critical emphasis is
in any case always on next year’s figures.) Each year, the machin-
ery is wound into action in June or July, in the operating compa-
nies. Each year, after countless meetings and reports and an
enormous amount of thought and effort, the board finally reaches
approval in December. Each year, the Unified Planning Machinery
delivers its estimates of future activity, and each year, the group as
a whole bases its investment decisions on those estimates.
There is only one problem. Whenever times are turbulent,
and anticipating the future is most critical, the Unified Planning
Machinery is wrong. Dead wrong. It failed to foresee the spike in
oil prices in the 1970s, when OPEC coalesced into a more pow-
erful cartel. It failed to anticipate the collapse of oil prices in the
mid-1980s, when the OPEC cartel devolved. And UPM failed to
predict the restructuring of the oil business in the 1980s and
1990s, as the greater source of profits switched from one part of
the business (production) to another part (downstream and mar-
keting) and back again. Even during the 1960s, which were rela-
tively quiet years for the oil business, those who looked could see
this type of problem with UPM already emerging.

Shell’s scenario experience

Fortunately for Royal Dutch/Shell, it was also developing an
alternative tool for looking at the future, alongside the Unified
Planning Machinery. This technique, called scenario planning,
was well suited for building memories of the future.
Unfortunately for a conventional manager, the scenario approach
presupposes that the future is plural. Scenario planning requires
 TOOLS FOR FORESIGHT 57

managers to abandon the one-line approach, the assumption that

there is only one predicted future. In scenario planning, there is
always more than one scenario.
Each scenario is simply an imaginative story about the
future — a sketch of the ‘lot of life’ as it could develop from the
present moment into the future. The name was chosen and pop-
ularized by Herman Kahn, the well-known futurist from the
Rand Corporation and the Hudson Institute. (He borrowed the
name from the movie industry; the original meaning of the word,
as the Oxford English Dictionary defines it, is a ‘sketch of the plot
of a play; giving particulars of the scenes, situations, etc.’) Kahn
was best known for his scenarios about nuclear war, in which he
advocated that people should ‘think about the unthinkable’ so
that, if nuclear war did become imminent, society would be less
vulnerable and less likely to slide into a holocaust.
Scenario planning has been practiced in some form since the
early 1960s, but even today, it remains surrounded by vagueness
and an air of mystery. People are unsure whether it is a process
for reaching better decisions, a way to know the future better, or
a combination of both. Many scenarios, like science fiction sto-
ries, look at the future of mankind as a whole or the economics of
an entire region; to a businessperson, it is not clear what to do
with these. When trying to sell soap powder in Canada or manu-
facture power generators in Switzerland, the businessperson
needs a translation of these grand, overarching developments into
something closer to and more recognizable in his or her own
world. The plot needs to be focused on its particular audience.
Learning to focus scenarios on a business purpose was in
fact part of Shell’s contribution to the practice. In 1968, soon after
the introduction of UPM, an ad hoc study group within Shell
undertook a study on ‘the year 2000’. The impetus was the same
predictive question that would later spark the long-term
 LEARNING 58

survivors study: how soon would the world run out of oil? And,
if those resources were depleted, would that condemn oil compa-
nies to an unexciting, low-growth future? In short, as Shell peo-
ple put it, ‘Is there life [for our company] after oil?’
The Year 2000 study was professionally done and produced
real business results in the Shell Group. A metal-producing com-
pany was purchased, a 50 per cent stake was taken in a nuclear
venture and an internationally trading coal division was started.
These outcomes were not all successful, and in the end Shell
retreated from many of them. In retrospect, it could have been
seen from the beginning that diversification would take place. It
was implicit in the question: is there life after oil? If you want the
answer to be yes, then you must enter some other business. In the
early 1970s, the only way to generate that business, at a scale
appropriate for a major oil company, was to acquire or instigate it
from the top, in a centralized move: diversification.
But the Year 2000 study also had a second outcome. A new
planning division was conceived, with a core group of some of
Shell’s more innovative people. These included Pierre Wack, Ted
Newland, Peter Beck and Napier Collyns — people who are now
known for their writing and work on planning, even outside
Shell. They were familiar with Herman Kahn’s work and they
started to build on it, developing their own form of the scenario
technique as a possible answer to two questions: how do we look
20 to 30 years ahead? How can we get people to discuss the
‘unthinkable’ together?
Like Herman Kahn, the scenario planners tried to grasp the
changing forces in the social values, technology, consumption
patterns, political thinking and financial structures of the world at
large. But, unlike Kahn, Shell’s scenario planners remained
focused on the future of the oil industry. As Pierre Wack, the
leader of the scenario team, never ceased to explain, ‘Scenarios
 TOOLS FOR FORESIGHT 59

must be relevant.’ To help businesspeople understand more about

what the future could mean, the plots had to operate in the world
of their own business. Only then would a manager see the rele-
vance of global forces and possible futures. Only then would
these stories help to raise the eyes towards the horizon.
In this sense, scenarios are neither a mystery nor a superior
way of ‘planning’. They are tools for foresight: discussions and
documents whose purpose is not a prediction or a plan, but a
change in the mindset of the people who use them. By telling sto-
ries about the future in the context of our own perceptions of the
present, we open our eyes for developments which in the normal
course of daily life are indeed ‘unthinkable’. Relevant scenarios,
brought down to the level of the individual player, help the man-
ager and his or her colleagues scout the lay of the land and see a
wider scenery. Scenarios bring new views and ideas about the
landscape into the heads of managers, and they help managers
learn to recognize new, ‘unthinkable’ aspects of the landscape even
after the scenario exercise is over. As Pierre Wack liked to express
it, ‘Good scenarios change the microcosms of management.’
To accomplish this, it is not enough for the scenario writers
to look out at the world from the narrow perspective of the com-
pany, as if from a window in the company’s building. One has a
rather limited view from most business edifices. Planners must go
out into the wide world. Once out there, they have to look back
at their company and ask themselves the question: what relevance
could the driving forces that we see externally have for the more
limited world of our own company and industry? This need for
wider perspective explains the exotic list of sources of inspiration
which Peter Schwartz (longstanding scenario planner and Royal
Dutch/Shell scenario team leader during the mid-1980s) propos-
es in his book The Art of the Long View. Schwartz suggests that
the budding futurist should learn to look at the fringes. Talk to
 LEARNING 60

people ‘with whom you disagree deeply, but can talk amicably’.
Read widely, ranging far outside your immediate business inter-
ests (from design magazines to self-published youth ‘zines’ to
government journals, he suggests), but don’t forget the
4
Economist.
From its beginnings in the early 1970s, the Shell planning
group developed an extensive network of outside contacts who
were chosen for their insights and understanding of what went on
in the world at large. Pierre Wack called them his ‘Remarkable
People’. Historian Art Kleiner describes his method this way:

[Wack] sought out acute observers with keen, unending

curiosity. These people devoted themselves to seeing: to con-
stant attention to the ways the world worked . . . Sometimes, a
remarkable person from outside Shell might stumble into a
scenario presentation, as an Iranian physician did in the early
1970s, looking for stimulating conversation . . . The two men
became close friends. Each year Pierre would visit and ask
how his perception had changed.5

There was little duplication with the planning still done by Shell
supply or finance people, who were looking only at oil-related
developments. The new planners did not ignore oil and energy
concerns, but they were looking for ‘driving forces’, which might
come from anywhere and ultimately work through to affect the
world of energy and oil. They analysed these forces to see
whether, and how, the resulting changes might affect their own
businesses. In short, scenarios provide tools through which the
non-fashionable and weak signals may be picked up and consid-
ered, without overwhelming the managers who use them.
Scenarios also hone managers’ judgement about the signifi-
cance of those forces, by providing new ways to group and con-
 TOOLS FOR FORESIGHT 61

sider them. In the big, wide world in which Shell (or any compa-
ny) operates, there are always many driving forces for change,
interacting with each other in a bewildering blizzard of mutual
causes and effects. To plot them all would lead to confusion and
(probably) despair: no one could ever make sense of all this! This
explains why well-crafted scenarios typically combine a number
of forces into a story that seems simple, but is actually quite
sophisticated.
Spending time discussing a scenario allows managers to see
long-term interactions — among, for instance, their own capital
investment plans, the energy efficiency of consumers and Middle
Eastern politics. In the process, the managers develop a language
in which they can later communicate among themselves about the
subject, to arrive quickly at decisions. The scenario gives them a
context for considering all of these forces — perhaps not compre-
hensively, in the manner of an academic dissertation, but dynam-
ically. A story of a falling oil price can bring all of these forces
vividly to life in the imagination, so that they linger in memories
of the future, in words that are understandable to their colleagues.
In the interaction among driving forces, there is always a
range of possible outcomes. Many listeners will ask, ‘What is the
most probable outcome?’ But the answer is that all of the possi-
ble interactions among the same set of driving forces in a complex
system have an equal probability of taking place. For that reason,
several scenarios will have to be written: enough to condition
managers to see past many of their blind spots. The number of
scenarios is important. Too many scenarios confuse the manager.
An uneven number gives the manager an unfortunate escape
route; it’s too easy to bypass the scenarios’ implications by
picking ‘the one in the middle’, the compromise future that is
seen as an alternative to the extremes. Two is probably a good
number for scenario exercises; it forces the manager to make a
 LEARNING 62

choice between them and thus to think through the ramifications

of both.
Once written, scenario stories are tested and quantified
with the help of simulation models and (in the case of the Royal
Dutch/Shell Group) the company’s data banks on energy and
economics. The quantification helps focus the scenarios and it
shows whether the stories are internally consistent. The end
result is a series of consistent, plausible futures, which don’t
merely provoke thought. If they’re successful, they should pro-
voke surprise and even emotion. ‘I never realized this could hap-
pen to us,’ people might say.
Having a low number of scenarios inevitably means that the
writing process is one of reduction. Conceivably, several thou-
sand pages of interviews and studies might result from as many as
two years of study and research. All of these reports have to be
reduced to a booklet which can be read in a sitting. This means,
preferably, a booklet of less than 70 pages. The person who writes
the final draft of this booklet must be a good storyteller, with a
nose for the main themes to be developed. The best-remembered
scenarios, in fact, have some of the characteristics of fairy or folk
tales. As Joseph Campbell remarked in his book The Hero with a
Thousand Faces: ‘The crux lies in the fact that our conscious
views of what life ought to be seldom correspond to what life
really is.’6
Indeed, the mythological elements of the Hero’s journey
seem to resonate fairly well for most modern businesspeople. The
Departure, the Belly of the Whale, Initiation on the Road of
Trials and finally the Return (having earned the Freedom to Live)
all have parallels in scenario stories. In my experience, some of the
scenarios that were best understood and longest remembered by
the Shell organization had elements of Joseph Campbell’s
description of the timeless adventure:
 TOOLS FOR FORESIGHT 63

The mythological hero, setting forth from his hut or castle, is

lured or voluntarily proceeds to the threshold of adventure.
There he encounters a shadow presence that guards the pas-
sage. The hero may defeat this power and go alive into the
kingdom of the dark . . . journeying through a world of unfa-
miliar forces, some of which severely threaten him. When he
arrives at the nadir, he undergoes a supreme ordeal and gains
his reward. The final work is that of the return . . . the hero re-
emerges from the kingdom of dread. The boon that he brings
restores the world.

We saw this theme, for instance, in Pierre Wack’s scenario of the

‘Rapids’ (an anticipation of the turbulence of the late 1970s) or
Peter Schwartz’s ‘The Next Wave’ (an early glimpse of the global
economy of the 1990s). In both of these worlds, the corporate
hero first had to conquer the rapids of an economic recession or
the threat of a world of low oil prices, before the boon (of
microchip technology) would reward the hero and restore the
world. Both scenarios helped to change the Shell corporate mind
and opened our eyes to the gloomy messages of, respectively, a
recession and a world of low prices. It was plausible that we could
survive and even thrive through the crisis. On these occasions, the
message was absorbed without the messenger (the planner) being
shot. Such is the power of a strong, symbolic story!
In the telling (through presentations), the story line
becomes stronger. Scenarios act as a signal-to-noise filter. The dri-
ving forces sharpen. The events depicted enter the mind with less
background noise and thus with a stronger profile and clearer
outlines.
The scenarios help managers see past their biases, but the
bias of the scenario storyteller must also be taken into account. It
is no surprise that, as creators of a work of art, different artists
 LEARNING 64

write different stories, based on their individual observations of

the same realities. Some scenario writers have a message. They
write ‘normative’ scenarios: pictures that they believe will con-
vince the listeners to join in creating or forestalling alternative
worlds. Other scenario writers have a strongly developed frame-
work for thinking about their world. They are so aware (for
example) of the enormous influence of technology that they can-
not easily imagine a world in which technology has little influ-
ence. Other scenario writers are more inductive; they tend to let
the story evolve during the course of their one or two years of
research, with little forethought about where the data may take
them. They listen to the themes that emerge as different people
talk about their observations. If they are successful, they produce
a work of remarkable insight and power; if not, they produce a
scattered mishmash of all the things they have heard.
Once the scenario is written, different people will present it
to audiences in different ways. How well the scenario writers are
heard depends, in part, on their skill and quality as presenters.
Some scenario writers are preachers. They want their message to
be heard. Some are almost mystical in their emphasis on the wide,
overarching themes. And others focus on many nuances, to let the
image emerge as if from the thousands of coloured points in a
Seurat painting.
Nearly always, if the scenario development has been con-
ducted well, the results will be disturbing. The scenario presenter
will have things to say which the line manager will experience as
unwelcome intrusions into his or her business thinking. To be
heard through a veil of emotional denial requires scenario writers
who have a high degree of acceptability in circles of line manage-
ment. That was why all of the chief scenario team leaders at
Shell — starting with Pierre Wack and Ted Newland, through all
of their successors — were carefully screened by every member of
 TOOLS FOR FORESIGHT 65

Shell’s Committee of Managing Directors. As a group, the CMD

‘endorsed’ not only the scenario leaders, but also every new set of
scenarios before they were published in the Shell community.
Endorsement did not mean that each and every one of the
managing directors agreed with the contents of the scenarios. Quite
the contrary, on several occasions. But they agreed that the major
themes of the scenarios deserved to be kept in mind, however dis-
tasteful or improbable the details of these themes might seem to be.
Endorsement would never have been forthcoming if the directors
had not had an opportunity to convince themselves of the integri-
ty and sound judgement of the principal scenario writers.
Public endorsement was not only necessary; it became offi-
cial practice. Early in the 1970s, the CMD issued a planning rule:
annual capital and operating budgets had to be defended against
the background of the scenarios en vigeur. For those of us in line
management, there was no way around it. We had to show that
we had at least considered the possibilities that the scenarios had
raised. If we wanted our budgets approved, we had to read the
scenarios with great attention and make sure that we had our
counter arguments ready.
This is a remarkable illustration of the power of corporate
rules, if issued with wisdom and followed with self-discipline.
The result was a widespread attentiveness to outside events that
was so thorough it was hardly noticed after a while. It simply
became ‘the way things are done around here’ to consider the
effects of external forces that might seem counterintuitive at first.
Without this rule, it is a real possibility that, today, the Shell
Group would still be a company with a one-track mind, blissful-
ly unaware that the future is uncertain and different from a
‘preferred’ one-line forecast.
Even so, endorsing and legislating attentiveness to the sce-
nario process was not enough. There had to be a head of planning
 LEARNING 66

who could stand in for the line managers and represent their con-
cerns and attitudes at every step of the scenario-creating process.
In the 25 years since the group decided to set up its scenario plan-
ning outfit, it has never appointed a professional planner to head
this activity. The six individuals who have been head or coordi-
nator of group planning (including myself) have been line man-
agers, mostly with a finance or exploration background.
Presumably, this helped to keep the activity rooted in the
business. It also helped to infuse it with a little credibility. The
colleagues of the chief planner’s previous ‘line’ incarnation were
inclined to give him some benefit of the doubt, at least in the
beginning. Even so, the role of chief planner was often that of an
unwelcome messenger of bad news, a doomsayer. Most managers
find it tiresome, at the moment when they are busy spending a
few hundred million dollars buying a concession or extending a
refinery, to stop and spend time ‘thinking the unthinkable’. The
planning coordinator’s little capital of fraternal goodwill dissi-
pates rather quickly in those circumstances. Perhaps this is the
strongest argument for appointing line managers, rather than pro-
fessional planners, to head the corporate planning activity. (But
don’t leave an ambitious young line manager in that post for too
long, because it could kill a career. Any planner runs the risk, in
the long run, of being seen as an irrelevant court jester, or a
Rasputin whispering mysterious schemes in the managing direc-
tor’s ears. Neither is an attractive career prospect!)
I hope that the foregoing paragraphs have demonstrated the
dimensions of scenarios as tools and the tool-building nature of
scenario practice. Since the early Shell experiences, there has been
a growing interest in this new practice of corporate foresight. The
first good books on the subject are now in the public domain.
Peter Schwartz’s The Art of the Long View7 explains how to con-
duct a scenario exercise. Kees van der Heijden’s Scenarios: The
 TOOLS FOR FORESIGHT 67

Art of Strategic Conversation8 is a solid handbook on the princi-

ples of scenario planning. Art Kleiner’s The Age of Heretics9
recounts how Herman Kahn, Pierre Wack, Ted Newland and
others developed the practice.

The bridge between scenarios and action

Ironically enough, a deemphasis on prediction seems to lead to
accuracy about the future. It is remarkable how often the scenar-
ios published at Shell over the years predicted important devel-
opments in the world around the Shell Group, often years ahead.
The timing and quantification of these events or trend breaks
were not always right, but the scenarios were often quite clear
about the results and implications of the change.
Shell’s scenario makers rightfully claim that they recognized
the future ahead of time. They foresaw the energy crises of 1973
and 1979, the growth of energy conservation and the reduction of
demand for oil, the evolution of the global environmental move-
ment, even the breakup of the Soviet Union. Shell’s management
had the opportunity to take cognizance of important changes in
the world, and it was early enough to make the decisions necessi-
tated by these external changes. Many people would say that this
gave the Shell Group an important competitive advantage.
Yet in the early 1980s, both top management and planners
felt uncomfortable with the scenario approach. We could see no
discernible influence, from this advance knowledge, on the major
decisions which had actually been taken during the previous
decade.
Top management saw an expensive planning outfit which
produced colourful stories, very useful for public relations but
too far removed from the real business. They remembered occa-
sions when the planners had sketched a future which had not
 LEARNING 68

come to pass. They showed little awareness of the occasions when

the planners had been ‘right’. Top managers began to say that this
planning effort should be ‘closer’ to the real business.
The planners, certainly, were less harsh in their judgement,
but they found it equally difficult to cite a convincing example of
a decision which had been taken after a scenario had highlighted
a critical change. In their defence, the planners maintained that
scenarios were not supposed to work in such a crude and direct
fashion. Instead, as noted earlier in this chapter, they argued that
scenarios served to change the ways in which managers saw and
understood their world. Scenarios were designed to oblige man-
agers to question their assumptions and reorganize their inner
mental maps of reality.
What did that mean in simple business terms? It meant that
scenarios were supposed to make managers say, ‘Aha! Now I
understand what’s going on!’10 As a result, the planners argued,
decision makers would make decisions different to those they
might otherwise have made.
However, how to achieve this ‘aha’ experience in the minds
of managers was not quite clear. This was seen as the real chal-
lenge of scenario analysis and was clearly not solely dependent on
the eloquence of the presentation and the beauty of the charts.
In short, we in Group Planning were facing a real enigma.
We could see, in retrospect, that scenarios had a proven track
record of reasonably reliable predictions, well ahead of time. But
we could not offer the sceptics any demonstrable evidence that
the Shell Group, as a whole, had changed its behaviour or become
more adaptive. There was at best a weak link between our
advance knowledge and the actual business decisions which had
been made during the previous decade.
Perhaps the issue was not scenarios at all, but the decision-
making process that they were intended to influence. Could it be
 TOOLS FOR FORESIGHT 69

that corporate decision-making processes needed some improve-

ment? And, if so, what was the real nature of decision making in
the Shell Group — and elsewhere? We resolved to explore these
issues at Group Planning. We would investigate whether the very
act of decision making could be redesigned, to influence man-
agers — and the company as a whole — to learn.
4
Decision Making as a Learning
Activity
IN RETROSPECT, PERHAPS I WAS TOO HASTY IN MY ENTHUSIASM
when we first began to reflect on the decision-making process
at Shell Group Planning in the mid-1980s. First, we had started to
think about the role of planning in the company’s decision-
making processes. Phrases like ‘planning is a catalyst’ came easily
to our minds, working as we were in the oil and chemical indus-
tries. Then we developed the idea that ‘planning is learning’. Only
gradually did it dawn on us that decision making itself could be a
learning process.
I can now see that I came to this understanding quite natu-
rally. In the previous decade, my two daughters had been of
school age and my attention had been directed to the subject of
learning. I had read the work of John Holt1 and it had revived
some of my old university interest in psychology. So I began to
talk about ‘learning’ and ‘accelerating learning’ for the whole
organization. We could gain, I argued, by speeding up the rate of
that learning. But I was not always heard with approval.
Organizations, I was told, do not learn: people learn!
Statements like these stem from a view of decision making,
 DECISION MAKING AS A LEARNING ACTIVITY 71

and another view of learning, that are embedded in the prevailing

culture of Royal Dutch/Shell and in society at large. These points
of view are inherited from the practices and theories of academia
and management. The two activities, learning and decision mak-
ing, are supposed to be completely separate.
In academia, decision making is called a science. The stu-
dent can follow courses on the subject; books are written about
it. The practitioners of this science, trained in the art of manage-
ment, sit in their offices and follow the expert steps for making
appropriate decisions. There is no need to learn during the
decision-making process; they have already learned everything
they need to know.
As for learning itself, according to the prevailing view, you
(if you are a manager, at least) are supposed to learn only during
a particular part of your life: the school years. This is a time of
fun, without too much responsibility. Then you move into real
life, into work at a company where you apply your knowledge.
Play stops and hard reality takes over. You are paid for what you
know. The more you know (or have learned in school), the more
you should earn. Education is not a vehicle for expanding your
capability, but simply a credential for bettering your lot.
Having more knowledge may ultimately mean that you
become a leader. Then, at last, people will listen to you. They will
be convinced by your logical arguments and the superior array of
facts at your disposal. If they still do not carry out your com-
mands (assuming that you have explained them clearly, reason-
ably and calmly), it is probably because other people in the
organization — wittingly or otherwise — put barriers in the way. It
is your job to have those removed. Leadership has as little to do
with learning as decision making does. Indeed, when a leader
says, ‘I learned something I didn’t know before’, it detracts from
his or her ability to appear certain and thus to inspire confidence.
 LEARNING 72

A leader who learns is a leader who is unsure.

This attitude is a caricature of intelligent human life. It por-
trays people as motorcars: you start at a service station (universi-
ty) and fill up your ‘brain tank’ with knowledge. Then you use
your intellectual fuel to advance down life’s motorway. In this
view, there is no need for institutions to make learning happen
more effectively or on a larger scale. All the knowledge of the
company is already embedded in the heads of its employees.
Learning, except perhaps for a bit of ‘touchup’ learning to stay
abreast of new technologies, is assumed to be already covered.
This view is reflected in the way we recruit, remunerate and
promote people. There is no place at the top for an actor who
seeks to anticipate outside events by (for example) bringing peo-
ple together to look at developments which might turn into a cri-
sis. There is no room for someone who admits that he or she does
not have all the answers. The idea that the company itself could do
some learning of its own does not enter into anyone’s thinking.

The reality of decision making

In the 1980s, as we studied the core intellectual activity of the orga-
nization around us, we began to reject the prevailing hypothesis.
Once attuned, I could easily see for myself that decision making
was a learning process. It was, in fact, hardly individual at all. It
was primarily a social process, simple, unheroic and unscientific.
After all, what happens in decision-making meetings?
People talk. Analytical techniques, such as net present value,
earning power calculations and optimization models, may be
included in the preparation of information for the meetings. But
in themselves these are barren soil for decisions. Decisions grow
in the topsoil of formal and informal conversation — sometimes
structured (as in board meetings and the budget process), some-
 DECISION MAKING AS A LEARNING ACTIVITY 73

times technical (devoted to implementation of specific plans or

practices) and sometimes ad hoc.
Suppose that you and I are part of a team, holding meetings
to make a decision. Look closely at what happens during such a
meeting. We talk. Ideally, we talk freely and openly. If we have
any hope of reaching a decision, we know the meeting can’t be
dominated by one person — certainly not by the boss. We know
that nobody in the room has the solution at hand. We will have to
struggle together to find an answer to a situation that concerns us
all. If the meeting is to be effective, therefore, none of us can lose
patience with the thought processes of our colleagues. We cannot
throw our weight around or stand on our stripes.
However, even if our meetings are not well organized and
well managed, they are still conducted through conversation.
2
This process of conversation goes through four stages. They can
be described best, I believe, in cybernetic terms:

✦ Perceiving. A meeting is called because somebody has seen or

heard of an event or phenomenon which is outside the normal
course of business. Sales may have slipped, a government may
have changed political colour or the competition may be offering
a new product. People say, ‘We’d better start figuring out what it
means to us!’ In short, the stage of developing a mental model —
 LEARNING 74

an internal interpretation, colouring the way we look at that new

event — has begun.

✦ Embedding. Now we spend most of our time explaining to

each other ‘how we see the problem’. We try to understand its
relationship to our known business world and to embed the
change within our previous understanding. If our team is suffi-
ciently diverse — with some technical participants, some from
finance, some marketers and some human relations people — we
can gradually build a multifaceted picture of the situation. We
invent language to ‘label’ the modules of the model that we are
beginning to share. We give birth to company jargon — shorthand
expressions for parts of our understanding of the situation on
which we now agree. We are externalizing and calibrating our
mental models so that we may reach a coherent decision together.

✦ Concluding. Gradually, shared understanding leads to plans

for action. Somebody is bound to ask, ‘Now, what would happen
if... ?’ From that moment on, the meeting becomes even more dis-
ordered, and yet more productive. We talk through the ‘what if?’
consequences of our options and potential actions. It is as if our
shared understanding has become a model on the table before us,
and we experiment with various conclusions. ‘What if we
launched a new product? Or changed the packaging? Or
switched prices?’ You could arguably see this phase as simulating
a model of the situation by running imaginary iterations of our
decision.

✦ Acting. Finally, we are ready for implementation and action.

The only relevance of the decision-making process resides in the
action which results from it. However, we design the action, as
best we can, to keep track of its effects, monitoring and evaluat-
 DECISION MAKING AS A LEARNING ACTIVITY 75

ing the process. How do we know it will be a success? What can

we do to be more certain of our observations? We are like scien-
tists, running a model, making our ideas real, so that we may per-
ceive the effects...
... and the cycle begins all over again.

As it happens, these four elements — perceiving, embedding, con-

cluding and acting — are seen by various psychologists as the
defining elements of learning. Whether they are managed effec-
tively or not doesn’t matter. Every act of decision making is a
learning process.

Learning by accommodation
Based on his studies of children’s development, the Swiss educa-
tion theorist Jean Piaget has proposed that there are two types of
learning, assimilation and accommodation:3

✦ Learning by assimilation means taking in information for

which the learner already has structures in place to recognize and
give meaning to the signal. (As David Ingvar would say, the learn-
er already has a memory of the past or the future with which this
information fits.) The learner can perceive, digest and act on this
information easily — in the way that a student, having looked
something up in a textbook, can use it to answer a test question,
or an artisan can look up a technique and use it on a project.
In companies, most information used in operational deci-
sions fits this category. For example, bank management instantly
recognizes the signal of an important increase in the interest rate.
The bank has all the procedures and structures in place to give
meaning to the signal. The institution, at all levels, is ready to
digest it — to come to conclusions and to act on them in decisions
 LEARNING 76

about deposits, loan transactions, money-market operations or

any other bank business.
This is the activity which most people have in mind when
they think of learning — being exposed to facts and assimilating
them intellectually. In this activity, there are ready-made ideas
and structures which fit the situation. These ideas can be trans-
mitted from one individual to another. This is the learning activi-
ty of the traditional lecture hall or classroom; it is so prevalent
that many of us are conditioned to equate learning with teaching.
In companies, the closest one gets to teaching is when an
expert or a consultant stands up in a management meeting and
doles out his or her wisdom. This teaching is not the dominant
method for training in companies. That’s just as well — it shows
that in companies, when performance counts, people recognize
that teaching is an ineffective route to learning.

✦ The other type of learning, as Piaget puts it, is learning by

accommodation. In this type of learning, you undergo an internal
structural change in your beliefs, ideas and attitudes. When we
learn by assimilation, says Piaget, the lectures and books of con-
ventional school learning are sufficient. But learning by accom-
modation requires much more. It is an experiential process by
which you adapt to a changing world through in-depth trials in
which you participate fully, with all your intellect and heart, not
knowing what the final result will be, but knowing that you will
be different when you come out the other end. This inter-
relationship with the environment actually makes you grow, sur-
vive and develop your potential.
Someone who undergoes a course of military training, for
example, will no longer think or act in the same way as before; the
same is true for someone who goes through the rigours of in-
depth professional school.
 DECISION MAKING AS A LEARNING ACTIVITY 77

Corporations also have a form of learning by accommoda-

tion — at least, successful companies do. As we saw in the previ-
ous two chapters, longlived companies find ways to respond to
signals of change in the business environment, by changing their
own internal structure. The same important increase in the inter-
est rate, which the bank management takes in its stride, has a very
different meaning for a heavily geared real estate company. It
should lead the latter to changes in its internal structure. It should
renegotiate its loans, divest part of its project portfolio, delay or
change the nature of individual projects, try to find a partner or
reorient its workforce.

Both these forms of learning, in the end, are successful precisely

because they are embedded in decision making. True decisions —
in which a new understanding is reached and an action taken — are
themselves examples of learning by accommodation.

The problems with conventional learning

If decision making is learning, then all companies learn all the
time. There is no need to ‘build’ a learning organization. You
already have a learning organization.
But the traditional time-honoured ways in which most com-
panies accomplish this learning are inadequate. The endless cycle
of meetings and discussions has some distinct disadvantages.

✦ It is slow. In one Shell study, we measured the speed of deci-

sion making when the decisions involved a change in the internal
structure of the company — shifting the product range, closing a
manufacturing gate or redesigning the organization. Such deci-
sions often took 18 months between reception of the signal and
implementation of the change. In some cases, it took five years or
 LEARNING 78

more between an item’s appearing on the agenda for the first time
and the final implementation of the required internal change.
Being slow is especially dangerous in a world of frequent
oscillations. We run the risk of reacting to the last disturbance or
‘fighting the last war’, when the next one is already coming
around the corner.

✦ It closes down options. Discussions on new business opportu-

nities or painful decisions on cutting parts of the business always
involve reallocation of resources. For example, if the company
contemplates the closure of a manufacturing site, or its move to
another country, the managers concerned feel threatened or gen-
uinely believe that they are asked to make a sacrifice. This brings
elements of negotiation into the decision-making process.
Negotiations, normally, can only have one outcome: the negoti-
ated or hierarchically imposed solution. This outcome becomes
the one option for which the company makes a single time path
into the future. It becomes ‘the plan’ or ‘the strategy’ and other
options are never explored.

✦ It depends on learning by experience, instead of by simulation.

This means that normal management requires constant experimen-
tation with reality. British Airways would never allow pilots to fly
a 747 without making them spend considerable time in a flight sim-
ulator; a pilot does not learn to fly an aeroplane by experimenta-
tion with a real plane full of passengers. Yet we find it acceptable
to let managers ‘fly’ their companies by trial and error.
Unfortunately, there are as many human fates linked to the deci-
sions of a manager as there are linked to the decisions of a BA pilot.

✦ It breeds fear. When we are asked to discuss an important

decision, in which fundamental change is involved, our minds
 DECISION MAKING AS A LEARNING ACTIVITY 79

tend to race ahead to think of the consequences. Fear of the risks

begins to permeate our thought processes. It cramps our imagi-
nation; innovative or adventurous options are often not seriously
considered.

As Piaget might put it, the preferred solution (in a fearful situa-
tion) becomes that of ‘assimilation’ rather than real change.
Managers continue to hope that the original distortion will be
only a temporary aberration. They make decisions that might
include cutting costs, cutting capital expenditure, cutting recruit-
ment or reducing the quality specifications of the products.
(Please note that these decisions are not always problematic. They
are perfectly good decisions if they are made because the compa-
ny has a poor position vis-à-vis its competitors, is overcapital-
ized, has too many people or suffers from some other
fundamental weakness. But if these decisions are made as a quick
reaction to a change in the outside world — for example, to a drop
in total demand — then they represent a solution of assimilation
where accommodation would be required.)
Decisions based on learning by assimilation can have dan-
gerous outcomes. If, indeed, there is a fundamental change in the
world outside, and managers sit there reassuring themselves —
‘Yes, it is a change, but life will return to normal and then we’ll be
leaner and meaner’ — then the risks are mounting rapidly. If pro-
longed, assimilation weakens internal systems. Cash flow
reduces. Employees walk away; so do customers. At some stage,
so do shareholders. If the original distortion was not an aberra-
tion, and it does not go away, then the company will slide into a
crisis with its internal systems weakened.
Fear also creates a preference for the repetition of previous
success formulas. ‘This reminds me of the situation we encoun-
tered 20 years ago,’ someone is bound to say. He or she goes on
 LEARNING 80

to describe the solution that worked then. The sigh of relief is

audible: ‘If it worked then, it may work now!’
History does, in fact, repeat itself — but never in quite the
same way. A 2 or 3 per cent deviation in conditions from the pre-
vious time to this time can make the difference between a good
decision and a mediocre one.
In short, the natural learning process tends to limit the
number of options — and it is slow. The speed, openness, inven-
tiveness and courage of our learning efforts must all be improved
for our organization to survive.
Thus the question before us is: how can we improve the
existing decision-making process? How can we accelerate and
fortify the learning which already occurs?

Learning to play and playing to learn4

My search for better modes of learning led me to the writing of sev-
eral noted educational researchers. The three sources whose work
helped most were British psychologist D.W. Winnicott (Playing
and Reality) from the Tavistock Institute, US teacher/writer John
Holt (How Children Learn) and the Media Lab’s Seymour Papert
(Mindstorms: Children, Computers, and Powerful Ideas), based at
the Massachusetts Institute of Technology.5
All three wrote primarily about classrooms and children,
but reading their books with companies in mind, one is struck by
the similarities. Classroom situations are easily turned into
boardroom situations, as we discovered before long in Shell’s
Group Planning.
All three writers had essentially the same theme: the essence
of learning is discovery through play. A decision-making process
which accelerated learning could do so only by making skilful use
of playing.
 DECISION MAKING AS A LEARNING ACTIVITY 81

The psychiatrists of the Tavistock Institute had known this

since the mid-1940s. They had been in charge of one of the most
massive training and learning efforts of the twentieth century —
the preparation of the Allied armies for the invasion of France in
1943. After the war, they formed the institute to continue their
efforts. From their pioneering work, we get a better idea of what
‘play’ really is and how it enhances the acquisition of knowledge
among individuals or groups.
D.W. Winnicott first published his book Playing and
Reality in 1971. In it, he coined the idea of a ‘transitional object’.
Play, he reasoned, is always conducted with a thing in hand: a toy.
Girls play with dolls, boys with Lego sets, and toddlers of both
sexes with Fisher-Price toys.
Playing with toys is very different to playing a game or
playing a sport. There is no way to win. The player is simply
experimenting with an object that in some way represents reality.
This brought home the difference between play and games. Play
is experimenting with a toy which the player accepts as repre-
senting his or her reality. This makes the toy a ‘representation’ of
the real world with which the learner can experiment without
having to fear the consequences. Underneath all the fun there is a
very serious purpose: playing with one’s reality allows one to
understand more of the world in which we live. To play is to learn.
Winnicott called these toys transitional objects because
they help the child to transit from one phase of life to the next —
from one level of understanding of the world to another. A girl,
for example, invests a doll with a part of her reality. In her mind,
the doll is her little brother or her friend. She experiments (plays)
with it. Because the doll is only a toy, she can do so without hav-
ing to fear the consequences. She can mutilate it and find out that
particular actions do indeed lead to mutilation. She can toss or
drop it, in ways her mother would never allow her to do with her
 LEARNING 82

real-life baby brother. In so playing, the girl acquires knowledge

about relationships and about cause and effect. She learns to begin
acting in real life at a higher level of development.
Businesspeople do something quite similar. When Shell (or
any other oil company) develops a new oilfield in the North Sea,
and a new oil drilling platform must be built, we do not experi-
ment with reality! We will not build the structure, put it in 100
metres’ depth of seawater and see what happens. Instead, we
build a scale model which we put in a model of the seabed. Then
we experiment with these scale models, sometimes for years on
end. We subject our ‘toy’ to all imaginable forces of waves and
wind and time, to see what would be likely to happen. Then, and
only then, we build the real thing.
Similarly, at a chemical company developing a new process,
models will be made and experimented with for months — at the
scale of the lab bench, in a pilot plant and sometimes at a sub-
optimal production facility, before the final design is constructed.
The same is true for a dyke built in the Netherlands, a massive
dam or any technical situation in which we do not want accidents
or victims.
The military, for which every error is a matter of life and
death, rarely moves without having played endless ‘war games’.
Many of the operations in the Gulf War of 1990, along with the
entire logistical supply chain, had been played out in computer
simulations before the moves were made in reality. Similarly in
business, whenever the risks are great, we do not experiment with
reality. We go through a protracted phase of experimenting with
a model, a representation of reality, before we enter ‘real’ life.
Even spreadsheet programs are ‘toys’ in this sense; managers can
simulate major financial decisions, trying different versions of
reality, before risking great sums of money in real life.
All of these examples should make one thing clear. We
 DECISION MAKING AS A LEARNING ACTIVITY 83

know extremely well in business that play is the best method of

learning. That’s why it never ceases to amaze me that, in most
business decision making, play is not even considered as a vehicle
for learning. Instead of simulating reality, we ‘learn from experi-
ence’ — we experiment with reality itself.
We perceive a problem. We put it on the agenda of the next
meeting. Come the day of the meeting, we do everything at once:
understand what this agenda item is about, think about what
might happen in the future around this subject, dream up an
action to be taken and launch its implementation. We do all this,
preferably in one meeting, with two more items to go through
before the two key people in the corner have to leave in a hurry
to catch their planes.
This Rambo style of management is quite pervasive. It is
mostly adequate, though not ideal, for operational decisions —
where knowledge and priorities fit in harmony with the outside
world. A bank reacting to an increase in interest levels, or a
wealthy company making comparatively small investment deci-
sions, can afford to make some decisions this way. However,
heroic decision making becomes a high-risk gamble in unhealthy,
persistent business situations where the only way out is an adap-
tation: closing a factory, a new direction for the research pro-
gramme, aborting a product launch.
We should therefore not be surprised that so many fatal
mistakes are made in the change management of companies, or
that so many managers have so little time to think because they
are constantly engaged in firefighting. To me, this prevalence of
mishap is a strong argument for running our companies with the
same low tolerance for error in management that we currently
employ in the more technical parts of the business. As many, if
not more, human fates ride on the quality of management’s deci-
sions in a takeover, merger, plant closure or product change as
 LEARNING 84

there are at risk on a North Sea oil platform. A marketing manag-

er in even a relatively small company, such as Shell Kenya, influ-
ences many more people through his or her decisions than an
airplane pilot does with 350 passengers in a 747 aircraft. We should
not be so willing to take more risks in non-technical work simply
because the results of our errors are less immediately violent.

Building the manager’s Lego set

Peter Schwartz, the head of the Shell scenario team during the
1980s, tells the story of how he was walking down the street in
Palo Alto, the California university town. Suddenly, a stranger
crossed the street and walked up to him.
‘How’s the future?’ the stranger said. When Peter looked
blankly at him, the stranger said, ‘You’re Peter Schwartz, right?
You write about the future, right? You’re interested in computers
and kids, aren’t you? You ought to read a book called Mindstorms
by Seymour Papert. You’d really like it.’6
The stranger walked away and Peter crossed the street (in
Palo Alto, all you have to do is cross the street and you’re in a
bookshop) and bought a copy of Mindstorms. Then he brought it
back to Shell, where it gave us the next step in developing transi-
tional objects for managers.
Papert was using the personal computer, which was still
only a few years old at that time, as a transitional object for learn-
ing by children. Using LOGO, a computer language that Papert
had helped invent, children could program a mechanical ‘turtle’
(or a turtle on the screen) to move in various directions or to draw
patterns. The world of turtles became, as Papert called it, a
microworld, which in turn helped children learn about the real
world. The experience of LOGO programming was as close to pure
play as one could imagine. For example, by controlling the tur-
 DECISION MAKING AS A LEARNING ACTIVITY 85

tle’s motion, using commands to set position and velocity, the chil-
dren developed a visceral, in-depth, ‘accommodated’ (in Piagetian
terms) understanding of concepts in Newtonian physics such as
the laws of motion. It was very easy to read Seymour Papert and
see his microworlds as another form of transitional object.
From Papert’s work, we concluded that we could put a rep-
resentation of reality into a computer and use that ‘play’ dynam-
ic to build depth of understanding among Shell managers. We
resolved to experiment. We brought in a computer scientist, a
young university graduate, who developed a model of the oil sup-
ply chain.
Unfortunately, the model was quite simplistic. Managers
turned on their computer screens, looked at it briefly and put it
aside. We learned a valuable lesson: managers are much less pliable
than children. Children have an immense capacity for imaginative
substitution: they can invest their reality in almost any object and
then play with it. But we seem to lose that capability with age; as
adults, we only want to play with ‘toys’ that meet reality halfway
and are representative enough to be recognized as reality.
So we tried to put the managers’ multidimensional world
on to a two-dimensional computer screen. We thus found our-
selves dealing with the problem of mapping — the same problem
that cartographers have when they must express the complex
reality of a country with only the dimensions of printed ink on
paper. To solve a mapping problem in cartography, one needs to
develop a simple, consistent set of symbols: black lines for rail-
ways, red lines for motorways and blue lines for rivers, for exam-
ple. We needed a similarly simple set of symbols with which to
map business problems.
When Pierre Wack heard about this stage of our quest, he
said, ‘Why don’t you talk to Jay Forrester?’ Under the name of
system dynamics, the MIT management professor Jay Forrester
 LEARNING 86

had developed a small set of simple symbols, built into a comput-

er programming language, that was perfectly capable of mapping
any business situation. When we visited Forrester at MIT, it
became clear that his programming language, then called
DYNAMO, was well suited for building models with the requisite
complexity.
Unfortunately, computer modelling was so complex that
there was no accessible real-time modelling available. Model mak-
ing required teams of highly trained specialists, who needed
months to make the model run, and by the time they came back the
situation had changed and the model no longer represented reality
for the managers. We spent months trying to find a computer mod-
elling system that we could use for real-time models: models pro-
grammed in the presence of the managers’ group, incorporating
their ideas and perceptions, to overcome their mistrust. Anything
built outside their presence was regarded as a ‘black box’ — put one
answer in, get another out and put it away. There was nothing
compelling about it and they didn’t want to play.
Building a DYNAMO model also required months; but one of
Forrester’s graduate students, Barry Richmond, had begun to cre-
ate a new piece of software called Stella. Richmond said that it
was 100 times more effective than DYNAMO, and we at Shell were
inclined to agree with him.
With Stella (later known as iThinkTM),7 we could build
8
microworlds of our own business — computer environments that
showed, for example, the key variables in an oil supply chain
(such as, for example, ‘producer price’ and ‘consumer demand’),
with their interrelationships defined as formulas that fit the way
they interacted in reality. We could enter decisions that a manag-
er might make (such as increasing certain types of supply alloca-
tions) and see how the results might shift over the course of
months or years. And, as our understanding of the real supply
 DECISION MAKING AS A LEARNING ACTIVITY 87

chain improved, we could make changes in the model, to make it

conform in ever more sophisticated ways to the reality around us.
Here, at last, was modelling software that could be managed in
real time.
System dynamics turned out to be ideal for bringing out the
mental maps that we had sought to influence with scenarios. We
would enlist managers to help articulate their business problems
and introduce them to the specifications of the computer soft-
ware. Forrester and his colleagues in the system dynamics group
had developed a few simple symbols which permit the modeller
to draw complicated business situations: arrows for lines of influ-
ence, squares for ‘stocks’ of various types of capacities (the
amount of oil held in our tanks, the level of current staff), circles
for ‘flows’ that governed the rate of change (the sales rates or rate
of hiring) and feedback loops that suggested how a part of a sys-
tem would either continue to accelerate its growth or move
towards equilibrium.
Up to now, the computer models that Shell managers knew
(and usually deeply distrusted) were the linear models of physical
situations like a refinery or a transport fleet. They were used to
calculate refinery optimizations and the like. But these models
were different. Based on non-linear equations, they described the
evolving causal relationships embedded in complex business situ-
ations as they took place over time. One model, set up for the
management team, demonstrated the value of establishing a com-
prehensive in-house oil commodities trading floor. Another
model helped the management of a small biotechnology firm in
the Shell Group chart its strategy options. A new automotive
retail policy was developed in the Netherlands, as was a new nat-
ural gas strategy after the oil price falls of 1986.
What difference did the models make? We set out to answer
that question. Because one cannot measure the quality of
 LEARNING 88

decisions, we decided that we would measure speed: how long did

it take to get from the perception of a changed external reality to
the implementation of a fundamental shift of operation? When
we measured it, we tended to find that the process of learning had
accelerated by a factor of two or three; it was now two or three
times as fast to implement a new internal system. The shift to a
full, comprehensive trading floor, for example, was a response to
changes that had taken place in the international oil market. The
vertical integration which had held sway for 40 years, under pres-
sure from nationalizations in the Arab world, was disintegrating
fast. There had been spot markets in oil trading before, but oil
companies did not immediately recognize the need to shift their
managerial approach: from optimizing the flow of oil within the
company only to being willing to say, ‘Every drop of oil that I
have is in principle for sale — not just to our own companies, but
to anyone.’
Shell was not the first company to make this shift; British
Petroleum set up a full-scale in-house oil trading floor well before
we did. Nonetheless, Shell made the switch in good time for our
own survival. This was aided by a Stella model, with which top
managers ‘played’ during the early 1980s. Since they were respon-
sible for the decision, the learning of Shell as a company depend-
ed on their perceptions. Thus, when the managers recognized the
shifting environment, the institution itself perceived a shift. From
that moment, Shell was reoriented to create a trading floor and to
develop the necessary support for the floor. It took us only six to
seven months. In the past, similar decisions had taken 18 months
or more. Within the following year or two, the amount of oil
traded in that system rose to become 40 to 50 times as much oil
as physically moved through the real system of refineries and
tankers.
 DECISION MAKING AS A LEARNING ACTIVITY 89

What computers couldn’t do

Despite these successes, we began to see, during the late 1980s,
that system dynamics models were not a panacea. They were par-
ticularly problematic in the first phase of every team learning
exercise: the capture of the mental models of the individual team
members.
There were three parts to this task: capturing the (often
unarticulated) understanding that a group of managers held about
their world; mapping this understanding in a visual, two-
dimensional form on the computer screen; and creating the com-
puter model in real time, with the managers present. When all
three steps came together, managers could see with their own eyes
that the symbols on the screen were commensurate with the
views they had expressed in words.
But the process often didn’t come together. More often than
not, when we began to combine two or three managers’ explana-
tions in the same model, calibrated against one another, we saw a
reversal to the old, suspicious attitudes about computer ‘black
boxes’. Those meetings were fiascos. It would be impossible to
get the group of managers to start playing. Instead of simulating
their experiences and learning about external reality, they became
critics of the model. They spent hours querying its assumptions,
pointing out omissions, criticizing the modelling techniques —
anything but learning about their environment!
I often wondered, in those days, why children had so much
more imagination than results. They were prepared to play with
toys which weren’t exact representations of reality; they knew
they could learn from those toys. Grown-up managers wanted
the model to resemble not just reality, but their own assumptions
about the external reality. If in doubt, they would refuse to play.
 LEARNING 90

Other companies have had similar experiences. Often, the

question ‘Whose model is it?’ takes precedence over the question
‘What is this model saying?’. Peter Senge and some of his MIT
colleagues, for example, developed a set of models of the claims
adjustment processes of the Hanover Insurance Company in
Worcester, Massachusetts. I have seen some of the Hanover peo-
ple play this. An element of competition immediately creeps into
the play situation: learning is only a by-product. Managers have
little concern for understanding the situation described by the
simulation. Their concern is to score more points than the boss.
They want to finesse the game instead of learning what the game
can teach them.
Clearly, the computer itself was getting in the way of our
primary purpose: understanding the system. We were asking
managers to get to grips with several unfamiliar processes in one
meeting. First, they had to talk together about their assumptions,
breaking the hallowed traditions of corporate meeting behaviour.
Second, they had to look at symbols on a computer screen, put
there much too quickly by some young computer expert. And
third, they had to put up with the on-screen spaghetti and accept
this as a reliable representation of the discussion. We were asking
too much.
So we eliminated the computer. Instead, we moved to a dis-
tinctly low-tech technique: noting our ideas on magnetic,
coloured hexagons, which are placed on a whiteboard so that
everyone can see what is written on them. We then cluster and
rearrange the hexagons at will to show related concepts or con-
nections between ideas.9 There are other ‘soft mapping’ tech-
niques; indeed, a great deal of interesting group dynamics
research is going on in the UK in this area of soft modelling and
cognitive mapping.
Even teams of very senior managers react positively to soft
 DECISION MAKING AS A LEARNING ACTIVITY 91

modelling techniques such as the hexagons, particularly at the

early stages of mapping a team’s mental models.10 We have found
it useful to move from there to a second phase, in which the
team’s concepts are converted into a systems model. The com-
puter models give the team its only way to discover what Jay
Forrester calls the ‘counterintuitive consequences’ of its actions —
the long-term, unanticipated, far-flung results of internal policies
and options. The computer reveals the underlying relationships
and dynamics of a business situation in a far superior way to sim-
ple soft modelling. But the managers do not need to become com-
puter scientists in the process. One does not learn how to ride a
bicycle by designing one; and a management team learns about its
environment, not by designing the computer representation, but
by using it.
Accepting this view means facing the difficulty of convert-
ing the shared hexagon map (or a team’s causal loop diagram, as
described in The Fifth Discipline Fieldbook11) into a fully quanti-
fied computer model. There is a long way to go before this prob-
lem is solved. Management teams, especially senior ones, grow
impatient with this process; they do not want to sit there while
their perceptions are quantified into the computer model. Thus
the temptation always exists to dismiss the team and hand over
the flipcharts and hexagons to a computer modeller, who will put
the project together in a back office. Neither managers nor com-
puter modellers seem to wish for anything better.
This temptation should be resisted. Any model emerging
from the modeller’s back office runs an unacceptably high risk of
being rejected by the team. The modeller simply doesn’t know
what the managers know; the modeller can’t make the model rep-
resent their understanding of their reality. I believe that the solu-
tion to this problem ultimately lies in designing a computer
language similar to Seymour Papert’s LOGO. If a computer
 LEARNING 92

language can be created which is simple enough to allow six-year-

olds to design their own microworlds, it ought to be possible to
create a language for managers.

Why don’t we play all the time in business?

I hope I have shown in this chapter how the decision-making
process is in fact a learning process in any company and that there
are ways to improve the speed, if not the quality, of the decisions.
The more in-depth the simulation, and the more that ‘play’ trig-
gers the imagination and learning, the more effective the decision-
making process seems to be. In companies which attempt
large-scale internal change, this is particularly true. Decisions can-
not be made in the old authoritarian manner. They need inter-
action, intuitive reflection and the fostering of collaborative
mental models. They need play. They need learning.
Despite its reasonable nature, this is a hard message for
managers to hear. It goes against the traditional way in which
most people look at their career position. They do not think
about their job in terms of learning. Intellectually they may agree,
but they still feel that their leadership depends on their ‘know-
ing’ — their ability to project self-assured confidence in their own
information. The corollary notion, that the best way to learn is
through play, makes the message even less palatable.
This was certainly the case in the mid-1980s, when I had
been planning coordinator for about five years. I still did not find
it easy to talk to my colleagues about decision making as a learn-
ing process, or about learning as play. Finally, I dared to give a
talk about the subject at a Shell planners’ conference in 1986, in
the distant village of Banff, Canada. That talk was well received,
but it was still not easy to discuss the subject at top management
levels in London or The Hague.
 DECISION MAKING AS A LEARNING ACTIVITY 93

Even the phrase ‘I learned’ was inadmissible in many Shell

circles — which made it very difficult for people to enter into the
kind of colloquy that would help us improve our decisions. To
have to ‘learn’ something meant that you didn’t know it in the
first place and (particularly for those of us who came of age in the
1950s) it was considered much better to lie and give an answer —
any answer — than to admit you didn’t know. This attitude is still
ingrained in many companies — sometimes deeply ingrained.
In the end, one of my colleagues, Napier Collyns, advised
me that I would find a more attentive audience if I talked ‘from
the outside in’. At his suggestion, I wrote a Harvard Business
Review article, published in 1988 under the title ‘Planning as
Learning’. The article stayed prudently in the area of planning,
and only touched obliquely on the subjects of this book: decision
making and the nature of companies. And it worked! It was now
possible to hear the word learn in speeches from senior managers
at Shell, and to develop more experimental approaches to man-
agement decisions. Since then, the debate in the Shell Group has
made important steps forward, and gradually there is a wider
application in more fields of the theories and practice of institu-
tional learning.
Ultimately, if companies do not embrace the hypothesis of
accelerating learning and the concept of play, they will suffer the
serious, long-term effect of learning more slowly than their com-
petitors. Thus the living company needs to find a way to cope
with inbred resistance to seeming ‘incorrect’. Something must be
done to make the members of the company feel secure in the
company’s identity, so that they can reveal themselves safely and
speak up with impunity. These are matters which move beyond
learning; they have to do with building a coherent identity.
Part II
Persona (Identity)
5
Only Living Beings Learn
WHEN WE UNDERSTAND THE CONCEPT OF PLAY, WE BEGIN TO
see how companies can learn beyond the learning of individ-
ual managers. Imagine that you have a team embarked on a steady
regime of ‘play’ within a company. People regularly join the team
or leave it to pursue other projects, so that, by the end of three or
four years, no original team members are left. Yet the team’s capa-
bilities continue to improve. The calibre of the projects the team
takes on continues to improve, because each new team member is
brought along to experience the same quality of learning.
Now imagine that there are similar teams throughout the
company. People still join and leave the company, but the high
calibre of learning continues because the practice of play is insti-
tutionalized. Outsiders, watching the company’s financial and
social results, are moved to exclaim, ‘It’s amazing how this com-
pany has developed! It is far more capable, and far more resilient,
than it used to be!’
What is the entity they are talking about? Is it the set of
teams that are learning? Is it simply the sum of the managers who
are active in the company? Or is this greater capability embodied
in the company’s factories, oilfields, ships and trucks? In other
words, does the balance sheet reflect the company’s ability to
 PERSONA (IDENTITY) 96

develop its capabilities? Why is the company worth more or less

than it was a decade ago? Is this change inherent in the assets on
the balance sheet? Is it possible that a bundle of assets can learn?
Assets are just dead objects. They have nothing to do with
the innate spirit that moves and propels a company. Nor is a com-
pany just a bundle of individuals or a combination of assets and
individuals. We have already seen how companies can survive the
loss of both assets and individuals, and more, and still keep their
essential nature intact. Thus it is impossible to talk about organi-
zational learning without trying to think about every company as
a living being. But this is an unusual line of thought. We expect
living beings to have cells and bodies; to be born, die and repro-
duce; to take in nutrients and give out waste. Do corporations,
with their abstract, legally created bodies, do anything of the
sort? If a company is a living being, what is that being like?
These questions became important to me around 1971. That
was the year that I reached a personal turning point — a moment
of intense soul searching that, in retrospect, defined the rest of my
working life and career.
I had just been appointed as a director of Shell Brazil. But it
was an inauspicious time to be placed in the top management of a
major oil company in South America. Two years after I came to
Brazil, the world was in the throes of the oil crisis. Embargoes
came into effect against many countries. Supplies were curtailed.
Long lines of cars with frustrated drivers lined up on the fore-
courts of service stations — primarily in the US and Europe, but
also elsewhere in the world.
At that moment in its history, Brazil was doing relatively
well. The country was in the middle of the ‘Brazilian miracle’: 15
years of uninterrupted economic growth. This growth required
lots of oil, which had to be imported. But Brazil’s government-
owned oil company, Petrobras, had been canny. It had acted
 ONLY LIVING BEINGS LEARN 97

quickly to set up supply lines. Since the oil kept on coming the
economy kept on growing, and Brazil’s exports continued to pay
for the ever-increasing price of oil imports.
Nonetheless, the oil crisis helped rekindle Brazil’s national-
istic fires, always easy to bring to flame. Years before, under the
slogan ‘o petroleo è nosso’ (‘the oil is ours’), the petroleum indus-
try had been all but monopolized in government hands.
Exploration and production of oil within Brazil, as well as refin-
ing and transport, were handled exclusively by Petrobras. Private
companies (like Shell) were grudgingly permitted to bring in oil
imports and to sell oil in the domestic market. But now Petrobras
had beaten the oil companies at the supply game; it had succeed-
ed, where they had failed, bringing in the oil which the country
needed for its growth.
Hence a chorus of voices emerged, asking to remove the for-
eign companies from their last foothold in oil retailing. Petrobras
could do better. In addition to Brazilian nationalism, the chorus
drew on an increasingly strident worldwide theme. In the court of
public opinion, multinational corporations were the culprits
responsible for the ‘outrage’ of the oil crisis. Individuals around
the world in those days were asserting their rights against large-
scale universities, armies, governments and multinational corpora-
tions. Small had been declared beautiful. From the point of view
of this rhetoric, it was easy to see that large global oil companies,
bigger than many nations, had power which eclipsed mere sover-
eign control. They used (or, rather, abused) their power, creating
artificial shortages for their own egoistic purposes and manipulat-
ing the markets to increase their already obscene profits.
Individuals were thus deprived (said the critics) of the natural right
to move around in a motorcar whenever they wished.
At that time you could hear such remarks coming from the
mouths of politicians, in reports in the press, in United Nations
 PERSONA (IDENTITY) 98

communiqués and even in the voices of some of my friends and

acquaintances. The scorn hit me hard. By then I had already been
working more than 20 years for one of these multinational oil
companies. My father had worked for the same company. Where
I came from, gaining employment in such a solid, large-scale
organization had always been an occasion for celebration, not
calumny. Moreover, I had worked for Royal Dutch/Shell in
Europe, the Middle East, Africa and now South America. I had
come to know hundreds of my colleagues at nearly all levels of
the hierarchy. I did not recognize myself or my colleagues in the
descriptions the press and politicians were giving.
I could, to be sure, write off many of the critical comments
as cynically self-serving political statements. But there was no
doubting the depth, or the sincerity, of the emotions which peo-
ple expressed.
My other colleagues, who worked in these maligned multi-
national institutions, felt similarly startled, hurt and misunder-
stood. From childhood onwards, we had joined a wide variety of
institutions: churches, clubs, trade unions, professional organiza-
tions and, finally, companies. Why would we be acceptable as
members of a church or a club, but socially suspect because we
worked in a large company?
Before long, however, we realized we had nothing to fear.
We could protect ourselves by remaining silent and keeping the
company below the parapet of public attention. The attacks were
all focused on the company, not on the individuals involved in it.
The company was seen (and caricatured) as a unit in its own right,
with its own purposes and its own characteristics. Outsiders
could not fathom its intentions and machinations — particularly
when a company like Shell had foreign roots, seemingly not sub-
ject to the control of the national society (of Brazil, in this case).
To outsiders, a company such as Shell had the mysterious power
 ONLY LIVING BEINGS LEARN 99

of making its employees do things which those individuals would

never do acting in their own right.
It all added up to an image of Shell as a sort of giant phan-
tom in a forest — difficult to see, with no specific contours, but
with enormous, uncontrolled, undefined powers that might well
do us harm. Moreover, it was a silent phantom. The multinational
entity told no story about itself. Outsiders could only guess why
it existed, what it did for a living, why it came to Brazil, and how
the world would be different if the company did not exist at all.1
Many managers at Shell, and at other multinational compa-
nies, ignored this outsider’s perspective; they were too busy man-
aging the company, perhaps, to pay attention to the rantings of
politicians and the press. From my Brazilian perspective, however,
it was impossible to ignore. It made me wonder whether the stereo-
type of me, as a Shell man, was true. Was my personal identity
characterized by the company for which I worked? Or was the
company shaped and formed by the individuals who worked for it?
It seemed to me, on reflection, that neither was true. The
institution was not a creation of its current members. It was a sep-
arate entity, a persona in its own right. It had its own character
and history.
Joining the institution therefore meant a certain element of
submission to a set of views and beliefs, which I might not have
taken on as an individual — just as joining a church, a trade union or
a political party might lead one to support a set of practices and atti-
tudes. But joining a corporation, any more than joining a church, a
trade union or a political party, did not mean surrendering my
capacity for judgement or critical opinion. It did mean, however,
that I would have to learn to exercise my judgement as a participant
in a large, collective endeavour; my voice and the entity’s voice
might be distinct, but they would not be separate. For as long as I
belonged to Shell, I would be associated with the Shell entity.
 PERSONA (IDENTITY) 100

I had been half-insulted by the Brazilian press and politi-

cians. I felt they were accusing me of things I had never done and
accusing my father and colleagues of the same things, by associa-
tion. But now I realized that I wasn’t being accused; the Shell
entity was on trial, not me. At the same time, I was part of the
Shell entity. I racked my brains to discover why we hadn’t got a
clear answer. Managers at some of our competitors clearly felt the
same way. This was the era, for instance, when Mobil Oil began
to write and publish its argumentative newspaper ads, airing the
company’s point of view on oil politics, international trade and
environmental issues.
But Shell never created such ads. We talked about the pos-
sibility and it became clear in our conversations that, although we
had some answers as individuals, there was no answer available
from the institution. We were part of an entity larger than our-
selves. If we were going to come up with a fitting response, we
would have to find a way to express the entity’s needs and spirit.
We would have to find a way to develop a healthy relationship
between the entity’s persona and its environment.

The persona of a living being

The concept of an entity’s persona had already been part of my
education for many years. I was introduced to it in my college
days in the 1950s, through the work of a significant German psy-
chologist named William Stern. Although he is largely unknown
outside continental Europe, Stern was one of the founders of
developmental and learning psychology — best known, perhaps,
for developing the intelligence quotient (IQ) formula. But it was
his more philosophical writing that had the most impact on my
understanding of the living company.
I was introduced to his ideas through an unusual path of
 ONLY LIVING BEINGS LEARN 101

study. Most business academics, particularly in the 1950s, began

with the science of economics. This science, still recovering from
its nineteenth-century struggle to gain scientific respectability,
was focused on measurability, predictability, causality and unam-
biguous answers. Academic credence, after all, could be acquired
only by copying the approaches of hard sciences such as physics.
Thus, when economists talked about human behaviour as part of
their theories, they postulated a mythical creature, homo econom-
icus: a perfectly rational person who always operated from self-
interest, with clearly defined reasons for every action and
decision. Economic theory could thus encompass sophisticated
formulas to describe complex, large-scale, aggregated activities,
which then could be translated into ‘managerial science’.
But the formulas said very little about the actual behaviour
of homo sapiens — which is immeasurable, unpredictable, unfath-
omable and deeply ambiguous. Even a 19-year-old student like
me could tell, from my part-time job at a Shell refinery, that,
whereas the management curriculum had no place for human
beings, the workplace was full of them.
Indeed, five years of German occupation had made the
workers of the Netherlands into great masters of passive resistance
to the pressures of autocratic power. The business results of Dutch
companies could clearly be affected by recalcitrance, passive resis-
tance and the lack of active cooperation. Why, I wondered, why
was so much time spent at university in understanding systems
and figures and so little in understanding humans?
Seeking an answer, I opted to include general psychology in
my doctoral studies. I was permitted to do this because, in addi-
tion to our major studies such as economics, finance and organi-
zational structure, we were permitted two ‘lighter’ electives. By
some quirk which I never understood, a subject called general
psychology figured in the list of electives. It was taught by a Dr
 PERSONA (IDENTITY) 102

Van der Spek, the head of a nearby psychiatric institute. He

seemed as surprised as anyone to find in his charge a student from
the economics school, with no background in psychology and lit-
tle time to follow courses (because my job at the Shell refinery
kept me occupied with nearly full-time hours). The only way to
combat ignorance was to give me a long list of books to read;
thus, for the next year, I read my way through the history of psy-
chological thought, through characterology and on into modern
schools like the existential psychology of Jean-Paul Sartre.
In this journey, my attention was caught, time and time
again, by references to a school of thought called Personalismus,
founded by William Stern. Stern became, with his wife Clara, one
of the pioneers of child psychology. Together they operated a
clinic and published a classic work on the language of children.
He was one of the founders of a new university in Hamburg after
the First World War and was a forerunner to Jean Piaget. Then, in
1933, he and his wife were among the first victims of the Nazi
prosecution. Their clinic was closed; their books were banned
and burned. The family took refuge in the US. Stern died in
Durham, North Carolina, in 1938, five years after he was forced
into exile.
As far as I know, Stern never wrote in English. That explained
why, despite his contributions, his name was little known. Outside
Germany, few could read his papers. Within Germany, after 1933,
his name, books and reputation were wiped out. Only nearby
Holland preserved his influence; there, with a sufficiently wide-
spread knowledge of the German language, a group of academics
seriously studied his ideas after the Second World War. His last
book, General Psychology on a Personalistic Basis, was republished
in 1950, in the German language, in The Hague.
I was drawn to Stern’s thinking because of his systemic way
of looking at human beings. Born in 1871, he had built a career in
 ONLY LIVING BEINGS LEARN 103

a discipline which, like economics, had to fight hard to earn the

predicate ‘scientific’. As with economics, academic respectability
could be acquired only by copying the approaches of hard sci-
ences like physics. Psychologists eschewed synthesis and general-
ization for analysis, specialization and narrow definitions of their
problems and studies. This had led to the establishment of many
specialized fields of study within psychology: Gestalt psycholo-
gy, behaviourism, depth psychology and more. All of these
focused on parts of the human being: the subconscious, the reflex-
es of behaviour, even the soul (as distinct from the body).
Complicated phenomena, such as ‘seeing’, were studied in the
same ways that physicists studied wave bands and electric cur-
rents. Fields of study were kept narrow, so that experiments
could achieve repeatable, predictable results. And there were,
indeed, important experimental results. But, just as with econom-
ics, conventional psychology had very little to say about the peo-
ple at the Pernis refinery, where I worked after university, and
where so much understanding was needed.
Stern, by contrast, was trying to develop a systemic theory,
a theory that would encompass the behaviour of people in the
refinery, the Shell system around them and the relationships
which bound them together. As one fellow psychologist, R.B.
McLeod, put it:

For [Stern] psychology was simply a road leading towards an

understanding of man, and ultimately towards an understand-
ing of the universe ... Stern preferred to regard his psycholo-
gy as a special branch of a more general science of
Personalismus, a science which he hoped would one day unite
within a single system the findings of all the sciences of man,
the physiological, the psychological and the cultural-
2
historical.
 PERSONA (IDENTITY) 104

To Stern, each living being has an undifferentiated wholeness,

with its own character, which he called the persona. A living being
cannot be understood unless that persona becomes evident. The
persona, in fact, is the essence of the living being. It is part of the
larger world, but separated by its ‘membrane’ from the larger
world — made distinct as a miniworld in itself, with its own values
and experiences. The persona represents body and soul together.
It has several key characteristics:

✦ The persona is goal oriented, said Stern. It wants to live as long

as possible and to realize the development of its potential from its
talents and its aptitudes.

✦ It is conscious of itself. A persona can perceive itself as ‘I’,

although it is composed of parts and elements which are personæ
in their own right. In its turn, it can be a part of a larger entity, as
the soldier is part of a platoon, the platoon is part of a company,
the company part of an army and the army part of a nation’s
armed forces. Stern condensed this concept into a Latin expres-
sion: ‘a persona is a Unitas Multiplex, a structure of structures.’3

✦ It is open to the outside world. Elements from the outside —

such as food, bacteria, dust, light and sound vibrations — con-
stantly enter the human system. But human individuals and their
ideas also constantly enter higher-order personæ such as a com-
pany or a corporation. At the same time, a persona is in constant
relationship with the outside world, in the sense that every expe-
rience represents one more exchange in a lifelong dialogue with
the forces of the world around it.

✦ It is alive, but it has a finite lifespan. One day it is born, and

one day it will pass away.
 ONLY LIVING BEINGS LEARN 105

In 1919, sixty years before the popularity of terms like living sys-
tems, human potential and holistic health, Stern proposed look-
ing at an entity’s behaviour, sociological environment,
psychological history and, by extension, economic life as compo-
nents of one existence, all interrelated.
This resonated with me. I had hoped to find some under-
standing of the human being who, in my economics textbooks,
had been abstracted to a lifeless marionette, the homo economicus.
Stern said instead that these lifeless marionettes were not people
at all. The central argument of Stern’s book Person und Sache, in
fact, was the nature of the distinction between persons and things.
‘Things’ are all the dead objects in the world, objects without a
will or a life force. Things are affected by events, but only persons
decide to make things happen.
When the temperature goes up, the rock on the mountain-
side expands. When it freezes at night, the rock contracts. It may
crack in the process, but in a very short time the rock is back in
harmony with its cold environment. In the world of things, the
world without a will, causality is measurable and repeatable. The
same force, acting on a thing in the same way under the same
conditions, once or a million times, will produce the same result.
Homo economicus was thus a thing. It did not have a goal or a
will; it existed only to react to other forces (such as supply and
demand). The forces of the outside world might affect homo eco-
nomicus once, or a million times; under the same conditions, the
result will always be the same.
By contrast, the members of homo sapiens — the real people
who worked in real jobs or attended real schools — are unfath-
omable precisely because they are wilful. They act towards their
own purpose, which economists cannot predict.
A living entity, such as a human being, is not merely a pas-
sive object, buffeted by outside forces. As people, we make
 PERSONA (IDENTITY) 106

choices. Our behaviour cannot be explained solely by cause-and-

effect relationships. One cannot really say, ‘Such-and-such a
thing happened to my next-door neighbour, and therefore, of
course, he [or she] reacted in such-and-such a way.’ No one can
guarantee how any of us would react in any given circumstances.
Our individual behaviour can be explained only by understand-
ing the internal force of our goals and teleological drive, together
with the forces coming from the outside environment. Even the
same forces, striking twice under the same conditions, might not
elicit the same reaction, because our internal goals might change.
(We might, for example, have learned enough from the first strike
to react differently the second time.)
Absorbing this insight from William Stern was very helpful
to me during the years that followed. I cannot point to any spe-
cific decision that emerged from it, but it coloured every decision
I took and every move I made. I knew that I could not count on
people simply to follow the rules, as if they were rational crea-
tures. I knew that I could never predict their actions simply
because they were alive. Although this unpredictability made the
risks of business greater, it also meant that I could tap potential
rewards that would otherwise be closed to me. It meant that peo-
ple would achieve, under encouraging conditions, great leaps of
invention and activity that homo economicus could never even
dream of — if, that is, homo economicus could dream at all.
But the events in Brazil in 1974, twenty years after my
introduction to Stern, shook me up even further. As a result of
my ruminations, I came to realize that Shell, as a whole, was also
an unfathomable being. It, too, was alive. It, too, had potential
rewards that could be tapped. Royal Dutch/Shell was not a thing.
In the sense articulated by William Stern, Royal Dutch/Shell was
alive.
 ONLY LIVING BEINGS LEARN 107

The ladder
William Stern, as it happens, had anticipated my insight. In his
work, he described a metaphorical ladder which might, if you
drew it, look something like this:

Stern, of course, was writing this in 1919. Today, we would prob-

ably add some rungs below the level of individual to include, per-
haps, ‘body subsystems’ and ‘cells’. Everything in the ladder,
larger or smaller than a human being, is a persona in its own
right — or, as Stern called it, a unitas multiplex.
Stern said that a living being always has a hierarchical struc-
ture. The ladder is the expression of that hierarchical structure.
There are always smaller components within our personæ, and
we are always components of greater personæ than ourselves. We
are a unitas multiplex, he said; we are one, looked on from the
outside, and subdivided as seen from within.
In Stern’s view, human individuals stand somewhere in the
middle of the ladder of personæ. Below them might be sub-
systems of the human body and even cellular identities. Actually,
 PERSONA (IDENTITY) 108

Stern was not specific about this; but more recent researchers,
such as biologist/cybernetician Francisco Varela,4 have shown
how subsystems such as the cell are actually organizations them-
selves, with collective entities and purposes.
Above the human being on the ladder of personæ are the
collective organizations and institutions through which people
join together. The family, the tribe and the national government
are all living systems in which people join together; so are the
trade union, the sports club and the non-profit organization.
These living entities nest within each other, like Russian
dolls. From the outside, one sees a large unit: Royal Dutch/Shell
or the Catholic church (to give two examples). The newspapers
speak of ‘Shell’s activities in the North Sea’ or the policies of the
Catholic church. And this point of view is valid; there is a Royal
Dutch/Shell persona and there is a persona for the Catholic
church. But that point of view is incomplete. For, seen from with-
in, the order of the Jesuits is a living system on its own inside the
Catholic church. Shell Brazil is a living system inside the Shell
Group. Both the Jesuits and Shell Brazil are driven by self-
preservation and self-development, as much as the higher-order
systems of which they are a subordinated part.
Further down the ladder within each of these entities, one
sees a variety of individual people, each with individual goals,
striving towards survival and self-development. The individual
people are often symbiotic, but equally often they end up clash-
ing with each other. Each has different characteristics and
potential.
A company like Royal Dutch/Shell, in short, has its own
ladder of personæ, looking something like the one opposite:
 ONLY LIVING BEINGS LEARN 109

Each unit of a company, at its own level, is a living system. Each

is distinct, visible and self-determined. At the same time, each
unit is embedded in the larger whole.
Consider how these units match William Stern’s criteria for
a living persona:

✦ Each of these units is goal oriented. Each is driven towards self-

preservation and self-development. No matter what individual
managers might feel, each business unit continues to act to pre-
serve itself and expand its scope of activity.
As with an individual person, the behaviour of a corpora-
tion (or of its units) cannot be explained solely through external
cause and effect. For example, one cannot say that Japanese com-
petition or lower profits caused General Motors to react by
changing its product line. GM’s behaviour, in the face of compe-
tition or lower profits, can be explained only by understanding
 PERSONA (IDENTITY) 110

the various living systems within GM. Perhaps Cadillac changed

its product line one way, because of its own goal to sell more
expensive cars, while Chevrolet responded in a thoroughly dif-
ferent way, because of a goal of better-quality cars. Or, in its
response, GM might merely have been part of a larger response
by the US automobile industry. Whatever happened, GM’s
behaviour is unpredictable; no one can guarantee that events
would play out the same way a second time.

✦ Each unit of the company is conscious of itself. Everyone asso-

ciated with any of these units is aware of its boundaries, of who
is included and excluded. The fate of Shell Brazil, for example, is
clearly linked to the fate of its customers, its suppliers, its retail
station franchise holders. But no one assumes that Petrobras,
which is Shell’s major supplier in that country, is a part of Shell
Brazil.

✦ Each unit is open to the outside world. People and ideas con-
tinually enter and leave a company, in the same way that a human
being exchanges information and material through his or her
digestive system, pores, eyes and ears. Just about every corporate
experience, from mundane transactions to intricate scenario exer-
cises, represents one more exchange in a constant, lifelong dia-
logue with the forces of the world around the organization.

✦ Each unit of a company, while alive, has a finite lifespan. One

day, each of the operating companies within Royal Dutch/Shell
will pass away. Some will die before Royal Dutch/Shell dies.
Some may live on past that time, as components that become part
of other corporations. Each organizational unit has a potential
lifespan, which it may or may not reach.
 ONLY LIVING BEINGS LEARN 111

You might argue that, whereas you are an individual being, a cor-
poration is simply a construct, composed of creatures. It may
contain many personæ, but it is not alive in the same way, for
example, that a human being is alive. But within your body, there
are cells, viruses, bacteria, intruders and parasites, often acting
without your conscious control and sometimes (for example,
when you get the flu) varying from your purpose. A company
contains managers, employees, shareholders, subsidiary compa-
nies, buildings, technologies and financial assets. Both of these
personæ, you and the company, thrive best when most of the
small entities are reasonably well dedicated to the survival and
potential of the whole.
For most businesspeople, the implications of this ladder are
a little harsh. Managers would prefer the world of business to be
like the world of things: always in harmony with its environment,
passive without a will of its own, just waiting for the manager to
give it a push, which would then produce a predictable and mea-
surable result. Instead, as a living entity, a company is always
insecure, never stable, always subject to shifting relationships
between the company and the outside world.
And with that understanding, in 1974 I began to see a way
out of my dilemma at Shell Brazil — a way to define the way that
Shell Brazil could stand for itself while remaining open to its out-
side environment.

Introception
There is another point about living entities: they are the only enti-
ties which can learn. Cybernetician/biologist Francisco Varela
expresses the idea this way: ‘Every living being that moves, has a
brain.’ A brain allows learning. Where there is movement, there is
learning.
 PERSONA (IDENTITY) 112

In my talks, I put forth the hypothesis that ‘companies can

learn’. I encounter little scepticism. I rarely add the rest of the
sentence: ‘Companies can learn because they are living beings.’ If
they were mere ‘bundles of assets’, they would be dead objects
and learning would be impossible for them.
I admit that it comes to me relatively easily to think of com-
panies, or indeed every institution (trade unions, clubs, churches,
government organizations), as ‘persons’ or ‘living systems’. The
same is true for legislation in nearly every Western country: cor-
porations tend to come under the legal definition of ‘persons’.
And everyday language grants corporations the same courtesy:
we speak of ‘General Motors deciding’ or ‘Unilever marketing a
product’. The way we speak of a corporation is much closer to
our language about human individuals than to the way we would
speak about a lump of ore or a sack of grain.
But why should it be important that corporations can
learn? In part, it is important because of a faculty called introcep-
tion. William Stern described this capability, common to all learn-
ing beings: the ability to be aware of one’s own stance and
position vis-à-vis the rest of the world.
Stern suggested that introception was one of three types of
interaction between an entity and the world around it. The first
level was simply a biological relationship to the threat, stimuli,
raw materials and accommodation of the environment. Is it hot?
Is there food? Does the environment excite us? Does it scare us?
Animals live mostly at this biological level.
The second level of interaction involves direct experiences and
encounters with the world when we feel harmony with it. These
experiences are stored in the memory of the persona, and recalled
selectively, to accentuate the positive. For as long as possible, we
seek to live in peace with our environment. We may feel tensions
with it, as mismatches occur, but we will try to put them aside.
 ONLY LIVING BEINGS LEARN 113

However, sometimes the tension exceeds a threshold

beyond which we cannot tolerate it any longer, and we reach a
crisis. Now we enter a third level: the level of values and beliefs.
We don’t need a crisis to reach this level; we can also get there
through sustained reflection. But once here, all our principles and
attitudes are open to question in light of the values and attitudes
of our environment. Do we agree, for example, with the prevail-
ing attitudes about ethics and virtue — attitudes which may be laid
down by authority figures, such as our parents or our bosses? If
we disagree, how strongly do we rebel against them? How do we
reconcile our differences from the prevailing sensibility of the
larger system? To Stern, questions such as these thrust people
(and companies) into a process of building greater awareness, a
process which he called introception.
Stern wrote that introception was a key function of the per-
sona of complex entities. They must find their place in the world;
they must develop a sense of the relationship between their own
persona’s ethical priorities and the values in the surrounding
world. For example, a living company is always engaged in ques-
tioning its own value system in relation to the ethics of the world
in which it lives. This process is directly linked to the company’s
persona, which may in fact date back to before the current legal
incarnation.
A consultant named Michael McMaster, for instance,
recently described how he visited a construction company in the
north-east of England, a company based at an old shipyard.
Already it was possible to see how the company’s current work —
manufacturing parts of North Sea oil-drilling platforms — would
disappear over time, as construction of new North Sea platforms
stopped. How would the company cope? The managers weren’t
sure until, after some prodding, they began looking back at their
own history. There had always been an economic activity based
 PERSONA (IDENTITY) 114

in that estuary, drawing its labour pool and membership from the
local community. They had built Viking ships there once; then
they had built windjammers, and then steam engines. Now they
were building oil platforms. What would come next? They could
not be sure, but it would involve finding a harmony between the
regional physical environment (the proximity to the sea) and the
human capability that had existed there for centuries. This land,
this community, was deeply entwined with the company’s per-
sona, and they could change that persona only with great,
wrenching effort and risk.
The historian Simon Schama describes a different sort of
dance between corporate values and the ethics of its environment
in seventeenth-century Holland, when a society in transition
tried to make space for new forms of commercial enterprise:

[The strong Calvinist] sense of the reprehensible nature of

money-making persisted, even while the Dutch amassed their
individual and collective fortunes. The odd consequence of
this disparity between principle and practice was to foster
expenditure . . . (notably) acts of conspicuous expenditure on
both pious objects . . . (and on) propitiatory gestures of
philanthropy.5

Sensitive to the world, the new commercial enterprises adapted —

not just by being more philanthropic than would have come nat-
urally, but also by fomenting institutional change:

The safest place of all in Amsterdam was its Wisselbank,

founded in 1609 . . . Its overriding concern was not to generate
funds for enterprise, but, on the contrary, to control the con-
ditions under which they could be exchanged . . . Its very exis-
tence testified to a determination to neutralize the worst evils
 ONLY LIVING BEINGS LEARN 115

associated with the unconfined world of money: usury,

default, counterfeit and other kinds of fraud. Its working
motto was ‘probity, not profit.’6

In the relationship between the persona of the corporation and

the persona of the local government, neither may be affected
immediately. One may overpower the other in the end. But both
sides will inevitably be affected by the other’s value systems,
because, as we will see, both are influenced by the same critical
factor: the values of the people who belong to them both.

When the persona and the world don’t match

Companies like Shell often find themselves in the sort of dilem-
ma that engulfed me in Brazil in 1974. There is a mismatch
between the company’s sense of ‘rightness’ and fair play, and the
ethics of the society around it. Such dilemmas can be very serious;
they can call into question the fundamental values of a manager’s
life.
At Royal Dutch/Shell, for example, the Statement of
Business Principles says that it is verboten to pay bribes or grant
favours to local politicians. But in many countries, bribing local
ministers is an accepted way of doing business. In other countries,
such as the Brazil I knew in 1974, the prevailing ethics of the host
country seem irreversibly bent on criticizing a company, just for
existing and doing business within its borders. And in still other
countries, such as South Africa in the late 1970s and early 1980s,
a politically reprehensible regime is in place and outsiders call on
corporations to leave the country entirely.
In these difficult situations, should a company refrain from
operating in that country? Should a company refrain from hiring
local people? Finally, what responsibility does a company have to
 PERSONA (IDENTITY) 116

an environment where its values and principles, its persona, seem

horribly out of place?
As I began to think about these questions in terms of
William Stern’s theory, it became apparent that the most intu-
itively appealing solution — holding fast to one’s own moral prin-
ciples and leaving the country — would often do the most damage
in the long run.
During the late 1970s, I was the coordinator in charge of a
region which included Africa. At that time, the senior manage-
ment of Royal Dutch/Shell was under great pressure from
activists in the United States, Sweden and the Netherlands to shut
down its South African subsidiary. Shell’s presence in South
Africa, and the oil it supplied, were seen to give legitimacy and
power to the apartheid system. For years, the senior managers at
Royal Dutch/Shell agonized about this. We were basically decent
people, and we knew as well as our critics about apartheid’s dev-
astating effects; we saw them directly and we didn’t like them. My
colleagues and I thought seriously about adding our voices to the
protests and supporting the removal of Shell from South Africa.
Yet, in the end, we chose to remain. Personally, I had two
reasons. First, protesting was not a role we understood at Shell.
We were very good at producing and distributing oil and gas. We
were not sure we knew how to manage effectively a withdrawal
of this scale and significance. The assets, such as refineries and
depots, could not be removed. The people to operate them were
mostly South African. Withdrawal risked being an empty gesture
of high morals, but low practical value.
The second reason had to do with introception. Stern’s the-
ory suggested that, if you want truly to change society, you could
only accomplish it from the inside out — from the clash of differ-
ent value systems in a single environment. Even while I agreed
that the apartheid system had to change, I was convinced that
 ONLY LIVING BEINGS LEARN 117

only working from the inside, as part of the South African com-
munity, could we change it. Kicking at them from the outside
would not work. The government which maintained apartheid
would not call on its introception to deal with us, if we moved
outside the country. We would simply become one more enemy
to be ignored. Within the country, we were one of the forces
affecting South Africa’s sense of its own identity.
In making this decision, I drew on two previous experi-
ences in African nations. When I took over as regional coordina-
tor of Africa in 1978, we had just shut down Shell Angola. There
had been a revolution, fuelled by Marxist ideology and East
German advisers. My predecessor had let himself be convinced
that the new environment was dangerous for Shell people.
Unwilling to risk the lives of any Shell employees, he shut down
Shell Angola and walked out of the country.
Shortly afterwards, our offices were visited by an Angolan
government delegation, led by one of the victorious generals of
the revolution, who had now become minister of energy. He still
called himself by his nom de guerre: General Monty. We made no
headway during the morning. He had brought a delegation, we
had a group of managers, and he spoke very little English. But at
lunch, I made sure we were sitting across from each other. After
my time in Brazil, I spoke Portuguese — one of the languages
common in Angola. Probably none of the other Shell people
spoke it. So I began talking to him in Portuguese over lunch and
his eyes lit up. Finally he could say what was on his mind.
‘You’re a bloody coward,’ he said. ‘You walked out of the
country.’
‘Sure,’ I said. ‘It was dangerous for our people. When you
marched up to Luanda [the capital] shooting, we had to leave.’
‘Maybe,’ he said. ‘But you’re Shell. You’re big, visible and
basically non-political. When you walked out, you gave the start
 PERSONA (IDENTITY) 118

signal for a flight out of Angola.’

It was true; in the 1970s, there had been a major refugee
flight back to Lisbon from Angola. Once Shell left, everyone with
a Portuguese passport wanted to leave as well. This meant that,
almost overnight, the middle class in Angola disappeared. The
shopkeepers, mechanics, engineers, station owners and the people
who ran the telephone company all considered themselves
Portuguese, even though they often had a great deal of African
blood.
‘When you whistled,’ the minister continued, ‘they fled.
And you did tremendous damage to our country. Admittedly,
Mobil took over, but Shell’s example was influential. When Shell
whistled, they fled.’
The infrastructure delivering aviation fuel to the airfields
and gasoline to the service stations began to break down as a
result, he said. Because they were still fighting the war, nearly all
of the available supplies that survived Shell’s exit were requisi-
tioned by the military. Now that the war was over, the country
had enormous difficulty maintaining and developing the infra-
structure to function as a nation-state.
‘That’s why you’re a bloody coward,’ he repeated.
That conversation stayed with me a long time. He had given
me something to think about, something on which I ruminated
for years.
Soon thereafter, another African revolution took place — in
Ethiopia. The emperor was thrown out and a group of Soviet-
supported Marxists took charge. They instituted a series of total-
itarian initiatives, including the ‘Red Neighborhood’ committees.
Anyone with a Western connection (such as working for a multi-
national company) was susceptible to being taken from bed in the
early morning, tortured during the day and shot at dusk.
It could have been very easy to shut down Shell Ethiopia. It
 ONLY LIVING BEINGS LEARN 119

was a relatively small company. But the top man, a young man of
Scottish descent, deeply impressed me. He had made up his mind
that he wanted to stay in the country. At 6:30 each morning he
stood by the door to the Shell office and counted the employees.
If one was missing, by 9 a.m. he sent out emissaries to find him or
her. They were often in the hands of a Red Neighborhood com-
mittee. By lunchtime, he was at the Ministry of War office. ‘There
is one of my people in the hands of that committee,’ he would say.
‘If he isn’t back by five this afternoon, I will stop supplying the
army with oil.’
We lost only one employee in the revolution. When it
ended, Shell was the only Western multinational company still
working in Ethiopia. All the others had packed their bags and left.
Except for Shell, the entire infrastructure and production capaci-
ty of the country were unmanned. Two years later, the revolu-
tionary communist regime allowed us to start paying dividends to
our shareholders. We still run that company, primarily with
Ethiopians as managers. It is one of their very few opportunities
to learn managerial skills without leaving the country.
Those two experiences taught me that we could only define
our responsibility situationally — vis-à-vis the larger communities
in which we worked. In thinking about South Africa, I had come
to the conclusion that it was an undeniably unpleasant regime.
But there were many unpleasant regimes in the territory for
which I was responsible, which stretched from Morocco to
Burma. The most unpleasant regimes were often almost invisible
to outside activist groups. We deeply disliked them, but we did
not step out of them. Our responsibility was to produce material
wealth, and we contributed, in the process, to the introception of
these repressive regimes.
Indeed, over the years Shell has seen many countries change
their value systems. South Africa moved into apartheid in the
 PERSONA (IDENTITY) 120

1950s; it moved out of apartheid 30 years later. Every time such a

change occurs, the company, like the other inhabitants of the
country, runs the risk that the value system of this part of its
world will begin to diverge from its own. The company has the
same choice that individuals have: either to adapt or to walk away.
Walking away, boycotting or sanctions would not move that
country closer to the company’s value system.
William Stern articulated the complexity of this relation-
ship, based in the introception of corporation and state.

The Persona has an influence on the world around it as an

example, a ‘role model’, but it can never equalize the world’s
values with its own. On the other hand, neither can the world
completely impose its values on an individual country.
Moreover, the mutual influence is weak in the present. The
most important effect of the radiation of persona on society and
the infection of the persona by society will take place in the
future. The examples given by ideas and work need time to be
absorbed.7

Is this always an argument for remaining, no matter how terrible

the regime? I think, with one exception, that the answer is yes.
And the exception is: you only remain if you are given a choice.
Every regime that starts killing, nationalizing or expelling you
does not offer that choice, and so you should leave. Examples in
Shell’s history include Russia after 1917 and China after 1946,
when Shell withdrew from those nations. Other examples include
the Jews who left Germany during the 1930s.
I have come to believe that a multinational company, work-
ing in more than 100 of the world’s countries, can only choose to
be itself and to remain truthful to its persona. If given a choice, it
should stay; but this means being neutral to the political values
 ONLY LIVING BEINGS LEARN 121

embraced in many of the host countries. If we chose to work in

only those value systems where we felt comfortable, in the end
we would operate in only about 15 per cent of the world’s coun-
tries. And we would give up, in the process, our persona’s oppor-
tunity to contribute to the world.

Who belongs?
During my Shell time, I was continuously aware of the level of
‘membership’ in everyone I met. I could encounter a salesman in
East Africa or a depot foreman in North Pakistan and know of
him — as he would know of me — that he was one of us, a Shell
person whom I could trust. On the other hand, I could meet a
non-executive board member, not recognize him as a Shell per-
son, and know automatically that he did not think of himself as
one of us either. This was not necessarily a bad thing. In some
cases, a non-executive director should not belong. Similarly,
external auditors who spent most of their working life auditing
the Shell account, learning more about the company than almost
anyone internally, were still not ‘one of us’.
However large or small the membership, the need for cohe-
sion between the members exists in all organizations, not just
companies. In a church, the harmony between the value system of
the members and that of the institution must be particularly
strong. A confession of faith is thus required of all new members.
A soldier in a country’s army cannot object, in principle, to the
values of that army; otherwise, the recruit is forced into civil ser-
vice instead of military service. And at the Royal Dutch/Shell
Group, a new member must subscribe to the company’s
Statement of Business Principles. Violation of these principles is
viewed as a serious offence, usually followed by expulsion.
The ham-fisted mergers and downsizings of recent years
 PERSONA (IDENTITY) 122

tend to tear companies apart precisely because they confound this

need for cohesion. These types of changes put enormous pressure
on the persona at the heart of the company’s identity. They strain
the linkages between the values of the company and the values of
its members. Managers are forced to ask themselves, ‘Am I still a
member of this firm? Is my membership secure?’
If this argument is persuasive, you may be ready to accept
the idea that a company is primarily a living being. But a question
probably remains. What does this mean for management? Is man-
aging a living system different from managing a company in the
traditional way, as a collection of assets? Can a living company be
managed in the same way as an economic company? Yes, a com-
pany may be alive. It may have a persona. But so what?
6
Managing for Profit or Longevity:
Is There a Choice?
DURING THE LAST FEW YEARS, THE PROBLEMS WITH WHOLESALE
staff reductions have become well known in business.
Managers are familiar with the costs of overloading the staff who
remain, of losing loyalty and of having to hire back new staff with
less commitment and capability, when the business expands again.
We understand all of these costs, yet we continue to cut staff rou-
tinely whenever return on investment (ROI) must be improved.
The reason is that we do not place the cutbacks in the con-
text of the real nature of our companies. There are, in fact, two
different types of commercial companies in existence today, dis-
tinguished by their primary reason for being in business. The first
type of company is run for a purely economic purpose: to pro-
duce maximum results with minimum resources. This sort of
‘economic company’ is managed primarily for profit. People are
regarded as ‘assets’ — extensions of the capital assets of the firm.
As with capital assets, investment in human assets is held to a
minimum in an economic company, to produce the greatest pos-
sible return in the shortest amount of time.
The economic company is not a work community, except
 PERSONA (IDENTITY) 124

incidentally. It is a corporate machine. Its sole purpose is the pro-

duction of wealth for a small inner group of managers and
investors. It feels no responsibility to the membership as a whole.
It provides no community for its employees and managers, except
as a by-product of its chief purpose: return on investment in cap-
ital and human assets.
This economic company represents a viable choice. Many
people in the business world may not want to create a work com-
munity. It is perfectly legitimate for anybody to want to have a
corporate machine with the sole purpose of earning a living for
that person or for his or her family. Moreover, it is also philo-
sophically viable. Assets are important and so is return on invest-
ment. Without these financial concepts, civilization would never
have developed its current ability to produce wealth. But making
this choice has consequences.
Here, as everywhere, there is no free lunch. People who run
economic companies have far fewer options in their managerial
practices. Only a small group of people qualify to be ‘one of us’ in
the inner circle. All other people recruited to contribute to the
corporate effort will become attachments to somebody else’s
money machine. They will be outsiders, recruited for their skills.
They will not be members. They trade their time and expertise for
money and they feel little loyalty to the company as an entity.
They may even feel little trust towards the people in the company
and certainly little desire to give their all on the firm’s behalf. This
means that hierarchical controls must be strengthened. Otherwise,
the economic machine will not work effectively. However, strong
hierarchical controls mean that it is less possible to mobilize effec-
tively the brain capacity of all the people in the firm.
If this economic business grows, then new members will
have to be admitted to the inner core of the company. These new
members will, inevitably, be chosen by the founders. They may
 MANAGING FOR PROFIT OR LONGEVITY 125

include family members or close friends. The ‘people who

belong’ will continue to consist of as small a group as possible.
Potential recruits will understand that, although they may be
recruited, they are not there to stay. They will work with their
eventual exit in mind.
A critical point will inevitably arrive, one day, when the
succession of the inner community will have to be addressed. The
founder, or his or her successor, will leave. This shows up the sec-
ond inherent weakness of economic companies. Not only is it
more difficult for them to be a learning organization, but they
also face considerable obstacles in the transition from one man-
agement generation to the next.
To me, an economic company is like a puddle of rainwa-
ter — a collection of raindrops, gathered together in a cavity or
hollow. These drops are the puddle. They remain in their posi-
tion, at the bottom of the cavity. When it rains, more drops may
be added to the puddle; its field of influence may broaden, soak-
ing the ground around it. But those original drops remain in their
position at the centre.
Paradoxically, this stability may lead to vulnerability.
Puddles of rainwater cannot survive much heat. When the sun
shines and the temperature heats up, the puddle starts to evapo-
rate. Even the drops at the very centre, in the middle of a heat
wave, are in danger of vaporizing. In fact, most puddles have very
short lifespans.
The second type of company, by contrast, is organized
around the purpose of perpetuating itself as an ongoing commu-
nity. This type of company has the longevity of a river. Unlike a
puddle, a river is a permanent feature of its landscape. Come rain,
the river may swell. Come shine, the river may shrink. But it
takes a long and severe drought for the river to disappear.
Yet from the point of view of the drops of water, the river
 PERSONA (IDENTITY) 126

is horribly turbulent. No drop of water remains at the centre for

very long. From one moment to the next, the water in one part of
the river or another will have changed. It will no longer be the
same. Finally, the drops of water run out to the sea. The river lasts
many times longer than the lifetime of the individual drops of
water which comprise it.
Instead of stagnating like a puddle, longlasting companies
seem to emulate the flow of a river. No one drop of water domi-
nates the company for long; indeed, new water drops continually
succeed the old drops and then in their turn are carried out to sea.
The drops of water are not destroyed; they are carried forward.
The river is a self-perpetuating community, with component
water drops that enter and leave, with its own built-in guarantees
for the continuity and motion of water within its banks. A com-
pany, by initiating rules for continuity and motion of its people,
can emulate the longevity and power of the river.
In such a ‘river company’, return on investment remains
important. But managers regard the optimization of capital as a
complement to the optimization of people. The company itself is
primarily a community. Its purposes are longevity and the devel-
opment of its own potential. Profitability is a means to that end.
And to produce both profitability and longevity, care must be
taken with the various processes for building a community: defin-
ing membership, establishing common values, recruiting people,
developing their capabilities, assessing their potential, living up to
a human contract, managing relationships with outsiders and
contractors, and establishing policies for exiting the company
gracefully.
To create a company that flows like a river, we must con-
sider the design of the channels that contain the flow.
 MANAGING FOR PROFIT OR LONGEVITY 127

The boundaries of identity

A river company is open to the outside world: there is tolerance
for a high entry of new individuals and ideas. It is, in fact, expect-
ed that new concepts and knowledge will flow through the com-
pany’s ‘stream of activity’ on an everyday basis.
At the same time, however, the river company maintains its
cohesive identity. Members know ‘who is us’ and they are aware
that they hold values in common. In a very real sense, they belong
to each other. In Chapter 5, I described ‘introception’ as the
awareness of one’s own values and how those mesh with the val-
ues of the outside world. But another sort of introception exists
within most companies. The values of the company coexist with
the values of individuals within the corporation — and every
member is aware of this coexistence.
A company has a collective sense of the answer to the defin-
itive question about corporate identity: who belongs? Who is
considered part of ‘us’? Conversely, who does not belong and
thus is part of the surrounding world?
There is no ambiguity about who belongs and who does
not. At the level of introception, the company’s members know
who is prepared to live with the company’s set of values. Whoever
cannot live with those values, cannot or should not be a member.
Whoever is not a member does not need to share the values.
However, they can share the values of some other institution, like
a trade union, and still be an employee-non-member of this com-
pany. Of course, these non-members are likely to act as non-
members, putting the needs of some other entity before the needs
of the company. Members must share the set of institutional val-
ues that rest at the core of the company’s persona.
At Shell, we saw this in our research on corporate longevity,
 PERSONA (IDENTITY) 128

the study described in the Prologue. Even in companies which

were widely diversified or decentralized, the employees and man-
agement often seemed to have a rather good understanding of
‘what this company stands for’ or ‘what this company is about’.
And they happily identified themselves with these principles.
In many cases, this value system had been brought in by the
founder. Sometimes it was even formalized in a sort of constitu-
tion. As the report noted:

One company saw itself as a fleet of ships, each ship indepen-

dent, yet the whole fleet stronger than the sum of its parts. This
sense of belonging to an organization and being able to identify
with its achievements can easily be dismissed as a soft or abstract
feature of change. But case histories showed that strong employ-
ee links were essential for survival and change. Successful com-
panies appear to maintain a cohesion at all levels.1

Cohesion is the force with which the molecules of a body cleave

together. In a company, cohesion represents the pulling together
(and keeping together) of the employees. To have cohesion,
employees must know ‘who belongs’ and ‘who is not one of us’.
In a living company, cohesion and diversity exist together.
The company is clearly a unit, with a single identity; but the peo-
ple and substructures within that unit show a rich variety. They
are composed differently from each other; they have different
characteristics and different potential. But they are all part of a
cohesive whole. Shell Brazil, for example, has quite different
characteristics from Deutsche Shell. As substructures, however,
they work together organically in the worldwide Shell Group. In
the same way, Deutsche Shell, subdivided in its turn, will expect
its own substructures — its divisions, refineries and marketing dis-
tricts — to work organically for the company as a whole.
 MANAGING FOR PROFIT OR LONGEVITY 129

This insight helps us maintain some optimism about wide-

ly divergent systems like a multinational company (or the
Catholic church or a cooperative of Raiffeisen banks). These sys-
tems can hold together. The substructures — the individual oper-
ating companies, monasteries and churches, or cooperating
savings banks — do not need to be made uniform for the whole to
keep together. On the contrary, there is value in diversity. Nor do
they need to be controlled by hierarchical submission. Cohesion
takes place on a more ephemeral, yet completely tangible, level.
Cohesion with variety may mean that the company has
members towards whom we do not necessarily feel a deep empa-
thy. Cohesion can mean being stuck with people we dislike. This,
I believe, is a blessing in disguise. One day, early in my career, my
boss (the finance controller of a Shell operating company) called
me in to give me an exciting new assignment: ‘We are going to
start a new department — Methods & Procedures — and you are to
be its boss.’
My department would undertake a thorough redesign —
what would now be called a ‘reengineering’ — of the old account-
ing and administrative processes. We would build a new
department out of nothing! I listened happily as the controller
read out the names of people who would be transferred to this
new department. But then he came, at the end of the list, to Mr Z.
I jumped from my chair. ‘Oh no, not Mr Z!’ Two years ear-
lier, I had worked directly under Mr Z in another section of the
accounts department. We had argued heatedly. I did not like him as
a person; I thought he was a blustery braggart. ‘All air, no sub-
stance,’ I judged, with the absolute certainty of a 23-year-old.
‘Young man,’ said the controller (who had always been a
mentor to me), ‘you will have to learn one or two things, and this
is a good time in your life to do it. First, if you want to be a leader,
you must realize that a manager is not God. A manager does not
 PERSONA (IDENTITY) 130

create people — certainly not in his own image. As a manager you

take people as they come, the way God created them, and you
learn to work with them.’
He let that message sink in before he continued: ‘Second, in
working with people, you should learn that it is inefficient to try
to make up for their "imperfections" (as you consider them) by
working harder or longer hours yourself. By doing so, you may
increase your output by 25 or 30 per cent over a limited period.
You might gain one third of a man-year. However, if you create
the conditions under which 10 people will each produce 10 per
cent more, you will have gained one full man-year.’
I did not contest the logic of his argument.
‘So, Mr Z is going to work with you,’ he concluded. ‘If, at
any time, I hear of a disagreement between the two of you, I will
not bother to find out the cause. It will be your fault! As a man-
ager you have to learn to work with people as you find them.
Your role is to create the conditions in which they will voluntar-
ily give their best.’
Mr Z and I worked together for two years. I forced myself
to listen to him, even when I disagreed. Only once did I explode.
I felt a red curtain of anger rising in me and I asked Mr Z, ‘Please,
go and close the door of my office on the outside.’ He did so. I
turned and walked to the window, picked up a steel ruler from a
table and broke it between my hands.
Mr Z, for his part, went straight from my office to the con-
troller. I only found this out at the end of the year, however, when
the whole episode was included in my performance report — no
doubt reducing my performance bonus.
Mr Z and I continued to work together for more than two
years. Indeed, I had to learn that part of my job involved creating
conditions in which even he was moved to give his best. Mr Z and
I, after all, were members of a common entity, and it was my duty
 MANAGING FOR PROFIT OR LONGEVITY 131

to that entity to find his strong points and help bring the maxi-
mum out of them. Mr Z and I needed more than mere tolerance of
each other. We would have to become aligned, with a sense of
common values and purpose. But that mutual sensibility could not
exist unless we knew, down to our bones, that we were members
of the same common whole and therefore could trust each other
even if we did not like each other.

Common values
Is there an alternative to killing the diversity of a company or
beating the organization into hierarchical submission? In other
words, what keeps the members of a living, river company in tune
with each other?
They subscribe to a set of common values. They believe that
the goals of the company will help them to achieve their own
individual goals. Or, as William Stern put it:

It is important that the individual goals of the substructures

are harmonious with and best served by the goals of the high-
er level system. In crude terms, this means that the overarching
structure should make it clear and prove to its components
down to the individual human beings, that their survival and
their self actualization are best served by working together
towards the survival and the development of the whole.2

This basic ‘egoistic’ principle is often misunderstood in business

governance. It means that anyone running a large, complex insti-
tution — such as a group of subsidiary companies, a set of joint
ventures or a company composed of business units — cannot sim-
ply dominate individual self-interest through the exercise of
power. As Stern might put it, each entity within a company, from
 PERSONA (IDENTITY) 132

the subsidiaries to the business units to the departments to the

individuals, is a persona in its own right. Each of these personæ
exists in constant exchange and dialogue with the world around
it. This means that each is continually testing its own values,
through introception, against those of the larger group.
Governance is a matter of assuring that the goals of the sub-
sidiary companies and of each employee are harmonious with
the goals of the larger whole — and vice versa.
A key governance issue, for example, involves the perenni-
al problem of the salaries paid to top managers. Press accounts
often excoriate CEOs and senior managers for making too many
times the salary of people at the lower end of the hierarchy. This
is a problem, but not for outsiders. It is a critical question for a
company’s introception about its own values.
Any executive’s remuneration is exaggerated if it is experi-
enced by the rest of the work community as exaggerated. One
work community might be quite prepared to pay its top people
fairly high amounts; that would be in harmony with the compa-
ny’s internal realities. Another work community would find high
salaries for senior managers unpalatable. The standard for any
pay scale is an internal standard. You may be my boss today, but
I’m the next generation. I know I’ll be the boss tomorrow, or one
of my peers will.
In short, the goals of each member must be harmonized
with the goals of the whole community. The whole and the parts
must understand that the interests of each are best served by stay-
ing together.
Many people seem to want this form of cohesion. It’s what
they’re describing when they say, for example, ‘I spend eight
hours or more every day at work. It’s a dominant part of my life.
And I want it to fit with the rest of my life!’
Our corporate survivor study at Shell revealed a very sug-
 MANAGING FOR PROFIT OR LONGEVITY 133

gestive link between longlived companies and a strong sense of

values. In some companies, these values hailed back to the
founder of the company. This person might even have codified a
kind of mission statement, describing a vision of what he wanted
the company to be. This would be a statement of what the com-
pany is about: a definition of self.
This should not be confused with the typical ‘corporate
vision and mission statements’ prominent in companies today.
Today’s mission statements are written in the future tense. They
talk about ‘what the company will be’. This makes them negative,
in the sense that they tell the members what their company is not
at this moment, but what it would like to be at some moment in
the future. They are explicit; they define the company as a ‘com-
puter firm’ or an ‘oil producer’ (admittedly the best in the
world!). Thus they overlook a critical question: what will bind
together the members of this work community, when the world
moves away from computers or from oil?
Founders and managers of longlived companies, a hundred
years or more in the past, did not link their values to a particular
product, service or line of work. They knew, or sensed, that the life
mission of a work community was not to produce a particular
product or service, but to survive: to perpetuate itself as a work
community. Their statements contained values and ethical rules
akin to a modern ‘statement of business principles’ or to the basic
tenets of a religion. Members, present and future, of that company
would have to make these values and rules their own, or else they
would not deserve their place in that company. More likely than
not, in deeply troubled times when nobody knew the answer to
totally new problems, the sharing of a set of common values helped
companies make choices to which all the individual employees
could subscribe. They were sailing blindly into an uncertain future,
but they could have confidence and belief in each other.
 PERSONA (IDENTITY) 134

Consider, for example, the story of the Japanese company

Mitsui, a drapery shop that became a moneylender and then a
mining and manufacturing enterprise. The founder, Takatoshi
Mitsui, left behind a large set of rules and guidelines when he died
in 1694. These included some organizational principles:

Those in authority should be kind to subordinates, who in

return should respect those in authority.

The essential role of managers is to guard the business of

the House. They should give appropriate advice if their mas-
ters’ conduct is not good and correct blunders that may be
made.

There were rules on personnel management:

Considerable amounts of silver shall be set aside as a reserve

fund for the benefit of elderly employees of the House who
have lost their property and also for the relief of those suffer-
ing from fire and other calamities.

In order to select for managers, keep an eye on the young

men and train promising candidates for that position.

Other rules describe the ethics of conducting business:

Farsightedness is essential to the career of a merchant. In pur-

suing small interests close at hand, one may lose huge profits
in the long run.

All kinds of speculation and new and unfamiliar business

ventures shall be strictly forbidden.

Persons in public office are not, as a rule, prosperous.

This is because they concentrate on discharging their public
duties and neglect their own family affairs. Do not forget you
 MANAGING FOR PROFIT OR LONGEVITY 135

are a merchant. You must regard dealing with the government

always as a sideline of your business.

These values and rules helped many generations of Mitsui people

to see who they were and what they stood for. Learning starts
with self-knowledge. A distinctive definition of self improves
cohesion.
The results in the case of Mitsui were impressive. The cor-
poration became the Japanese government’s official money-
changer in the eighteenth century. Yet in the nineteenth century,
it was adaptable enough to switch political allegiance and thus
survive the Meiji Restoration. In the twentieth century, Mitsui
covered almost every type of commercial, industrial and financial
activity through more than 100 subsidiaries. Then came the sever-
est test of its cohesion and adaptability.
In 1945 President Harry S. Truman issued a directive: the
industrial/banking combines, such as Mitsui, which dominated
the Japanese economy would be dissolved. The Shell report
describes the consequences:

Holding companies of the Zaibatsu combines were liquidated

and the wealth of the Zaibatsu families was drastically reduced
by a heavy capital levy. The two greatest trading houses, Mitsui
Bussan and Mitsubishi Sjohi were forced to liquidate in 1947.
To prevent them starting up again, the dissolution order pro-
hibited any company from employing more than two former
officers or one hundred employees. As a result, Mitsui Bussan
dissolved into 170 separate companies. In addition, the use of
such trade names as Mitsui and Mitsubishi was forbidden.

One might expect that this would be enough to annihilate any

corporate entity. Mitsui’s money was taken away, its managerial
 PERSONA (IDENTITY) 136

capacity dissolved, its name rubbed out and its central nervous
system (management structure and trading facilities) liquidated.
What could possibly still be alive to grow back into a living sys-
tem? The Shell report recounts the sequel:

A movement towards reunion of the Zaibatsu groups started

after the termination of the occupation (1952). Although the
top holding company was still missing, leaders of the former
Zaibatsu companies organized regular meetings for the pur-
pose of information exchange, and the bank, to which several
companies looked for lines of credit, began to perform some
of the central functions. The companies resumed old names
with Mitsui (in it) and the separated dissolved trading compa-
nies merged to form Mitsui-Busan (1959).

These days, Mitsui is thought to be the name of a group

composed of about 30 companies. Some of these put Mitsui
with their own names. The relationship among these compa-
nies is informal but it can sometimes be substantial. (For exam-
ple, Mitsui Petro Chemical was founded by eight companies of
the Mitsui group in 1965.) Although the Mitsui Zaibatsu has
been dead since the dissolution, the identity of ‘Mitsui’ seemed
to be kept alive within the companies of the group.

To what extent did the sharing of a set of common values make

this remarkable resurrection possible? It could not have been
negligible. Only a sense of common identity — a feeling of belong-
ing to a social system with a strong definition of self — could have
surmounted the destruction of the original unit into its separate
cells and genes. This collective sense of self was even stronger
than the ambitions and greed of the managers who had been
placed in charge of the individual companies after the dissolution.
They could have ‘gone it alone’ and in the short run they would
 MANAGING FOR PROFIT OR LONGEVITY 137

have been successful. But they were too infused with Mitsui’s val-
ues to leave the name and the entity behind.
You might be inclined to dismiss this example because
Mitsui is Japanese, and thus culturally inclined towards collec-
tivism. But the Deutsche Bank in Germany was similarly frag-
mented by occupying forces and it returned at full strength,
under its original name. At the same time, other companies in the
Axis nations did not survive the dissolution of the postwar peri-
od. The industrial empire of I.G. Farben, for example, was bro-
ken up into smaller chemical companies, including BASF and
Hoechst, which never came together again. In many instances
they engaged in strong competition with each other.

Joining the flow: recruitment policies

Living companies are very selective in the people they admit to
their membership. But they cannot be closed. Like the water in a
river, the membership of the community is constantly changing.
Sometimes members are forced out, when their value system is
apparently not harmonious with the company’s values.
Sometimes a small group of members redefines ‘who is us’ and
‘who is not us’. Attention must always be given to ensuring a reg-
ular stream of new talent to refresh the company.
Cohesive recruitment is not so much a matter of policy; it
is a form of reasoning. It starts by defining the desirable size and
shape of the human community. You may assume, for example,
that the total size of the company’s membership will stay more or
less constant during the foreseeable future. How many people
must then be fed into the stream to compensate for all the people
lost during the next 25 years to retirement, resignations and ill
health?
Then repeat this exercise under a number of different
 PERSONA (IDENTITY) 138

assumptions. Assume 10 per cent growth, for example. Assume

that one particular corporate unit will grow disproportionately.
As you consider each of these scenarios for potential growth, you
will discover that a figure begins to crystallize: a target for annual
and regular recruitment. This is a target figure, because a compa-
ny can never be sure that in each individual year it will find (or will
be able to afford) the exact number of people with the required
qualifications — the necessary academic or technical backgrounds,
the required mobility and the necessary value systems.
Things will go wrong if, for whatever reason, a management
generation starts interfering with this ratio of expansion. If times
are hard and it is decided to cut back recruitment sharply for a
number of years, then a severe price will be paid by the successors
of the managers who took this decision. Twenty-five years later,
those successors will sit around the table to choose new leaders.
They will find out that the generation from which they have to
select their successors is very thin in quality.
Things will go equally wrong if one corporate unit starts
growing faster than the rest of the company. If recruitment must
follow this growth, then the type of people recruited for that unit
(engineers or accountants or physicists) will increase dispropor-
tionately in the total mix. Twenty-five years later, the company
will find that it has lost diversity in its succession stream. Too
many people speak the same language and agree too easily with
one another. The company runs the risk of groupthink.
Entry-level recruitment should thus be seen not only as a
vehicle for bringing in new workers. Twenty-five years from now,
the quality of the most senior leaders of the company will depend
in part on the quality of today’s entry-level recruitment. This is
particularly true in longlived companies, which have demonstrat-
ed the value of promoting from within. James Collins and Jerry
Porras, for example, report in their book Built to Last that: ‘In
 MANAGING FOR PROFIT OR LONGEVITY 139

seventeen hundred years of combined life spans across the vision-

ary companies, we found only four incidents of going outside for
a CEO — and those in only two companies.’
In an economic company (a ‘puddle company’), recruitment
simply means finding the right people to serve the asset base of the
company. The pace of recruitment is handled by the numbers. If the
company has more demand for its product than it has capacity, it
admits new people (and machines). When demand slackens, the
company reduces capacity by letting people go. At the same time,
the people are defined in terms of skills: ‘We need 250 metal bash-
ers’ or ‘We have a surplus of paper pushers’. The language of recruit-
ment in economic companies betrays their underlying reasoning. It
is not people who are hired or fired but skills: ‘hands’ to handle the
machines or ‘brains’ to make the right types of calculations.
In a ‘river company’, by contrast, recruitment is a rite of
passage. It represents the first moment for testing the fit between
the new member and the community. This entry into the work
community deserves, and receives, a great deal of attention. It is
as if people are being admitted to a club, a professional body
(such as a medical association) or a trade union. New members
must carry the right (professional) qualifications, but the harmo-
ny between the member and the institution is equally important.
Do the values of the institution harmonize well with the values of
the prospective new member?
In the spring, representatives of major corporations appear
at the job markets at European universities. Only particular com-
panies graze these academic meadows, because the cost (in man-
agerial and recruiters’ time) is high. But there is no other effective
way to judge these very green graduates in terms of their poten-
tial for fitting well into a company. At Shell, recruiters do not
simply look for computer specialists or people with a biochem-
istry background. We ask ourselves about each candidate: ‘What’s
 PERSONA (IDENTITY) 140

in there, inside that person? Where could I start him or her?’ We

ask them questions about their attitudes: ‘Are you prepared to
work in other countries?’ And we may show them our statement
of business principles and ask if they’re willing to adopt it.
The harmonization of values does not mean that the orga-
nization is looking for clones of its present membership. This
would be dangerous, because it would stand in the way of acquir-
ing the human diversity necessary for long-term survival.
Looking for sympathetic clones at the moment of recruitment is
one of the temptations that successful managers steel themselves
to resist, difficult though that may be.
Even being critical and keeping our standards high, we will
often have a choice among various candidates. That is the moment
of risk. Having shortlisted the candidates on the basis of verifiable,
more or less objective criteria, how are we going to make the final
choice? We can only guess at the future of the human potential
before us. So, we tend to select the recognizable. Anyone who
seems vaguely ‘like me’ or has a similar background is sympathet-
ic. He or she, after all, will probably turn out more or less like me.
The risk is that going for the recognizable profile, the com-
fort of the known background, will reduce the diversity in the
company’s population. The diversity of people is probably
already too narrow. It is the sign of a living company’s maturity
when managers begin to look for people who are not like them-
selves — who may come from a different ethnic or national back-
ground, for example, and will thus bring a new set of attitudes
and talents to the corporate body.

Development of people
And the initial recruitment is only the first step. During the first
two or three years, several of the new recruits depart. The com-
 MANAGING FOR PROFIT OR LONGEVITY 141

pany and the new person look at each other and decide that they
don’t, in fact, get along. They don’t like each other. They don’t fit.
But the majority of new recruits remain. In Shell, several
hundred people per year remained worldwide during the 1980s.
They were now recognized as members. They had entered into
the system.
At any moment in time in any living company, there are at
least three generations in place, poised to succeed each other.

✦ The generation of recruits, up through their early 30s, is con-

tinually ranked and rated by their supervisors and peers. There
may be wild swings of approval and disapproval in these ratings:
‘Two years ago we thought Jeanne would make it up to managing
director, but last year she was a bloody failure.’
✦ Then, around the age of 37 to 40, the successive rankings of
potential seem to stabilize. People enter the second generation,
the generation of those in their late 30s and their 40s who are
identified as high potential and exercise quite a bit of leadership
in the company. (I was in this generation, for example, when I
became a senior manager of Shell Brazil.)
✦ Finally, there are people in their 50s, the senior managers of
the firm. (At Shell, the mandatory retirement age is 60.) They set
the course of the company and should spend a great deal of time
considering which people will succeed them.

This stream of succession does not imply a contract for life. Some
members of the company will drop out or fail. But, after the first
few years, the dropout rate becomes very low indeed. Moreover,
those in the stream can feel confident that they will at least get a
chance. And they know that the company will be willing to invest
in their development.
A top manager needs to know that there will be a supply of
 PERSONA (IDENTITY) 142

good managers in the future. Look at the most brilliant planner,

marketer or production person in your operation. Where will his
or her successor come from? It will take several years to grow
such a person, and one or two more years to identify them. An
organization with an awareness of this problem will have an
imperative to develop the human potential within the firm.
It might seem, at first glance, that a living company, with
continuity, would not allow much movement of people through
the enterprise. But continuity implies movement. Over time,
members are guided through the ranks in a process which aims at
developing them up to their ultimate potential. To some extent this
can be done through training, but it also requires a system of tak-
ing risks with people. Every time they take a new position, man-
agers are, in effect, given another test about how far they could go.
What is it that they could eventually do? What is the highest level
at which they could effectively function? What aspect of their
potential will be brought forth by this new position?
When I was a regional coordinator, I often had opportuni-
ties to conduct these tests. When a key position opened in, say, an
operating company in Kenya or Brazil, I had to develop as clear
an idea as possible of where this operating company was going.
Where did we want it to go? Then I would look for a particular
type of person — the person for whom this set of conditions
would work as a tonic, calling forth capabilities that he or she
never knew were there. If the position involved marketing in a
growing market, I might not necessarily look for an experienced
marketing manager. I might look for someone with the potential
to get things done under pressure and arrange to send him or her
on a three-week marketing course.
This type of development cannot be bought. It cannot even
be developed as a programme within one’s own company. It can
be produced only by taking risks and giving people time to come
 MANAGING FOR PROFIT OR LONGEVITY 143

to fruition. When a new general manager of an operating compa-

ny was appointed in my region, I would generally tell that person,
‘You can be reasonably sure of having this position for at least
two years. Therefore, don’t try to do too much in the first six
months.’ I wanted them to give themselves time to learn in the
position.
When managers get older, more creative efforts are needed
to draw forth their potential. People in their 40s, for example,
often go through a crisis when they begin to see that their work-
ing life will end in less than two decades. They begin to wonder
what they have done with their lives so far. At this stage, in a liv-
ing company, they will often flourish with a transfer to another
country or another function. The company shows them, in this
fashion, that it doesn’t consider them merely a specialized set of
hands or brains, but a person with a continually expanding poten-
tial. They may have to come to terms with the fact that they will
never be CEO or managing director, but the company has shown
that it is important to keep their interest going and their involve-
ment alive.
I required all my managers to spend at least 25 per cent of
their time on these types of issues related to the development and
placement of the people who reported to them. General Electric
CEO Jack Welch claims to require managers to spend 50 per cent
of their time on these sorts of development issues. Whatever the
percentage of time you spend on it, this could be the most critical
component of an executive’s work.

Assessing human potential

This emphasis on developing people also means that there must
be reliable ways to evaluate the potential and performance of peo-
ple — not to discipline them (for fear inhibits learning), but to
 PERSONA (IDENTITY) 144

appreciate better how to develop them.

A simplistic way to estimate ultimate potential is to ask the
boss. Slightly more sophisticated methods include asking a suc-
cession of bosses, while the incumbent moves through the sys-
tem. These methods are always unsatisfactory, because they rely
on the opinion of a very few people.
The best way that I have seen took place at Royal
Dutch/Shell. It consisted of assessments by ranking, once every
two years, done by successive teams of people (usually the man-
agement team of the relevant business unit), the peers of the per-
son being assessed and a personnel officer. Because of the fast job
rotation in the company, the composition of the team changed
over the years. The result was that not only were people judged
by a team (rather than being at the mercy of one person’s opin-
ion), but the team evolved through successive exercises.
This made the outcome much more acceptable to the
incumbent, even if the ultimate potential was lower than he or she
expected. Furthermore, the individual had the consolation of
knowing that the exercise would be repeated every two years. An
error made in one evaluation might be corrected the next time.
Most important of all, however, was the way in which the con-
tinual shifting of evaluation allowed for a kind of ongoing human
‘flocking’ (described in the next chapter), in which people contin-
ually came together to talk about the nature of human compe-
tence and capability.

Trust and the human contract

In an economic company (a ‘puddle company’), there is an
implicit underlying contract between the company and the indi-
vidual. Often unwritten, it is nevertheless universally understood:
the individual will deliver a skill in exchange for remuneration.
 MANAGING FOR PROFIT OR LONGEVITY 145

This agreement is usually ratified under the umbrella of the coun-

try’s social legislation or some collective labour agreement. It is
based on the underlying premise that most people, in the end,
really want economic reward — a higher paycheck — before all
other goals.
Similarly, the living, river company has an underlying
implicit contract. It, too, may never be written down, but it is
obvious in every personnel decision made by the company.
Individuals will deliver their care and commitment in exchange
for the fact that the company will try to develop each individual’s
potential to the maximum.
I have already described some of the methods of this devel-
opment. But its implications are equally important. Money is not
considered a positive motivator in a living company. It is, as psy-
chologist Abraham Maslow put it, a ‘negative hygiene factor’. If
money is insufficient, then people will grow dissatisfied, but
adding more money (above the threshold of sufficient pay) will
not motivate people to give more to the company. To give more,
the individuals need to know that the community is interested in
them as individuals, and they need to be interested themselves in
the fate of the larger entity. To give more, both the entity and the
individual need to care about each other.
Karl Weick, author of The Social Psychology of Organizing,
has written that what people really want from the workplace is
the ‘removal of equifinality’, by which he meant that people want
to see that they have brought order, design and quality to the
incoherent, ambiguous raw material of their work. They want to
see that their decisions and efforts have had a positive impact. If
they are treated only as ‘hands’ or ‘brains’, then they have no
sense that they are removing equifinality and they will look for it
anywhere they can find it. They may concentrate on collective
bargaining or getting a better salary — at least they can achieve
 PERSONA (IDENTITY) 146

that! — or they may go into secondary activities. They will

become the local organizer for the Red Cross or run the cricket
club. And that’s a shame for their employer. Running a local
cricket club is not easy, especially in terms of human relations.
People who can do that effectively could accomplish a great deal
for a company. The implicit contract of a living company guaran-
tees (not in words, but in actions) that they will have the oppor-
tunity to improve the world.
Robert Putnam, in his book on governance in democratic
societies,3 makes the point that no amount of hierarchical disci-
pline and power can possibly substitute for the absence of civic
behaviour and mutual trust in a community. Once the main struc-
ture and its substructures (subsidiary companies, departments,
individual members) agree that they have the same interests, goals
and purpose in life — once there is the conviction that everybody is
better off staying together — then we have true control. The mem-
bers of a community can be quite certain that all of them will try
seriously to reach their common goals, without needing the waste,
stress and rigour of coercive discipline. I would go even further, to
argue that a living company cannot abide coercive discipline. It
will not last, at least in today’s business environment, with those
sorts of strict controls.
If that is true, then living companies absolutely depend on
an implicit membership contract. Without the implicit contract,
there is no guarantee of continuity. Without continuity, there can
be no trust between the community and its individual members.
Without trust there is no cohesion and, thus, no living company.
Consider, for example, some of the highly politicized situa-
tions where Shell managers have worked. During one of the tur-
bulent periods before apartheid ended in South Africa, I went on
a visit to the country. When I landed at Johannesburg airport, I
was greeted with a letter from the police. It said that if we asked
 MANAGING FOR PROFIT OR LONGEVITY 147

questions of our own executives about our oil allocations while we

were in South Africa, we would end up in jail. We could not check
on the details of their supply system. We could rely only on trust.
This is an extreme example of an everyday dynamic. If you
run a multinational company with operations in over 100 coun-
tries, there’s only one way to run it. You cannot sit on the backs
of your managers. You cannot send auditors every week. You
must have mutual trust. Often decisions must be made in which
the managers in that company have no time to check their deci-
sions with superiors. They must move with the speed of basket-
ball players, aiming with as clear an idea as possible of the
company’s goals, roaring across the court to reach those goals, an
individual acting on behalf of a much larger entity that all of you
comprise together.
Think of the ramifications if you can’t trust people any
more because they don’t trust you, because they think you will
throw them out to save ten dollars.
Without the mutual trust that comes from an implicit con-
tract, the managers in the company will pay as much attention to
their salaries and compensation as they will to the needs of the
company. Therefore, there is much more probability of serious
error simply because their attention has been diverted. Any bas-
ketball player can tell you that, if you take your eyes off the ball
for even one-tenth of a second — for instance, if your eyes are con-
tinually diverted to the scoreboard — then you will lose points. If
you do business at the $100 billion level, as in the Shell Group of
companies, and your eye is diverted for one-tenth of a second,
then you can put millions of dollars’ worth of activity in jeopardy.
Moreover, the implicit contract of the river company can
accomplish miracles of profitability in its own right. We saw this
at Shell in the early 1980s, when the turbulent new oil economy
led us to construct a full-scale in-house commodities trading
 PERSONA (IDENTITY) 148

floor. When you have a commodities trading floor you need com-
modities traders, and we did not know at first where we would
find such people. Traders are very highly paid people. Their
salaries soared far above the Shell remuneration scales. In the City
of London, they were typically poached from other firms, and
they could therefore be expected to be poached, in turn, from
Shell. They did not fit at all into our human community.
We dealt with the problem, not by hiring traders, but by
having people from the Shell community work as traders within
the company — at salaries considerably lower than the City of
London was paying. We could never have kept those people, even
if they had formerly been Shell managers, by offering them the
contract of an economic company. Some other company would
have poached them. And, in fact, we lost some people, but far
fewer than the salary differential might suggest. ‘With us,’ we
could tell our trading managers, ‘you will be a trader only for two
to three years. It will be a very useful experience, and from there
you may well get promoted to marketing manager. We are inter-
ested in your long-term development.’ As we had done with
computer programmers in the late 1960s, when that skill demand-
ed a premium, we undoubtedly paid traders a little more than the
Shell scales warranted, but we were still able to integrate them
into the Shell system. This was very powerful. It even made the
traders accept the idea that they would drive to work in a Ford
Escort instead of a BMW.
Some companies do not institute an implicit living compa-
ny contract because executives worry that they are creating
sinecures. Is there not a danger, they ask, that people who have
been managers for 20 or 30 years will regard their jobs as a licence
to coast? Indeed, that is a possibility. But at Shell, we found that
the existence of an implicit contract allowed us to engage in reg-
ular dialogues about this. One manager can say to another, ‘I have
 MANAGING FOR PROFIT OR LONGEVITY 149

doubts about your ability to pull your weight.’ And the first
manager can respond by asking about the reasoning behind those
doubts and contesting the impressions which led to them.
Conversations like these can take place only in an atmosphere of
mutual trust.
This implicit contract does not necessarily mean lifelong
employment. The value systems of company and individual may
turn out not to be in harmony. As an employee, you may decide
that the company cannot help you meet your aspirations. Or the
company may come to realize that every time someone succeeds
you in a position there is a mess to clean up, because you tend to
choose quick solutions with unfortunate long-term conse-
quences. More likely than not, after ten years, there will be no
more career in this company for you; you have little potential left
for this company to develop.
Nonetheless, the contract affirms that there is at least a sta-
tistical probability of lifetime employment. As an employee, you
realize when signing on that there are no guarantees; you might be
laid off at any point. But you also know that, statistically, you are
likely to be valued for your experience, knowledge and ability to
produce results for the entity as a whole.
When people must be laid off, the implicit contract leads to
a very different discussion to simply saying, ‘We’re terribly sorry,
but you have half a year to find another job.’ In a living compa-
ny, a manager might say something like this: ‘Yes, the institution
is in dire times and we have to do something about it. One of the
things we have to do (having taken some care to reshape our cost
structure everywhere) is to eliminate some jobs, including yours.
Having said this, we still have an implicit contract with you. Are
there other ways to develop your potential that do not stand in
the way of developing the potential of the company?’
 PERSONA (IDENTITY) 150

Outsiders
The implicit contract of a living company makes one think very
carefully about the position of the outsiders who work, over the
long term, in service to the company’s purpose. This includes sup-
pliers, distributors, franchisers, contractors and even many cus-
tomers — who are not included in the membership.
In these relationships, there is often a very clear under-
standing that, for a limited period of time, the company and the
outsider will commit themselves to a relationship with each other.
Joe Jaworski, for example, served as the head of the scenario team
at Royal Dutch/Shell for three years. This put him in one of the
most significant, most visible roles at Shell Centre, the headquar-
ters of the global group of companies. As he entered into that
activity, Joe became a fully accepted full-time member, an ‘us’
member of the team. But everyone concerned knew that, at the
end of his tenure, that relationship would end. For these types of
relationships to work, there must be a special state of commit-
ment, the sort of mutual concern without formal structure, that
Joe writes about in his book Synchronicity.4
There are many such middle-ground positions. The details
differ. For example, someone in that position might not get a pen-
sion, but they might get an added payment which allows them to
develop their own pension contract. The important things are
that the contract exists, that it is two-sided — both parties must
consciously think about it and accept it — and that it recognizes
the value of the two entities, the contractor and the company, to
each other.
Outsiders are very important to a living company. With its
required recruitment policies, it cannot meet capacity shortfalls
by recruiting more members. Immediate capacity needs should be
 MANAGING FOR PROFIT OR LONGEVITY 151

met by subcontracting resources from the outside world. This

form of handling personnel matters is increasingly common. In
Italy, Benetton handles only a minor part of its manufacturing
with its own people; all but 20 per cent is subcontracted.
Relatively few members are admitted to the inner core of the
work community.
Some businesspeople fear that, by contracting out the phys-
ical operations of their business, they may lose control or become
more vulnerable to takeover or imitation by competitors. This
fear is not entirely unfounded, but it deserves a closer look.
The Shell Group runs one of the world’s biggest commer-
cial operations with ‘only’ about 110,000 people. But the number
of people working on Shell projects is far greater. Shell performs
very few of the necessary operations by itself. Its people do not
drill the wells, lay the pipelines, build the refineries or transport
most of the oil. The vast majority of Shell’s retail outlets remain
in the hands of independent operators. Shell itself is mostly a
management club. Members of its work community have a
unique ability: to get any operation going, anywhere in the world,
with most of the physical work handled by independent compa-
nies. This job — combining internal and external operations into a
productive whole — is more difficult to execute and to imitate
than if Shell people handled every job themselves.
We saw evidence of this during the spate of nationalizations
of the oil industry in the 1970s. National governments, with vary-
ing levels of technical sophistication, took over oilfields and
refineries. We consoled ourselves at Shell by saying that they
could not manage these assets as well as we could. That was prob-
ably true. Nonetheless, none of the nationalizations went so far
off track that it had to be reversed (although some countries were
clever enough to let the original owners run the oilfields and
refineries). The truth is that it is nearly always possible to hire
 PERSONA (IDENTITY) 152

somebody to run an asset like a refinery or an oilfield.

The only nationalizations which were reversed were mar-
keting companies in a few South American and African countries.
This work involved distributing our sophisticated product to
logistically difficult places — and making sure that they got paid
for the delivery. Local transporters, resellers and storage handled
most of the physical work, so the local governments (and we)
assumed it would be the easiest job to take over. It was the hard-
est. The government of Argentina could not learn to fit together
an activity which brought high-specification jet fuel to an airport
in Buenos Aires, gasoline to a petrol depot near Bariloche and
motor lubricant to a store in Rio Gallegos, in the right quantities
and at the right time. Only a work community with a highly
trained, coherent membership and high levels of learning can han-
dle a job like that: a job managing outsiders.

Exit rules
The need for rules about continuity is important not only when
people enter the community, but when they exit in retirement. Exit
rules, for example, stipulate that there is a fixed moment of retire-
ment for each and every member — without exception. At Shell, for
example, all managers leave the company at age 60. Living compa-
nies are the opposite of that old cartoon which shows 12 geriatric
board members nodding slowly to the chairman, who is proposing
to extend the retirement age by one more year!
Strict exit rules require the incumbent management to rec-
ognize that they are there for only a limited amount of time.
Leadership becomes stewardship. Just as you took over from
somebody, you will hand your leadership over to somebody else.
Your legacy will depend on whether you kept the shop as healthy
as you found it or made it just a bit healthier. Strict exit rules are
 MANAGING FOR PROFIT OR LONGEVITY 153

thus good for humility.

Of course, the rule-enforced departure is seldom without a
little tension, but an exit from a living company is not felt as a per-
sonal defeat. On the contrary, many alumni of Shell and other
companies find good use of their knowledge in academia, states-
manship or consulting. Consolation prizes, such as ‘honorary
jobs’, should be avoided; elephants’ graveyards are pathetic places.

When a river company changes course

And what of premature exit — when people are forced to leave the
company before they wish and before they have reached retire-
ment age? That, to me, is a symptom of a deeper problem. If a liv-
ing company seems suddenly to turn on a dime and traumatically
begins to lay off people in violation of the implicit community
contract, then this is a signal. The river company is trying to
become a puddle. It is being turned into an economic company.
In my present capacity as an adviser, I frequently find
myself in meetings at companies which have made this shift. I
have learned to recognize the mood in such a company. It is
understood, for the first time, that only a few dozen people, at
very senior levels, are the true ‘members of the company’.
Everyone else is subject to performance evaluations, with their
jobs at risk.
Sometimes I find myself at lunch, or in a meeting, with
managers who have recently found themselves moved into that
second group — the people at risk. The mood in these meetings is
invariably sombre, with a palpable sense of insecurity. Most of
them are not really at risk; they know that their skills are valued
enough that they will remain employed, at least for the moment.
But they do not have the assurance that they once had, the assur-
ance that the organization is concerned about their development.
 PERSONA (IDENTITY) 154

Their future with the company is on the line. Perhaps one day
they might be reelected into membership, someday, but even then
there would be a scar — a memory of the time when they had not
been selected to belong.
Often, in these meetings, it takes half an hour or more for
people to start talking about their problems and ideas. In the old
days, they rarely needed more than a minute or two before engag-
ing each other. But now they are not working for the community
any more. They are simply working for themselves.
Other times I meet with managers who have been selected
for the elite, the people whose membership is secure. The atmos-
phere at these meetings is ebullient; it evokes the same feelings of
camaraderie (‘Us with a capital U’) and identification with the
whole that had existed throughout the organization in the past.
There is, of course, a bittersweet awareness of the people who
have been laid off, and the conversation often regretfully turns to
the circumstances which have made this move necessary.
‘What are you going to do,’ I sometimes ask at these meet-
ings, ‘about the people who will stay?’ How will they make sure
that the people remaining, throughout the organization, recog-
nize that they are still members of the company? How will they
make people feel that their identity is valued as part of the whole?
I often find that the thought hasn’t even entered their heads.
In many cases, the management team’s thinking has been focused
on the people who will be laid off. How will they make their
selection of who goes and who stays? What will the company say
to the trade union? How will they treat the ‘downsized’ people in
terms of severance pay or helping them get another job? There is
rarely much thought left for the people who remain. Implicitly,
the message of the company to those people is this: ‘You may
stay, but it does not necessarily mean that you are still a member.’
In those circumstances, what happens to trust? What hap-
 MANAGING FOR PROFIT OR LONGEVITY 155

pens to the resilience that the company needs in its managers?

What happens to productivity, to discipline, and to the company’s
capabilities as a whole?
It takes a long time to build a river company. But if you
have a river company in place, you can demolish it in less than 12
months. Simply follow these easy steps:

1 Declare that the company isn’t profitable enough. Henceforth,

your goal will be a specific amount of return on capital employed.
2 Develop an action plan in which all assets will be trimmed
back across the board to meet these goals.
3 Follow the plan.

History shows that many companies which go down this route

face repercussions after a year or two. Exxon let 15,000 people go
in 1986, in the wake of the oil-price collapse. It concentrated
power in a narrow chain of command and took away one side of
its organizational matrix structure. In the process, it considerably
reduced its managerial capacity. A year later, the Valdez oil spill
took place. It took the company 48 hours to react. That 48 hours
has cost it $3 billion so far in cleanup costs, bad publicity and
legal fees. And the ticker is still counting.
For a while, it was difficult to find Exxon people with
whom to conduct business. Shell has a 50-50 partnership with
Exxon in a North Sea operation. Both sides meet regularly in
joint project groups and operating committees, asking, ‘We are
thinking of doing this or that in a certain way. What do you
think?’ Suddenly, it was difficult to find people at Exxon who
would answer these questions. They would no longer stick their
necks out. They no longer seemed confident of their relationship
with the parent company, and their ability to take constructive
risks had been destroyed.
 PERSONA (IDENTITY) 156

At another company, which had lost a great deal of people

suddenly, a manager told me this: ‘From one day to the next, our
productivity decreased by 15 per cent. People come to the office
and go to the coffee machine right away. They spend the first hour
out of the day asking what they can do today for themselves.’
Senior managers who inadvertently turn living companies
into economic companies are not wilfully destructive. They are
simply trying to do their best against the background of wide-
spread misapprehension about corporate purpose. They are strug-
gling with a terrible dilemma. When shareholders and outside
regulatory bodies ask about the senior managers’ results, they do
not ask about efforts to improve community. They do not enquire
about the company’s prospects for a long and prosperous exis-
tence. They ask, ‘What is your return on capital employed? Aren’t
you overcapitalized? What is your productivity?’
Being seen by the outside world as ‘managers of capital
assets’ places an almost irresistible pull on some living companies,
despite the internal need to build and maintain a human work
community. It is understandable that these companies evolve,
more or less quickly, into companies which are like puddles. Top
managers are pulled by the outside world in one direction and by
the needs of the work community in the other.
To manage a corporation effectively, we must learn how to
treat institutions as living ecosystems — set up with the recogni-
tion that they will live or die according to the same natural laws
that govern human growth and development. As we will see in
the next chapter, this sort of management does not take place pri-
marily in the arena of information systems and reengineering. It
is a social matter. It has to do with providing opportunities for
managers to learn together.
Part III
Ecology
7
Flocking
ACCORDING TO THE CONCISE OXFORD DICTIONARY, ECOLOGY
is the branch of biology dealing with the relationships
between organisms and their surroundings, including other
organisms.
I first learned about ecology in organizations during a visit
with Allan Wilson, a zoologist/biochemist based at the
University of California at Berkeley. I had come to Wilson only
by happenstance; Peter Schwartz, then the head of our scenario
team at Group Planning, heard a reference to him during a trip to
California. Peter thought that Wilson’s work might complement
our other investigations into accelerating learning.1
Wilson’s work in evolutionary biology would gain him the
MacArthur ‘genius’ prize several years later, but at that time we
knew him only as someone who studied the way animals learned.
He knew just as little about us. Indeed, he was bemused to find a trio
of Royal Dutch/Shell planners knocking on his door one morning.
Clearly, he was not used to businesspeople showing an interest in his
work — let alone managers from a multinational oil company. So we
explained, as best we could, that we were trying to understand the
nature of learning in large organizations like our own. We thought
there might be a clue in the nature of learning among animals.
 FLOCKING 159

‘Ah, well, then . . .’ Professor Wilson said. He was not sure

that his work was relevant to ours, but he was quite happy to tell
us about the role of learning in the evolution of life. We then
passed a fascinating couple of hours in which he told us, in much-
simplified language, about the ‘genetic clock’ embedded in the
molecules of genes of all species. The molecules of genetic mate-
rial, he said, change at a surprisingly constant rate in the evolution
of organisms, even those whose anatomy is evolving at very dif-
ferent rates. Through biochemical analysis of genetic material, it
was possible to measure each species for the number of genetic
‘ticks’ that it had undergone to reach its current state. Although
not all his colleagues subscribed to the idea, Allan Wilson was
convinced that the genetic clock runs at the same rate in every liv-
ing organism, even in bacteria. A species with more ticks of the
genetic clock in its structure would have evolved further than a
species with fewer ticks.
‘Do you understand,’ Professor Wilson asked us, ‘that in
this way it is possible to establish a table of all the species on
Earth, and see who would come out as the "most evolved"?’
Yes, we did understand.
‘Well, then,’ he said, ‘you would not be surprised to hear
which species was number one — the most evolved species on
earth.’
No, indeed, we would not be surprised. And, indeed, the
most highly evolved was the human species.
‘But what about number two?’ asked Wilson. ‘Which
species, genus or family is the second most evolved?’
The Shell executives could not guess. It turned out, Wilson
said, that number two in the evolutionary race, at least as highly
evolved as the higher primates, were members of the bird family.
Songbirds, in particular, show a high rate of anatomical evolution.
‘Isn’t that surprising?’ he asked.
 ECOLOGY 160

Why would it be surprising?

‘Because the common ancestor of all birds is, in evolution-
ary terms, a relative newcomer. Birds evolved out of the reptile
family. They had little time to reach the evolutionary level where
they are nowadays. It is especially surprising if we think of evo-
lution in terms of a Darwinian battle of survival and selection
through the generations.’
Conventional natural selection theory, he told us, posited
that changes only occurred between generations, as successful
individuals reproduced more frequently and the new generation
carried forward the most successful genes. But that theory could
not account for the songbirds. They had simply had too little time
to evolve to their current number of ticks on the molecular clock.
There had not been a sufficient number of generations. How was
it that, during a given timescale, one species (like songbirds)
seemed to have evolved so much further than others?
Could something else happen during the life of a generation
that accelerated the evolution of that species? That question was
the focus of Allan Wilson’s current research. He had developed a
hypothesis about ‘intergenerational learning’ — that species
behaviour, rather than environmental change, was the major dri-
ving force for evolution. In other words, certain species evolved
‘faster’, according to the ticks on the molecular clock, because
they exhibited a particular type of behaviour.
And what behaviour did primates and songbirds share, to
put them at the top of the table of evolved species? Wilson theo-
rized that accelerated evolution took place in species with three
particular characteristics:

✦ Innovation. Either as individuals or as a community, the

species has the capacity (or at least the potential) to invent new
behaviour. They can develop skills that allow them to exploit
 FLOCKING 161

their environment in new ways.

✦ Social propagation. There is an established process for trans-
mission of a skill from the individual to the community as a
whole — not genetically, but somehow through direct
communication.
✦ Mobility. The individuals of the species have the ability to
move around and (more importantly) they actually use it! They
flock or move in herds, rather than sitting in isolated territories.

The blue tit and the milk bottle

To test this hypothesis, Wilson had turned to a well-documented
case involving the blue tit, a small songbird common in British
gardens. The UK has a longstanding milk distribution system in
which milkmen in small trucks bring the milk in bottles to the
door of each house. At the beginning of the twentieth century,
these milk bottles had no top. Birds had easy access to the cream
which settled in the top of the bottle. Two different species of
British garden birds, the blue tits and the robins, learned to siphon
up cream from the bottles and tap this new, rich food source.
This innovation, in itself, was already quite an achievement.
But it also had an evolutionary effect. The cream was much rich-
er than the usual food sources of these birds and the two species
underwent some adaptation of their digestive systems to cope
with the unusual nutrients. This internal adaptation almost cer-
tainly took place through Darwinian selection.
Then, between the two world wars, the UK dairy distribu-
tors closed access to the food source by placing aluminium seals
on the bottles.
By the early 1950s, the entire blue tit population of the
UK — about a million birds — had learned how to pierce the alu-
minium seals. Regaining access to this rich food source provided
 ECOLOGY 162

an important victory for the blue tit family as a whole; it gave

them an advantage in the battle for survival. Conversely the
robins, as a family, never regained access to the cream.
Occasionally, an individual robin learns how to pierce the seals of
the milk bottles, but the knowledge never passes to the rest of the
species.
In short, the blue tits went through an extraordinarily suc-
cessful institutional learning process. The robins failed, even
though individual robins had been as innovative as individual
blue tits. Moreover, the difference could not be attributed to their
ability to communicate. As songbirds, both the blue tits and the
robins had the same wide range of means of communication:
colour, behaviour, movement and song. The explanation, said
Professor Wilson, could be found only in the social propagation
process: the way blue tits spread their skill from one individual to
members of the species as a whole.
In spring, the blue tits live in couples until they have reared
their young. By early summer, when the young blue tits are flying
and feeding on their own, the birds move from garden to garden in
flocks of eight to ten. These flocks seem to remain intact, moving
together around the countryside, and the period of mobility lasts
for two to three months.
Robins, by contrast, are territorial birds. A male robin will
not allow another male to enter its territory. When threatened,
the robin sends a warning, as if to say: ‘Keep the hell out of here.’
In general, robins tend to communicate with each other in an
antagonistic manner, with fixed boundaries that they do not
cross.
Birds that flock, said Allan Wilson, seem to learn faster.
They increase their chances of surviving and evolving more
quickly.
 FLOCKING 163

Flocking in organizations
Any organization with several hundred people is bound to have
at least a couple of innovators. There are always people curious
enough to poke their way into new discoveries, like the blue tits
finding their cream. However, keeping a few innovators on hand
is not enough, in itself, for institutional learning. The organiza-
tion must leave space for them, so that they do not feel squashed
and their innovations have time to develop. (This is the purpose
of such well-known innovations as the ‘skunkworks’ of
Lockheed Aircraft — spaces set aside for innovators to work with-
out interference from the rest of the organization. It also raises
deeper issues of control and freedom, to which we will return at
the end of this chapter.).
Even if you develop a high-calibre system of innovation,
you will still not have institutional learning until you develop the
ability to ‘flock’. Flocking depends on two of Allan Wilson’s key
criteria for learning: mobility of people and some effective mech-
anism of social propagation.
Consider, for example, the most effective possible forms of
training and development. Some managers see conventional train-
ing and development as merely an opportunity to acquire some
new skills. However, if it is given the wider definition of ‘devel-
oping individuals up to their potential’ (as discussed in Chapter
6), then training and development become powerful vehicles for
institutionalizing learning. Over time, the capabilities of the orga-
nization as a whole increase, more than you would expect mere-
ly from summing together the increase in individuals’ capabilities.
What qualities must this training have to be effective? First,
it must encourage mobility. At Shell, for instance, executive
development programmes run parallel to a person’s career. The
 ECOLOGY 164

organization spends about $2400 per employee each year on edu-

cation; half represents the pure cost of five to six days’ training,
and the other half consists of the trainees’ salaries.2 As always
with training, whereas the costs are substantial and quantifiable
the results cannot be measured. But the intangible results are
undeniable: Shell people know that, at every stage of their career,
they will be encouraged to move forward or to move into new
endeavours or to bring themselves (and the company) new skills.
Even more significantly, most of the training they undergo
is collaborative and related to real-world activities. Managers from
all over the world meet in collaborative problem-solving exercis-
es, so that the firm is constantly improving its own capabilities,
even during ‘timeout’ periods for education.
I have found it very important for teams of disparate peo-
ple to undergo intensive training together at regular intervals.
Apart from knowledge transfer, such an intensive training pro-
gramme brings together many groups of people, learners and
trainers, all from the same corporation, but coming from very dif-
ferent cultural backgrounds and many different professional and
academic disciplines. The flocking is intensive; course attendees
nearly always tell you afterwards, ‘It was not so much what I
learned in the official sessions but what I picked up from my col-
leagues during the breaks that was important.’
As with the blue tit’s innovation, when it learned to siphon
cream from British milk bottles, a well-designed programme of
development can have an evolutionary effect. The innovation
spreads rapidly through the organization, without being com-
manded to spread. Somehow, people just seem to know what to
do. They gain and spread the knowledge because they have been
given structures that encourage flocking.
 FLOCKING 165

Job mobility
One question often pops up in debates among human relations
and human resources managers: should people be thoroughly
trained to do a particular job and, once they have learned to do it
more or less decently, left in place so that they provide a return on
the investment of their training, or should we move people around
many jobs during their career and let them accumulate experience?
Although the two approaches to job rotation are not neces-
sarily in contradiction (one can move people around and still train
them thoroughly for each job), the philosophy underlying each is
quite different. The first approach is analytical; it sees the compa-
ny as a combination of machines and labour, organized to produce
the highest possible proceeds at minimum cost. The organization
is positioned to gain the most value possible from its investment in
the ‘asset’ of ‘human capital’.
The second approach sees the company as a self-
perpetuating work community. Each employee has an ultimate
potential, and it is in the company’s interests to help the individ-
ual reach that potential. Thus, people move from job to job with-
in the enterprise — in part so they accumulate the maximum
experience available during a working life and in part so that,
through ‘flocking’, the organization gains from their experiences.
The military in many countries has long learned the advan-
tages of mobility. Promotion by merit allows the creation of a top
command level from a much wider recruitment base in the popu-
lation. It considerably increases the chances of having a more
capable officer corps. Today’s general is not necessarily the son of
yesterday’s general. A lifelong emphasis on training — from offi-
cers’ schools to continual learning for enlisted soldiers — is also a
key component in achieving this end.
 ECOLOGY 166

Social propagation
Does mobility mean only that you move as an individual, from
group to group? Or can it also involve groups and teams that
move from situation to situation? It probably means both.
Most innovative companies are run by teams. This is
because teams have a higher capacity to learn than individuals. In
fact, in most companies with a certain degree of complexity, most
decisions are taken by teams.
The capacity for a management team’s learning is influenced
by the way the team is defined. It should include all the people
(directly or indirectly) who together have the power to act on their
common interest. Ideally, a management team at any level of the
company should include all people who are necessary for the imple-
mentation of that team’s decisions. They should be able to work
together on common problems, each with his or her individual con-
tribution and technical speciality. This would be an ideal ‘flock’.
Some companies facilitate flocking of their management
teams; other companies have stronger territorial tendencies. They
classify members by their speciality, skill, or mandate — production
engineers in one function, marketers in another. Then they appoint
a management team of people from various functions and give each
a specific written statement, spelling out in detail exactly what their
assignment should be. Each member is carefully instructed to avoid
encroaching on the others’ territories; marketing people do not
oversee production and production people steer clear of marketing
concerns. Finance managers concern themselves with measurement
and money handling and do not permit themselves to get involved
in process concerns, while process managers ignore the imperatives
of finance, except where they are given direct orders. Each robin is
allocated his or her territory in the corporate garden.
 FLOCKING 167

We should therefore not be surprised when these teams com-

municate as antagonistically as robins, squabbling at the boundaries
of their territories. The amount of institutional learning is limited.
As in bird species, the resulting social transmission will be
different in a territorial company. Both the territorial and the
flocking company may employ equally innovative individuals,
but the chances that the innovative ideas will become company
policy are much reduced in the territorial company.
A caveat, however, is necessary. Flocking is hardwired into
the blue tit species. Robins cannot be trained to flock; flocking is
not part of their genetic makeup. Therefore, anyone who tries to
apply this metaphor literally might be tempted to argue that com-
panies, too, are genetically predetermined. Some cultures are like
songbirds: they can learn to flock more easily because they have
institutional learning bred into them, whereas other companies are
more like molluscs. In that case, it’s hardly worth trying, because
the capacity for learning is innate and unchangeable . . . isn’t it?
I doubt it. Surely, corporate life is not a Greek tragedy in
which the outcome is hardwired into the characters by the
Olympian gods, and the play can climax only in its inevitable trag-
ic ending. Human organizations have resources for evolution, as
we will see in later chapters, that songbirds do not. And, even if
they have not participated in designing a company from its birth,
many managers will find themselves in a position with influence
on some part of the business. From there, they can begin to
remodel the company’s structures and policies in a way that facil-
itates flocking and improves the company’s ability to learn.

Innovation and the dilemma of freedom

Many managers shrink from institutional learning. They fear
flocking and they are terrified of innovation. And they are quite
 ECOLOGY 168

right to feel this way. They are stuck on the horns of the age-old
managerial dilemma between control and freedom.
Innovation and flocking require organizational space —
freedom from control, from direction and from punishment for
failure. Experiments must take place with relative safety.
Conversation must be free and candid, without fear of reprisal.
Employee movements must be largely self-determined; no one
can ‘command’ a bird to flock in a certain direction, because the
travel pattern of the flock emerges from its own movement.
This is terrain where many managers fear to tread. In many
companies, creating space is seen by most managers as losing effi-
ciency, or even losing cohesion. It is not a simple matter to decide,
one Monday morning, ‘We are going to create an atmosphere of
space in this company.’ Having made that decision, you lie awake
at night. God knows what some idiot is doing this very moment,
you think, in Malaysia or Chile or Sweden — or in your own office.
Because the worry is so agonizing, the average manager is
inclined to err on the side of control. A restful night’s sleep is a
very compelling motive. In the process, however, the organiza-
tion’s ability to flock is compromised.
Just as a car buyer expects a new car to behave in a known
and predictable way, a manager wants a company to produce pre-
dictable results and to give timely warnings when it is running on
a dangerous course. Both car buyer and manager require an
acceptable degree of control before they will entrust their lives
(literally or metaphorically) to their new vehicle. This need for
control is so fundamental that it has dominated management lit-
erature for 100 years. Books abound on financial control systems
and on organizational theory for effective managerial control.
The apogee came in the 1950s, with Taylorism, ‘scientific man-
agement’ and the widespread adoption of time-and-motion stud-
ies. Companies, these mixtures of people and capital assets, were
 FLOCKING 169

reduced to machines. Managers were reduced to being machin-

ists. The search for total control, however, could not be sustained.
The cost of maintaining a company without flocking and innova-
tion, where every adaptation had to be ordered from the highest
levels of the hierarchy, was too great.
Nowadays most managers recognize that cost. Almost
everyone is in favour of decentralization and empowerment — in
other words, in favour of increasing freedom. But even today, few
dare to risk the accompanying loss of control.
Most of those who dare will show their fears in a crisis. They
will recentralize quickly, pulling power back into the centre and
into the top. After all, beneath the rhetoric about empowerment,
most managers trust themselves infinitely more than they trust
anybody else. They will have to live or die with the consequences.
This dilemma, as it happens, is very common in ecological
matters. To behave with ecological concern often requires a leap
of faith: that you will be better protected by harmony and flock-
ing than by territoriality and force of will.
Ecology, after all, is itself a process of Piaget’s learning
through accommodation. Learning in ecosystems takes place
constantly, as entities adapt themselves to new understanding,
based on changing conditions in their environment. The entities
which survive in a turbulent ecosystem are those which are able
to adapt in this way.
Thus, what sorts of qualities would position an entity well
for this sort of learning? Would it be better to be strong and stal-
wart, to dominate a niche? Or would entities, like companies, do
well to cultivate more modest, adaptive qualities, such as toler-
ance and an appreciation of internal space?
 170

8
The Tolerant Company
ROSE GARDENERS IN A TEMPERATE CLIMATE FACE A CHOICE
every spring: how to prune our roses. More than any other
single factor under our control, the long-term fate of a rose gar-
den depends on this decision.
This choice, in turn, depends on the result you want to
achieve that summer. If you want to have the largest and most
glorious roses in the neighbourhood, you will prune hard. You
will reduce each rose plant to a maximum of three stems. Each of
these stems will be limited to three buds. Everything but those
nine strongest buds will be clipped away, to get the best results:
the biggest rose. This represents a policy of low tolerance and
tight control. You force the plant to make maximum use of its
available resources, by putting them into the rose’s ‘core busi-
ness’. You may expect, in June, to see some sizable flowers with
which to dazzle the neighbours.
However, if this is an unlucky year, you might get a severe
night frost in late April or early May. Then your rose plants may
well suffer serious damage to the limited number of shoots which
remain after your pruning. If the frost is serious, or if deer visit or
if there is a sudden invasion of greenfly, you may not get any
roses at all this year. You may lose the main stems or the whole
 THE TOLERANT COMPANY 171

plant! Pruning hard is a dangerous policy in an unpredictable

environment.
Thus, if you know that you live in a spot where nature may
play tricks on you, and if your primary desire is to maintain your
roses, year after year, you may opt for a policy of high tolerance.
You will leave more stems on the plant and more buds on each
stem. You may even leave some buds which seem, at first glance,
like they might produce distinctly unspectacular roses.
You will never have the biggest roses in the neighbourhood,
but you have a much enhanced chance of having roses every year.
You also achieve a gradual renewal of the plant. By leaving young,
weaker shoots on the plant, you give them the chance to grow
and to strengthen, so that they can take over the task of the main
shoots in a couple of years’ time. In short, a tolerant pruning pol-
icy achieves two ends:

✦ It makes it easier to cope with unexpected environmental

changes.
✦ It leads to a continuous, gradual restructuring of the plant.

This policy of tolerance admittedly wastes resources. As they grow,

the extra buds drain away nutrients and energy from the main stem
of the plant. But in an unpredictable environment, this policy of
tolerance makes the rose healthier. It allows the rose and the envi-
ronment continually to engage each other, without endangering the
rose’s permanent growth. Tolerance of internal weakness, ironical-
ly enough, allows the rose to be stronger in the long run.
Questions of tolerance are a fundamental part of a compa-
ny’s ecological stance. Ecology does not only concern the rela-
tionship between a company and its surroundings. Equally
important are the company’s relationships with the different per-
sonæ inside itself: its individual members, its subsidiary
 ECOLOGY 172

companies and its branches. Tolerating a variety of life forms

within itself gives a company the resilience to withstand stress
and even disaster.

Decentralization and tolerance

We saw this in our Shell study about corporate survivors.
Longlived companies, we concluded, were tolerant. However, at
the time, we did not talk about ‘tolerance’ or ‘adaptability’.
Instead, we used the terms ‘decentralization’ and ‘diversification’.
Companies which had managed to survive for a long time, we
wrote, had done so by letting things happen in the margin: allow-
ing activities outside the core business to be set up by not coming
down like a ton of bricks on every diversion in which local peo-
ple seem to believe fervently:

[The companies] have made full use of decentralized structures

and delegated authorities. The companies have not insisted
upon a relevance to the original business as a criterion for
selecting new business possibilities nor upon a central control
over moves to diversify.1

Every company we found that had been in existence for 100 years
or more had gone through a period of adaptation so profound
that it had had to alter its core business portfolio thoroughly.
Some had made this change several times:

✦ Booker McConnell, the ‘Guyanese company’, was a British

company formed in 1900. Its first business was sugar production
in South America. It then began spreading its investments, under
an outward-oriented management, in anticipation of the nation-
alization of its original core business. It moved into shopkeeping
 THE TOLERANT COMPANY 173

and shipping, as well as into publishing (through an ‘Authors

Division’.) Its annual Booker prize for literature is famous,
although Booker McConnell’s corporate identity remains largely
unknown.

✦ W.R. Grace, founded in 1854 by an Irish immigrant in Peru,

traded in guano (a natural fertilizer). Then it moved into sugar
and tin. Ultimately, the company established Pan American
Airways. Today, it is primarily a chemical company, although it is
also the leading US provider of kidney dialysis services.2

✦ In 1590 an ancestor of the Sumitomo family named Riemon

Soga opened a copper casting shop in Kyoto. From copper cast-
ing, Sumitomo moved into trading and then, in the seventeenth
century, into mining. In the nineteenth century, the company
developed a strong manufacturing orientation. Nowadays, it is
composed of 15 main companies which include banking and
chemical businesses.

To be sure, most of these changes involved some sort of crisis that

affected the entire organization, from its roots to its far-flung
branches. But, from what we can tell, for the people running the
enterprises at the time these changes were probably far more grad-
ual than we would guess. Some of them might have been almost
imperceptible from the outset, even from the inside. All of them
were accomplished without losing the company’s corporate iden-
tity or having the company subsumed by another organization.
As I pondered the link between this decentralization and
corporate longevity, I reflected that these companies had devel-
oped their values and organizational principles during the seven-
teenth, eighteenth, nineteenth and early twentieth centuries — long
before words like decentralization and diversification had even
 ECOLOGY 174

entered common usage. How had they described their own poli-
cies to themselves? What did the leaders, owners and managers of
centuries past think about their own efforts to build internal har-
diness through diversity and openness?
I do not know how they would have described it, but the
question led me to understand it more completely in my own
mind. Companies such as Booker McConnell, W.R. Grace and
Sumitomo — as well as others such as DuPont, Mitsui and Stora —
found it easier to adapt because they had tolerance. Tolerance was
the core quality which made it possible to diversify and decen-
tralize, yet still manage the entity as a whole. These companies
were particularly tolerant of activities in the margin: small, seem-
ingly strange businesses that might have been pruned off the cor-
porate rosebush elsewhere, but here were given enough resources
to straggle along until the corporation needed them as an outlet
for endeavour.
When a DuPont needed to move into chemicals, a
Sumitomo into banking or a W.R. Grace into aviation, there was
already a budding nexus of capability within the enterprise, ready
to move into a new status as a core business. Moreover, because
the company had been tolerant of this ‘bud’ of new activity, it had
been given time and room to emerge organically from the core
structure. Thus its presence in itself demonstrated where the cor-
porate entity as a whole might naturally and profitably move next.
In short, systems that deliberately introduce diversity into
the product line — even at the expense of short-term proceeds —
and allow activities to go on undisturbed in the margin of the field,
have greatly enhanced chances of survival across the generations.
These systems are tolerant. Tolerant systems survive.
At first glance, tolerance of diversity might seem to contra-
dict the need for cohesion. But, as we noted in Chapter 6, cohe-
sion itself is improved by diversity. Royal Dutch/Shell is stronger
 THE TOLERANT COMPANY 175

because of the differences between Deutsche Shell and Shell

Brazil, not in spite of them. The differences between the operat-
ing companies have forced strength into the global parent; it must
be a strong enough container to hold all of those differences with-
out cracking. This strength has been built up gradually, over time.
It is made possible by the fact that the individual goals of the sub-
structures (the operating companies) are harmonious with and
best served by the goals of the higher-level system. At the same
time, full use can be made of the wealth of diversity available in
the system as a whole.
In addition, a cohesive system must be open to change and
diversity. Tolerance is a measure of the openness of a system. The
more tolerant a company, the more new people and ideas it can
absorb and foster over time. Tolerance is a dynamic characteris-
tic; it changes the composition of the company. Diverse people,
products and ideas require us to be patient with them; indeed, tol-
erance is patience. It requires time.
At the moment, as it happens, the prevailing managerial
attitudes tend to discount the value of tolerance. Small, budding
businesses in the margin are not seen as an organizational asset.
During the 1970s, many corporations learned that diversification
is a dangerous route to take. The lessons were so painful that we
are still in a long period of reaction, an antidiversification move-
ment that has already lasted 20 years. ‘Stick to your knitting,’ we
are told. ‘Go back to basics.’ Managers concentrate on their ‘core
competencies’ and ‘core businesses’. They eliminate all business-
es which do not perform first or second in their markets. Yet,
amid these fashionable attitudes, there are strong indications that,
in order to survive in an age of technological, political, economic
and social upheavals, a company must be able to change the
emphasis of its business portfolio completely. There may be a
‘core competency’ at any given moment, but in the next moment
 ECOLOGY 176

a new core competency may be needed.

Why then was the last round of diversification so devastat-
ing? During the 1960s and 1970s, most of it took place by dictum.
Diversifications were initiated from a central control point at the
top of the firm. They were financed through great allocations of
‘new business’ investment.
By contrast, the long-term corporate survivors we had
studied at Shell had made their changes in gradual, incremental
ways — almost always in anticipation of customer needs. New
business was not required to be relevant to the original business
and, above all, there was no central control over the diversifica-
tions. They were often minimally financed or self-financing;
often, they were simply the natural result of letting some group
of inventors or creators within the firm have time to experiment
and take risks.
Historically, diversification by tolerance for activities in the
margin has a much better track record than diversification by dic-
tum. Could it be, in fact, that the very top of a company is not the
best place to take initiatives — except in a crisis? Under ‘normal’
conditions, to judge by the record of longlived companies, the
senior manager of a company should make fewer decisions about
the business itself and spend that time instead focused on creating
conditions in which other people within the company can make
good decisions about the business.
Tolerance, in other words, is derived from a value system. It
can only exist in a company where people recognize the value of
creating space for innovation. This is the reason that some compa-
nies set aside pockets of innovation: the ‘skunkworks’ of
Lockheed or the famous sidelines at 3M and Motorola, which
grew into main lines of business. In essence, these companies are
creating pockets of organizational space in which innovation can
emerge. The pockets tend to be hidden away in backwaters of the
 THE TOLERANT COMPANY 177

company. The senior managers trust them, don’t oversee them,

keep them generally out of sight and out of mind, and don’t worry
about them — until they are needed.

The parable of the Chilean potato

A few years after the Shell study was finished, I mentioned some
of its findings and the thoughts it called forth in my mind at a din-
ner party in San Francisco. Paul Hawken, author of The Next
Economy, was one of the other guests, and he told me the story
of the Chilean potato.
There was a time, he said, when the balance of payments of
Chile deteriorated to the point that foreign exchange became a
problem to the country. The causes seemed clear: Chile could no
longer produce its own food and had to rely increasingly on
imports. The US decided to offer a helping hand and dispatched
a team of agronomists to study the problem.
The team flew to Santiago and proceeded from there direct-
ly to the Andes mountains. The Andes are where the potato orig-
inated; it is still a main staple in the Chilean diet. Potatoes have
grown for thousands of years at considerable heights in the
mountains.
The US agronomists climbed these heights and looked at
the potato fields. The fields clung precipitously to the mountain-
sides. They had highly irregular shapes and were interspersed
with boulders. Within each field, the agronomists discovered ten
or more varieties of potatoes growing. There were round potatoes
and elongated potatoes, red, white and blue potatoes; and — more
perturbing to the scientists — some plants which bore many pota-
toes and others which bore only a few tubers. This seemed terri-
bly inefficient.
Harvesting time came and the insights of the agricultural aid
 ECOLOGY 178

team increased even more. They noticed that the peasants were less
than systematic — almost ‘lazy’ — in the way they reaped the crop. A
considerable number of plants in the nooks and corners of the
oddly shaped fields were overlooked and left to grow wild. By then
the team had most of the elements needed to write a convincing
report. Out came the handheld computers. The calculations
showed, beyond doubt, that a more careful selection of seed pota-
toes, a switch to higher-yield varieties and more systematic weed-
ing and cropping of the fields would increase the annual crop by at
least 15 per cent. Because this equalled the shortfall in the country’s
food production, the team took their plane back to the US with the
feeling of a job well done.
But the advice was wrong. However scientific the agrono-
mists’ approach may have been, they could not compete with the
accumulated local experience, based on thousands of years of
potato growing in the Andes.
Chilean peasants, based all their lives in the mountains, know
that a wide variety of terrible things could harm their potatoes.
There may be a late night frost in spring or a caterpillar plague in
summer. Mildew might destroy the plants before any tubers have
formed or winter might come too early. Over the years, each of
these calamities has taken place from time to time.
Whenever a new calamity strikes, the peasants walk up to
their fields and look everywhere — in the corners, behind the boul-
ders and amid the weeds — for the surviving potato plants. Only
these surviving plants are immune to the latest plague. At harvest-
ing time the peasants will carefully dig up the survivors and take
the precious potato tubers back to their huts. They and their chil-
dren may have to go through a winter of famine, but at least they
have next year’s seed potatoes from which a new start can be made.
They are not locked into a particular set of farming practices or a
particular type of potato; they may be inefficient at times, but they
 THE TOLERANT COMPANY 179

have diversity bred into their everyday practice, diversity which

allows them to meet unforeseen disaster.
Paul’s story about the Chilean potatoes illustrated, to me at
least, that tolerance towards the margin is a generic survival strat-
egy. Apparently there are many examples in the arena of agricul-
ture, particularly where ‘efficient’ agronomic practices overrun
the delicate interwoven tolerances of indigenous agriculture.
Ecologists have a word for this type of agricultural efficiency:
they call it ‘monocropping’. By reducing the number of varieties
in, say, a potato patch or a wheat field, the monocropping
approach guarantees a much greater yield in the short run. But, in
the long run, it depletes the soil, diminishes the variety in the sys-
tem and threatens the health and life of the plants and animals liv-
ing there.
Corporate systems that aim at maximizing short-term pro-
ceeds and disturb or cut out activities that operate in the margin
of the company’s ‘field’ represent the business equivalent of
monocropping. In the long run, the parable of the Chilean pota-
to suggests that these companies have greatly diminished chances
of survival.
In retrospect, this comment from the Shell report seemed to
resonate particularly closely with the parable of the Chilean
potato:

Successful companies tended to perceive other [internal]

resources as being capable of development in addition to the
existing resource rather than instead of . . . Many successful
moves were made when companies did not see themselves
locked into a particular business, but in business, with talents
and resources that could be used profitably to meet a variety
of consumer needs. Successful moves were relatively free of
immediate pressure.3
 ECOLOGY 180

Successful companies, in short, were free to go against the grain

because they had been cultivating, within themselves, a wide vari-
ety of potential activities.

Tolerance and the ecology of companies

The choice between tolerance and intolerance, in the end, is a mat-
ter of ecology. The choice depends in part on the organization’s
interrelationship with the other living beings in its world, and also
on the organization’s relationship with the living beings — including
other organizations — within itself. By maintaining a certain level of
variety within itself, the living company is far more adaptable,
because it is far more capable of responding effectively to the vari-
ety of forces which exist in its environment.
Intolerant companies can have long and healthy lives, pro-
vided that they have an appreciable amount of control over the
world in which they live. The banking and insurance industries in
many European countries lived for a long time in this sort of
world. So do many postal, telephone and telegraph (PTT) com-
panies, as long as their telecommunications bailiwicks are strong-
ly regulated. Under monopolistic conditions, a stable market or
other conditions in which the company maintains control, the
managers will do well to optimize efficiency. They will go for
maximum results with minimum resources. This minimization of
resources inside the company requires an intolerant management
style. There can’t be much room for delegated authority and free-
dom of action.
Instead of growing potatoes in an open environment like
the Andes, these intolerant companies are growing potatoes in a
glasshouse. In a glasshouse, the horticulturist controls the
amount of heat, light, fertilizer and humidity in the environment
24 hours a day. Over time, like the horticulturist, managers of
 THE TOLERANT COMPANY 181

intolerant companies become more and more clever at finding the

optimum, most efficient methods for growing the potatoes in a
controlled environment. There is a great deal of learning by
assimilation — Piaget’s term for taking in new information with-
out changing one’s fundamental way of thinking or acting. The
managers’ structures and knowledge base get honed over time to
deal with a familiar world, but there is little learning by accom-
modation (making internal changes to fit a changing world).
The company will undeniably thrive for as long as the
world remains stable. High tolerance, after all, is wasteful of
resources. That is why a company with a lot of control over its
environment has few reasons to be open or tolerant. Such a com-
pany may be lucky and live all its life in a world with a stable mar-
ket for its products or services.
When the environment becomes unstable, however, there is
a need for fast learning. Now, suddenly, the glasshouse cracks.
The external environment, with all its unpredictabilities, reasserts
itself. The managers must return to growing potatoes in the
Andes. Diversity and openness are much better management dic-
tums in those conditions, as the sample companies in the Shell
study of corporate survivors demonstrate.

Balancing freedom and control

Why then is it so difficult to switch from corporate monocrop-
ping to diversity? Why do managers resist the shift, even when
they are keenly aware of the benefits that a practice of tolerance
might bring?
Some might argue that monocropping remains compelling
for managers because it continues to produce immediate results.
Even if managers know that those results stem from temporary
efficiency at the expense of long-term survival, they will hang on
 ECOLOGY 182

to the rewards they receive for bringing in ‘good numbers’. But

let’s beware of easy criticism. As I said in the last chapter, in a cri-
sis the natural reaction is to centralize, to decide quickly and to
impose solutions.
Therefore, I think that there is more: the deeper reason that
managers resist the notion of tolerance is the dilemma between the
need to control, a need we discussed in Chapter 7, and the need for
tolerance and freedom. These two needs, both equally desirable, are
at first glance mutually exclusive. Freedom and tolerance are neces-
sary to increase the learning abilities of the organization. Yet con-
trol is necessary to maintain its cohesion.
When push comes to shove, most managers will choose
control. In fact, it is emotionally difficult, in most companies,
even to relax the emphasis on control. Managers who are doers,
accustomed to getting things done, will tend to trust themselves
more than anybody else. They feel an emotional pull toward
exercising power and domination, and they fear the chaos and
uncertainty that come with letting go of the reins. If we let peo-
ple have freedom, they’ll simply do what they want and the
results will suffer.
And yet, the more changing and uncontrollable the world,
the more evident become the dangers of control. Relying on a
management policy of high internal and external controls puts a
tremendous burden on the few ‘thinking’ managers at the top. The
capacity to take decisions becomes severely limited and everything
waits on the judgement of the few ‘trusted’ people. By definition,
the neck of a bottle is always at the top! This very basic dilemma
is often couched in terms of the juxtaposition ‘to centralize or to
decentralize?’ Nowadays, almost everybody is in favour of decen-
tralization, but few dare to risk the accompanying loss of control.
That is why rational debates about centralization versus decen-
tralization are seldom followed by rational decisions. Instead,
 THE TOLERANT COMPANY 183

managers tend to make knee-jerk decisions, in which they abrupt-

ly move towards freedom and then suddenly retract back towards
control, oscillating continually in one direction or the other.
For all these reasons, creating space — increasing the mobi-
lization of internally available brain capacity — is always a tricky
business. The dilemma between tolerance and control can only be
solved, of course, by finding a way to meet both goals. Space must
be created for people to experiment and take risks. At the same
time, people cannot simply do what they like at the expense of the
organization’s common purpose. Clearly, one needs both:
empowered people and effective control.

The management of tolerance

If you are a manager, to control can mean to keep down the cost
per unit of output. In contrast, a management that empowers is
trying to increase the output per unit of costs. Good management
knows that it will have to do both. Whether it aims simultane-
ously at both targets, or whether it will ‘wave’ (oscillate) over
time from more control to more tolerance and back,4 an
‘either/or’ attitude — opting for either control or tolerance —
could be fatal in the long run.
Pursuing two contradictory goals is not easy. It should
therefore not surprise us that, between the horns of this particu-
lar dilemma, a pseudosolution has sprung up over time. It is called
‘strategic planning’. It sounds quite neat: if you have a strategy
that tells you where you are going, you don’t need to worry too
much about loss of control. You can let people get on with their
jobs, giving them freedom and space, because management sup-
posedly has control over the company’s direction and destination.
Disappointments with strategic planning were not slow in
coming. As Henry Mintzberg describes in his history, The Rise
 ECOLOGY 184

and Fall of Strategic Planning, complaints and carps about strate-

gic planning began to appear around 1973 and the criticism gath-
ered momentum through the 1970s and 1980s. Moreover,
strategic planning advocates could point to few successes. Even
strategic planning in the military became discredited, as Robert S.
McNamara’s Planning-Programming-Budgeting System (PPBS)
5
was blamed in part for the failures of the Vietnam War effort.
The fact is that the word strategy tends to be misused. It
should not be a noun; you should not ‘have’ a strategy, in the
sense of a document which the organization follows. Rather,
strategy should be a verb: strategy is something you do, rather
than something you have.
One thing management can do in a company is steer. This
is a very popular concept, both in management literature and
among managers. In this concept, strategy (or ‘steering’) is the art
of management. It consists of the daily activity of steering the
institution in a direction that will ensure that it, and the entities
within it, move towards their full development. Henry Mintzberg
noticed this himself in 1973; in his book The Nature of
Managerial Work, he reported that managers, whom he had
observed on the job, rarely sat back to plan in any abstract, cere-
bral sense. Instead, ‘strategies’ emerged out of the give and take of
day-by-day management activity.6
So let us talk about the art of strategy and steering in a com-
pany which supposedly balances tolerance and control. When
one thinks of steering, the metaphor of a ship comes easily to
mind. Many a chairman, in his public statements, talks about
‘weathering the storm’ or ‘changing tack’ and ‘setting a course’.
The metaphor seems, at first glance, to fit well. In a company, as
on a ship, there is a defined command structure with everyone in
some sort of specialized activity. Some people run the machine
room, some lift the anchor and hoist the sails. The boss, known
 THE TOLERANT COMPANY 185

as the ‘captain’, exercises the necessary degree of control and dis-

cipline to ensure that the ship and crew act in unison. It is clear
that the skipper is boss.
In this metaphor, the ship (or the company) is an asset
manned by people. It sails from destination to destination to
make profits for the owners. It has no interest in its own longevi-
ty. On a real ship, it would be inadmissible for the first mate or
the boatswain to have the freedom to change to another tack or
set another course. Chaos and anarchy would result; the ship
could well run on to the rocks. Of course, the skipper might sit
down with the first mate and discuss the destination and how best
to get there. He may even have a word with the helmsman. Yet
the skipper carries the ultimate responsibility. Even so, the skip-
per knows that he is on this voyage only because another human
voice has decided to send the ship to this destination. That voice
is the owner’s.
A living company, by contrast, is a living being. It moves
from birth to death, seeking to extend its own potential. There is
no one steering. Instead, to change the metaphor, the living com-
pany takes one step at a time. Each decision is followed by an
action, and then new observations about the effect of that action,
and then another step tomorrow. As I described in Chapter 3, the
company develops memories of the future to assist in its deci-
sions. Before taking each new step, it looks up and decides where
to put its foot in the light of the conditions of the moment. There
are no maps and no final destination, except death.
Such a company knows that it is capable of only certain
accomplishments at this moment in its history. These capabilities
restrict the number of places where it can put its foot today. It
may also have some untapped potential for future actions.
Developing that potential will increase the number of places
where it can step tomorrow.
 ECOLOGY 186

In which direction are we steering?

How do you steer a company like that without dissolving into
anarchy? If you are a manager, the poet Machado has a quote
which you might find relevant.
Life is a path that you beat while you walk it.7
To me, this line embodies the most profound lesson on planning
and strategy that I have ever learned. When you look back, you see
a clear path that brought you here. But you created that path
yourself. Ahead, there is only uncharted wilderness.
You do not navigate a company to a predefined destina-
tion. You take steps, one at a time, into an unknowable future.
There are no paths, no roads ahead of us. In the final analysis, it
is the walking that beats the path. It is not the path that makes
the walk.

Who is doing the steering?

In 1994, an article in the Financial Times described how Britain’s
fifth largest building society, a savings and loan institution, had
remained without a chief executive for 18 months. The lack of a
CEO, noted some outside bankers and financial analysts, could
affect the society’s credit ratings. Yet, continued the article, ‘The
existing management has acquired a better-than-adequate track
record.’8
The society had been run all that time by its finance direc-
tor, its commercial director and its information systems manager.
This triumvirate worked closely with a ten-person management
committee and a chairman who was available two days a week.
 THE TOLERANT COMPANY 187

Acting without a clear power centre, this banking institution

increased its profits in 1993 by 22 per cent and in the first quar-
ter of 1994 by 37 per cent. Remarked the Financial Times: ‘The
longer this has continued, the more Leeds executives may ques-
tion what even a paragon [CEO] will bring to the organization.’9
Only one doubt, in fact, remained: ‘Competitors accept that [the
triumvirate] has done a competent job and have the backing of
Leeds’ staff, but there remains a question about strategic
10
issues.’
Think carefully about this latter concern. Why should there
be a problem about strategy? It was clear that the organization had
one, and an ever-evolving one at that. Otherwise it could not have
made decisions for a year and a half. In other words, in this com-
pany a CEO was not necessary to elaborate a strategy.
But in fact a CEO might have had an immediate negative
effect. An incoming chief executive, who was likely to come from
the outside, would probably take a thoroughly different view of
the way forward. He could use his position of power to make a
U-turn away from a demonstrably profitable policy. All he would
achieve would be to create uncertainty in an otherwise well-
performing organization.
The moral of this story is clear. A company does not uncon-
ditionally need a single hand at the tiller. The personal use of
power to steer does not necessarily serve a healthy purpose. It
could reduce the number of brains engaged in the ‘planning-as-
learning’ activity, and it could be seriously disorienting.
Many companies are similar to this savings and loan insti-
tution. To be sure, there may be a CEO who espouses strategy
and sets policy. But the real decision making occurs in a diffused,
tolerant, ‘planning-as-learning’ environment.
 ECOLOGY 188

Setting the context as an alternative to steering

In all but the smallest of companies and in all situations but a cri-
sis, it would be wise to be careful with an action-oriented man-
agement style. This does not mean that we do away with all
leadership and hierarchy. Management plays an indispensable
role, but it is not in the first place one of looking out in the mir-
ror, deciding on the route and opening the throttle.
It is the company as a whole that beats Machado’s path
while it walks. The company as a whole should scan the environ-
ment, decide on the next step, determine where to set a foot and
when to do so. To engage as many brains as available and possi-
ble in this process is a complicated matter.
The art of managing — of coaching such a community along
its path — becomes a matter of setting the context for the rest of the
organization’s members to perform that task at their level.
This cycle of seeing, concluding, deciding and acting is, of
course, the cycle of continuous learning described in Chapter 4.
In this sense, strategy is simply the development of the organiza-
tion’s ability to learn. The organization’s ability to learn faster
(and possibly better) than the competition becomes its most sus-
tainable competitive advantage.
Senior management must set the context and the process
within which the maximum of the organization’s available brain
capacity engages in continuous learning. The design and opera-
tion of these processes are line management responsibilities
which cannot be delegated. Rachel Bodle, reporting on a series of
workshops on change management, describes the demands on
leadership which occur in a turbulent world. The new forms of
leadership, she writes, involve:
 THE TOLERANT COMPANY 189

demonstrating willingness to ‘let go’ in a new, open, more

informal and less hierarchical style of management; allocating
time to get wide recognition and acceptance of the need for
change; encouraging risk taking by ensuring that mistakes are
not penalized if there is learning; dispensing with authority
and suspending hierarchy so that learning can take place; pro-
viding necessary feedback; [and] creating an environment for
teamwork.11

Management can be helped by planners in this effort. If manage-

ment is a learning process, then planners can be helpers or
enablers in the learning process — and no more than that.
Learning starts with perception. Planners can encourage the com-
pany to look up more often and to look further ahead. They can
help make the learning process more deliberate and make sure
that more of the available brainpower is engaged. They can devel-
op methods for making sure that individual inventiveness
becomes shared knowledge before the next corporate foot is put
forward.
So, if strategy is something you do, I have little doubt in my
mind that this doing actually constitutes learning, not steering.
This embodiment of learning within the strategic process, in the
end, also determines the organizational role of the scenario work
that we examined in Chapter 3. Scenarios institutionalize a
process of learning (and exchange of learning) throughout the
company. With that in place, the company can afford to tolerate
its diversity of marginal activities, because the act of learning
keeps the company cohesive, without the need for strict regula-
tion by authority.
9
The Corporate Immune System
NO MATTER HOW VITAL TOLERANCE MAY BE FOR A LIVING
company, there are dangers in tolerance and openness. These
are not the dangers that most managers associate with tolerance:
they have nothing to do with loss of internal control or with inef-
ficiency. Like human beings corporations have immune systems,
and when the corporate immune system cannot cope with the
openness of a tolerant company, the resulting stress puts the
entire organization in danger.
In the human body, the immune system is built into the
cells within the bloodstream. Its role is to maintain equilibrium
with the intruders from the outside world which continuously
enter the body. The active cells of the immune system can detect
these outside organisms and, if necessary, secrete chemicals to
defend the body against them. The more capable the immune sys-
tem, the more resistant the host body is to harmful effects from
bacteria, viruses, fungi or parasites. These intruders from the out-
side are not prevented from entering, nor are they systematically
destroyed on every incursion. The immune system keeps them in
check while they exist within the body.
Francisco Varela describes the functioning of the immune
system in this manner. If you take an aeroplane, for example,
 THE CORPORATE IMMUNE SYSTEM 191

from New York City to Rio de Janeiro, you move your body
between two very different molecular soups — two environments
of completely different bacteria, fungi and viruses. When you get
off the plane in Rio, your immune system begins to ‘recruit’ new
cells from the regular flow of lymphocytes that your body pro-
duces in the spinal column and releases into the bloodstream.
Some of these potential antibody cells will be called into action,
depending on the new molecular environment in which you are
wading.
Each day, your immune system recruits enough new cells to
constitute up to 20 per cent of its own population. Your body,
during that recruitment process, rebuilds an equilibrium with the
unfamiliar intruders of the Rio atmosphere. This makes the
human immune system, according to Varela, into an extremely
open system — and thus a good learning system.1
Unfortunately, there are limits to the number of intrusions
that the human body can handle. The tolerance of any body has an
upper limit. Thus imagine that you are put into an environment in
which more than 20 per cent of the invaders consist of a new type.
Or suppose you are invaded by a particularly virulent organism.
In that case, the intruder from the outside is numerous or power-
ful enough to exceed your immune system’s capacity to learn.
Now you have an infection. Instead of learning and incor-
porating the new molecules, the immune system must repel them.
It increases the body temperature to make life more difficult for
the new intruders, and it develops a composition of antibodies
and killer cells to cope with new types of intruders.
Without thinking, your body has moved into a mode of
resistance and rejection. This is not necessarily beneficial for you.
You experience a fever, exhaustion, headaches and tension in your
digestive system. The reaction to the entrance of a foreign sub-
stance may even have more severe effects. Your own resistance
 ECOLOGY 192

and rejection may upset your own equilibrium so much that it

could lead to shock or even death.
This state of affairs is as true for corporations as it is for
individuals. The health of a company is under constant attack
from the inside and the outside. The attacks come from individu-
als or groups of individuals who do not want to be part of the
whole. They are there for their own purposes. It does not matter
how honourable or dishonourable these purposes may be. The
health of the community is under threat.
On the inside, the threats may come from groups of indi-
viduals who are not full members of the community. (They may
see themselves that way, or they may have been deliberately cast
as non-members at the time of recruitment.) They are insiders
and intimately familiar with the company. They are the ‘hands’
that operate the machines. But they are not part of ‘us’.
Recently, these internal non-members have grown in num-
ber. During exercises of staff reductions and downsizing, many
people who had thought themselves members have been told of a
sudden alteration in their status. They may still be needed for
capacity reasons or because the law made outright dismissal diffi-
cult and time consuming. But they have been told, explicitly or
implicitly, that they should no longer see their future as aligned
with that of the company.
We should not be surprised to discover that these internal
non-members will often feel detached from the company or
angry with it. Many will feel the need to affiliate with another
institutional body such as a trade union, if only to defend them-
selves. If that happens, there are then two institutional bodies, a
company and a trade union, cohabiting the same space with very
different goals. This will inevitably have some impact on corpo-
rate health; possibly benevolent, possibly malevolent, depending
on whether these two personæ will elect to have a symbiotic or an
 THE CORPORATE IMMUNE SYSTEM 193

inimical coexistence.
When a company tells some of its members that they are no
longer ‘us’, the company also defines a new group: ‘who is left of
us’. If that definition is made very narrowly, the company may
find that it has few real members left with loyalty to the whole. In
some cases, too few members are left. Those who still fall under
the definition of ‘us’ should understand beyond doubt that they
are part of the whole. The company should make it doubly clear
to them that the development of their full potential is still reason-
ably assured by holding together in a symbiotic relationship.
Other threats to the health of the company come from out-
side. These might include customers with a different set of expec-
tations; new teams of decision makers from a merger or
acquisition; or key members of the corporate community who
just don’t seem to ‘fit right’ with the prevailing ethos.
Just like the human body, the corporate body needs an
immune system which can treat different types of intruders
appropriately. For one thing, just as with the human body, the
vast majority of intruders are beneficial. Humans are constantly
invaded by bacteria and viruses whose impact is symbiotic with
our own health and survival. The same is true of companies; most
invaders bring with them new perspectives and capabilities.
Moreover, even if we wanted to, we could not keep intruders out
of either a human body or a corporation; they cannot be kept out
of an open system. Openness inevitably means that something
other than yourself enters your body.
But you can react to these intruders discriminatingly. Some
may indeed have to be repelled, even at the cost of upsetting the
company’s equilibrium. Some should be managed, so that their
encroachment comes gradually enough to let the immune system
build up a response. And some should be embraced as necessary
vehicles for the organization’s learning.
 ECOLOGY 194

Acquisitions and mergers

Consider, for example, the case of acquisitions and mergers.
These represent one of the irresistible temptations to which man-
agers of economic companies are subjected. Acquisitions and
mergers carry the double promise of ‘buying knowledge off the
shelf’ (gaining capabilities without having to go through a change
or conduct any learning yourself) and making the company big-
ger (and thereby, one hopes, stronger) in an adversarial world.
But ask people who have lived through an acquisition or a
merger about the experience during the first few years. It is not
unlike the first three or four days after catching the flu. The tem-
perature of the institution is quite high; it’s as if there’s a ‘fever’ in
the air. There are headaches; people feel an uncanny, unpredicted
exhaustion. And there may be a sense that the larger company
can’t quite ‘digest’ the smaller one. In fact, all sorts of rejection
mechanisms are in place. Most notably, during the first four or
five years after a merger and acquisition, there is an enormous
shift of individuals entering and exiting the affected units. The
institution is mobilizing its immune system — and rejecting a great
deal of the incoming culture.
In biological terms, a merger or acquisition represents an
intrusion of foreign bodies, ideas and values into the host organi-
zation. If you acquire a competitor who is 25 per cent of your
size, you have accepted an intrusion of 25 per cent of your exist-
ing population. A 50-50 merger with a foreign partner means an
intrusion of 50 per cent for each of the two partners.
At these proportions, many acquisitions and most mergers
are likely to rise well above the tolerance limits of both partners.
Acquisitions and mergers are infections. That is why the temper-
ature goes up and the corporate body goes into resistance mode.
 THE CORPORATE IMMUNE SYSTEM 195

Many cases of this resistance have been well documented.

Management professor Michael Porter analysed 2700 mergers
and acquisitions by 33 major US companies over a 36-year peri-
od (1950-86); his report, published in 1987, found failure rates
between 50 and 75 per cent (failure meaning ‘disposal’). A Dutch
study in the prestigious journal Economisch-Statistische Berichten
found failure rates of up to 60 per cent in similar situations.2
Moreover, many managers carry the scars of their own per-
sonal experience. After a merger, two internal tribes look at each
other in distrust, personnel policies remain affected by pre-
merger values for a long time after the original intrusion, and staff
will feel deep levels of distrust and suspicion about every major
personnel move.
I am well acquainted with a number of people who were
involved in the merger of two banks in the Netherlands. This
merger took place 20 years ago and the people I know are now 60
years old or older. Yet they can still remember vividly which of
the pre-merger employees worked for the other bank. To them,
the other bank still represents an infection that was never fully
assimilated.
The tribes within the merged institution continue to see
themselves as separate entities. If I come to your country as part
of the merger of our companies, you will still see me as an outsider.
You will trust me less than someone from your home institution.
On the next promotion you make, you will seek out someone
from your own bank; and on my part, I will attribute my failure
to get the job to the fact that you are a member of the other tribe.
Thus we will trust each other even less as time goes on.
Royal Dutch/Shell experienced this problem for 50 years. It
was formed through the merger of two parent companies: the
Royal Dutch oil company and the British firm Shell Transport and
Trading. They merged in 1907; but when I joined in 1951, the two
 ECOLOGY 196

parent companies still had quite separate organizations based in

The Hague and London, respectively. It was as if two single par-
ents had come together in marriage, putting all their children
together in one household, but sleeping in separate bedrooms. The
marriage was not consummated until a series of personnel policies
introduced in the mid-1950s began to integrate the worldwide
Shell community, followed by a McKinsey reorganization which
finally created one integrated organization spread over the two
central offices.
Another Shell example dates from the period in the early
1970s when ‘diversification’ was fashionable. Royal Dutch/Shell
bought a medium-sized metal company, Billiton. In terms of peo-
ple, capitalization, turnover or any other measurement, the Shell
Group was so much bigger than Billiton that Shell could absorb
this acquisition without any difficulty. However, by any stan-
dard, Shell’s intrusion was way above Billiton’s tolerance level.
Shell did not show a rejection mechanism, but Billiton did.
The Billiton entity ailed and, in effect, died. Less than a decade
later, nearly all of the senior Billiton managers had left — notwith-
standing the utmost care Shell had exerted to leave the original
Billiton management in control. We had tried to prevent Shell man-
agers from overpowering the new family member, but the ecologi-
cal relationship between the two companies made that impossible.
For Billiton, the merger represented an infection and had to be
repelled. This meant that Shell was unable to reap the benefits that
we had hoped might emerge from the Billiton merger. Shell was
simply too big for Billiton.
There is no easy way to manage mergers and acquisitions,
once the bankers have left. The infection analogy is useful, in my
mind, because it shows why money cannot buy intelligence,
knowledge and innovative new behaviour in large quantities.
Learning is not a matter of ‘filling up the tank’ by buying a new
 THE CORPORATE IMMUNE SYSTEM 197

partnership. Learning is a process. Partnership takes time. And if

one partner has too much power to interfere, whether it intends
to interfere or not, then the merger and acquisition process will,
in itself, set off resistance mechanisms which defeat the original
purpose that brought the merger into being in the first place.

Parasites
The less a company operates in control of its environment, the
more open it should be: foreign bodies and ideas will be able to
enter easily. That is as it should be; indeed, it becomes a strength
of the company. However, the company can never be sure how
these bodies and ideas will behave, once inside. Every intruder
has a choice: it can select a symbiotic relationship or it can pursue
its own benefit, to the exclusion of all others.
All intruders are not alike. Richard Dawkins, who has writ-
ten at length about the role of intruders and parasites in evolu-
tion, describes them all as egoistic. None of them ‘cares’ about the
welfare of the host body, except as a vehicle for its own survival.
They serve their own genetic interests. At the same time, how-
ever, many of them serve the host body well: they are symbiotic,
increasing the sophistication and capability of the host body at
the same time that they remain dedicated to their own interests.
There are bacteria which live in beetles, for example, and use the
beetles’ eggs as transport into the bodies of new beetles. They do
not obstruct the reproductive process of the beetle; indeed, they
depend on it.3
Others are parasitic: they bring lasting damage to the host
body. Dawkins offers a fascinating explanation of the difference
between a symbiotic intruder and a parasite. The parasite plans its
exit on its own terms. Any other intruder exits through the host’s
natural functions and systems, such as excretion and procreation.
 ECOLOGY 198

A parasite breaks through the host’s natural functions, exiting in

a way that may destroy or harm the host.
Dawkins gives the example of a snail that is infested by a
parasite (a fluke of the genus Leucochloridium) which burrows
into its tentacles and causes them to swell. Those swelling tenta-
cles (which happen to contain eyes) in turn protrude more obvi-
ously and look like tempting morsels to a particular type of bird.
In this way, the snail becomes more susceptible to being caught
by that bird. And the fluke now gains a larger host body, the bird,
to enter. This is part of the lifecycle of that particular parasite.
If you see a snail with a protruding eye and you want to
know why that eye evolved, you should not necessarily ask,
‘Why is a protruding eye in the interest of that snail?’ Instead,
says Dawkins, you ask, ‘Whose interest is served by the protru-
sion of this eye?’ And that question will lead you to the parasite.
Destructive parasites can also exist anywhere in the corpo-
rate host body. They can be excluded individuals or even individ-
uals in positions of power, but planning their exit on their own
terms. Power can be used to manipulate the definition of ‘us’ in
the service of someone else’s strategy. A senior manager manipu-
lating a situation to make his or her own résumé look good, but
leaving all the rest vulnerable, is behaving parasitically. Similarly,
when a division of a company resents being part of the whole,
that division can easily become a parasite in the host body. It does
not matter whether this resentment is justified. (For instance,
when the added value of the company’s head office is less than the
costs which the host office imposes on a resentful outpost, that
outpost is still a destructive parasite.)4 All of these people and sub-
systems, in Richard Dawkins’s sense, are serving their own self-
interest at the expense of the natural functions of the host
company.
If a company begins to perform seemingly self-destructive
 THE CORPORATE IMMUNE SYSTEM 199

acts, you should not ask, ‘Why is this activity in the interest of the
corporation?’ You should ask, ‘Whose interest is served by this
self-destructive act?’ Is it the small group which has misused its
power to define the company as only the five or six top people?
Is it the large intestinal snail called a partner company, a division
or a trade union?
This biological view does not distinguish between inherent-
ly good or bad behaviour — even if it were possible to define what
is good or bad. Consider, for instance, a highly intolerant institu-
tion which hires a group of consultants to improve its practices.
New ideas and new people have now entered the firm, with the
intention of opening up the institution’s behaviour and improv-
ing its chances of survival. This intent is assuredly ‘good’ — but to
the existing (intolerant) membership, the intruders’ behaviour
will seem ‘bad’ and parasitic. The corporate immune system will
go into action. The temperature will go up and killer cells may
finish off the intruding ideas. When this happens, it should not be
taken as a comment on the quality of those ideas. It is a reaction
to the strength of the host system’s immunity mechanisms.

Corporate symbiosis
We normally think of intruders as parasites and of parasites as
entering with the intention of weakening their hosts. This need
not always be the case. On entry, every intruder has the same
choice: a symbiotic relationship or a parasitic one. Indeed, there
is a great deal of leverage available from cultivating symbiotic
relationships with organizational parasites. To understand such
relationships, the critical question is this: why would certain peo-
ple or substructures work with a kind of group loyalty to one
another when others would not?
Dawkins answers this question as Stern did before him: all
 ECOLOGY 200

the members who stand to gain from tying their fate to the host
institution will ‘cooperate’. They will work together to make the
whole institution behave as a single coherently purposeful unit.
The primary difference between members and parasites will thus
have to do with their method of exit. Members will retire, where-
as parasites will serve for their own sweet time and leave by a dif-
ferent route.
Under those conditions, it is clear that management’s
responsibility for guarding the corporate health is best served by
preventive medicine and setting a context for mutual cooperation.
Make sure, when a new member enters, that there is a shared
value system in place. Ensure that there is a contract based on
long-term harmonization of goals, as we discussed in previous
chapters.
In this way, the company has the highest certainty that
entrants will select member status, instead of becoming parasites.
Money is not enough of an incentive. If the salary and bonus lev-
els represent the sole or the most important condition of the con-
tract between company and individual, the chances are increased
that the lure of larger amounts of money will lead to parasitic
behaviour. This will be even more true for people who are in a
closer position to higher amounts of money. The management
levels in a company are the most propitious, from a parasite’s
point of view!
Money, in fact, needs a lot of attention in a living company.
As every businessperson knows, money has as many attractions
as it has roles to play within a company. It is the means by which
the company gains access to resources — by which it remunerates
its people and its shareholders. It can be the life-blood that
attracts predators and parasites, but it also plays an important role
in the evolution of a company over its lifetime. Let us, therefore,
turn our attention there next.
Part IV
Evolution
10
Conservatism in Financing
IN PART 3, ECOLOGY, I MADE THE CASE THAT TOLERANCE AND
learning create the necessary conditions for a company’s
longevity and survival. Openness improves learning; without an
effective learning capability, a company cannot hope to ‘evolve’
effectively in an unpredictable world. Evolution is the process of
a company’s development, and managers who are attuned to this
can influence the speed and the means (the ‘how’) of the evolution
which they set in motion.
Intuitively, we can see that money must play an important
role in a company’s survival and evolution. With money, a com-
pany can buy resources (and time) which allow it to evolve more
rapidly. In addition, people who see companies as machines for
earning money measure the success of a company’s evolution by
the extent to which it can make more money than its competitors.
Equally, the success of entrepreneurs (those particularly
successful participants in the corporate game) is generally mea-
sured in terms of the amounts of money they can produce — either
in the size of their business’s revenues or in the amount they can
return to the company’s owners.
If the amount of money is the primary measure of corpo-
rate success, then it is obvious that no company could ever have
 CONSERVATISM IN FINANCING 203

enough cash coming in. But is that the only function money
serves? Could it also take part in governing the evolution of a
company? Could it be that having too much money could lead to
quick or unbalanced growth and development? Or that leaving
too little money in the company by siphoning off too much for
individual members (or parasites) could reduce the survival
chances of that company?
Many people will immediately agree with the latter state-
ment: an unbalanced distribution of money could well threaten
the future viability of the company. But how easily that can be
done and what form it would take are illustrated by a talk that
Dick Onians, a managing partner of the Barings Venture fund,
gave at the Royal Society of Arts in London in 1994. Barings, he
said, had invested in about 200 company startups during the pre-
vious decade. But only 40 of them had developed into profitable
and sustainable businesses. Of the remaining 160 companies, 20
per cent had failed outright. The other 120 companies, he said,
faced three probable destinies: ‘to be acquired by a larger compa-
ny; to merge with competitors; or to regress to being life-style
businesses for a small group of owner-managers.’1
In short, although they might survive, they would not sustain
themselves as the sort of entities that had been foreseen. To under-
stand the high mortality rate of these newborn startups, and to dis-
cover the factors behind the few successes, Mr Onians conducted a
review of ten successful startups and ten failures. Marketing, strate-
gic positioning and product development, he found, were impor-
tant; but they were subsidiary to another set of factors. The
companies’ survival rose or fell on the way they managed their
resources: their people, their information and their money.
In earlier chapters, we have looked at the way companies
manage their information (in the part on learning) and their peo-
ple (in persona and ecology). But the nature of money is
 EVOLUTION 204

particularly misunderstood. Every living entity consumes; and

money serves, in a large corporation, as the way of measuring
what has been consumed. As a result of this role, when properly
managed, the financing of a company becomes the governor on a
living company’s growth and evolution.
In this case, I do not use the word governor in its purely
political sense: to imply a state leader or authoritarian director.
The financing of a company is a governor in the sense of a regu-
lator — like the valve on a furnace, which adjusts the flow of fuel
and thereby indirectly controls the production of heat.
For the startup businesses that Mr Onians studied, financ-
ing would act as a governor by regulating the incoming flow of
cash and thereby indirectly modulating the growth of the compa-
ny. Apart from its own cash generation, there were only three
ways that further financing could become available. The compa-
ny could borrow. It could accept equity capital in return for
shares. Or it could engineer a hybrid of the other two methods.
Borrowing to finance the startup was an almost irresistible
temptation. For, as Mr Onians put it, that would allow the
founders to maintain control of the firm:

By working with borrowed money, management can retain a

larger share of the equity — indeed, ideally keep 100%. Thus,
by remaining the dominant owners, management can award
themselves salaries, perquisites, bonuses, pensions, and even-
tually, dividends with no deference to other shareholders or
their board representatives.2

This may seem a sensible way to build a business: get the bank to
cough up the money, pay it its due in the form of interest, and keep
all the value accumulation to the inner circle of founders and ini-
tiators. At the same time, with borrowed money, there is a much
 CONSERVATISM IN FINANCING 205

wider limit to what a company can accomplish in its early years.

But consider the results, as Mr Onians and his colleagues
found them in practice:

Entrepreneurs [with] high debt, high expense, [and] low equi-

ty under perform entrepreneurs who go for high dependence
on share owners’ funds. Nine out of the ten failing businesses
were highly dependent on short term debt. In five of these
cases it was their so-called ‘friends,’ the bankers, who pulled
the rug from under them and dictated . . . the terms and condi-
tions for the demise of their businesses.3

The picture was very different in the ten successful companies, all
of which eventually became significant international businesses.
Eight out of the ten had never had a loan. They were entirely debt
free and always had been. The two companies which had bor-
rowed money had done so to meet specific short-term needs.
They had since repaid the debts in full.
Conservatism in financing, in short, is not merely a conceit
of a former, less credit-happy age. It seems to be an essential con-
dition for companies which hope to survive to a ripe old age.
When companies know how to ‘listen’ to their financing, they are
ready to follow the path of a natural, longlived evolution.

Money as the governor of evolution

At Shell, we had found something similar in our report on longlived
corporations. Nearly every company over the average age had a
conservative approach to its financing. If not debt free, then they
were rigidly careful about their borrowing and investment capital.
In short, they knew the value of having money in the kitty.
At first, it may seem paradoxical to limit the speed of a
 EVOLUTION 206

company’s growth by its capacity to generate its own funds. In its

growth stages, any company which is willing to borrow can the-
oretically operate without constraints. Such a company has the
options that stem from not having to rely on its own money
reserves or the equity it can raise from investors. This makes rev-
olutionary change possible, to be sure; the results are more
instantly dramatic and the change may indeed be more effective.
But growth through borrowing money, or through mergers
and acquisitions, is dangerous precisely because it is not con-
strained. At some point, the pendulum will shift. Having to ser-
vice your debt, you lose the options that come from having ‘spare
cash’. You can no longer choose your moment.
Longlived companies know that having money in hand
means that they have flexibility and independence of action, when
competitors do not. Having built up their business organically,
they can grasp opportunities without having to convince third-
party financiers of the attractiveness of their decisions. They can
even make business decisions without having to depend on pure-
ly financial considerations. Money in hand has made them mas-
ters of their own timing.
In this way, conservatism in financing serves as a governor
to maintain your company’s evolution at an appropriate speed.
That does not necessarily mean a low speed. James Collins and
Jerry Porras report that Hewlett-Packard managers eschewed
long-term debt, a seemingly irrational policy. But:

by refusing to take on long-term debt in order to fund growth,

HP forced itself to learn how to fund its 20-plus percent aver-
age annual growth . . . entirely from within. Such a mechanism
. . . produced a whole company of incredibly disciplined gen-
eral managers operating with a level of leanness and efficiency
usually only found in small, cash-constrained companies.4
 CONSERVATISM IN FINANCING 207

To follow the model of evolution, a conservative finance policy

helps a great deal, and may even be necessary. A businessperson’s
life is full of irresistible temptations, and the most irresistible of
all is probably impatience. We often get into positions in which
we have the power to foster quick growth, with impressive short-
term results — at the expense of the long-term health of the enter-
prise. If there are negative consequences of this growth, the
consequences won’t be felt for months or years. So, instead of
evolution, we go for revolution. Instead of building an enterprise,
we set out on a quick adventure. We take a gamble.
But good businesspeople are not gamblers. They should be
the opposite; they are stewards and custodians of the company
they manage. Conservatism in financing helps them avoid the
temptation of gambling.
Whenever I mention this point in speeches, I immediately see
roughly half the audience nodding in agreement. Those of us who
have managed the pursestrings of an organization know full well
how virtuous conservatism in financing can be. Why then do many
managers have such difficulty putting it into practice? I believe that
the difficulty begins with our definition of corporate success.

Money as the measure of corporate success

In an economic company, one which sees itself as existing primar-
ily to maximize profits and assets, the criterion for success is clear.
The larger you grow, the more assets you can command. The less
resources you use in the process, the more profits you can deliver.
In this way, the dominant school of thought on business adminis-
tration measures success purely in terms of quantity: the maxi-
mization of revenues, market share, share value or proceeds. A
place in the Financial Times 100 or Fortune 500 is the symbol of
such success, and you only reach that position by growing larger.
 EVOLUTION 208

These criteria belong to the concept of a company as a

money-making machine. Managing money-making machines is
reassuring and comforting. It makes the company feel rational, cal-
culable and controllable. The economic company is thus a coun-
terpart to the homo economicus of my business school professors
in the 1950s: a perfectly rational creature, making choices based on
self-interest — and entirely unrelated to anything in real life.
But a healthy, living company does not measure its success
in terms of money or profits. In the research conducted by Collins
and Porras, for example, there was no indication that success fac-
tors included such economic measures of success as low-cost mar-
keting, being at the cutting edge of technology or being a high
value-added producer. ‘Existing first and foremost to maximize
profits’ was even specifically relegated to a much lower priority!5
Ten years earlier, the researchers of the Shell longevity
study had used soft, non-economic words to describe the suc-
cessful longlived companies. These companies were ‘financially
conservative’, we wrote, ‘with a staff which identifies with the
company and a management which is tolerant and sensitive to the
world in which they live.’
William Stern had written 70 years earlier that the basic dri-
ving force of every living system is the development of its inher-
ent potential. The longlived companies we studied at Shell seemed
to realize this force and to live up to its demands.
Everything about the company — its physical business, its
assets, its policies and practices — was a means for living. None of
these constituted the purpose of the company. Success for the
company meant evolving into the best possible thing it could be
and, in the process, being good at whatever it happens to be doing
in order to survive.
Shell’s purpose is not to deliver oil, to produce energy, or
even to better the material wealth and capability of industrial
 CONSERVATISM IN FINANCING 209

society. It has to be good at those activities to make the profits to

achieve its primary purposes: to survive and to develop new
potential as necessary in an evolving society. Shell does not exist
to pump oil. We pump oil in order to exist.
This view runs contrary to a lot of what is said and written
about companies, both by insiders and outsiders. But it is very
consistent with the way companies act (independent of what they
may say).
It is also very consistent with the way most managers’
minds work when they think about the ultimate purpose of their
company. When executives retreat for a weekend to formulate a
corporate purpose in terms of a mission statement, they rapidly
climb a ladder of abstraction. The phrase ‘British Gas is a compa-
ny to distribute natural gas in Great Britain’ is quickly replaced
with ‘British Gas is a worldwide energy company’.
Mission statements are often justifiably criticized because
they don’t seem to say anything, once all the abstractions are in
place. But the phenomenon of this abstraction is worth noting; it
is so consistent, from company to company, that it must exist for
a reason. I believe the reason is that every businessperson knows,
in his or her heart, that too narrow a definition of the business is
literally life threatening. For long-term survival, a company can-
not be defined in terms of the business it happens to be doing at
this very moment. Like the long-term surviving companies out of
the Shell study, it may have to change its business portfolio sev-
eral times over. In addition, to develop the options it needs for
survival, it may have to go through periods in which return on
investment to shareholders takes second place to reinvestment to
further develop the company’s long-term capabilities.
A long-term survivor, in short, does not define its life in
economic terms, but in terms of its own evolution: the develop-
ment of the entity as a whole, including all the people who have
 EVOLUTION 210

joined the contract with it, so that it and they are part of each
other’s identity.
To live with a sense of purpose so far removed from the
economic definition of companies may be refreshing. For exam-
ple, it may allow for patience. Evolution, in human living systems
and in corporate systems, takes place over long periods of time.
Paradoxically, however, this long, slow process allows relatively
fast adaptation to a changing environment, creating conditions
for the development of the organism’s potential.
I do not argue that all companies should live according to
this principle. Many companies, and their managers, have thrived
by remaining economic companies, especially where they were in
control of their external environment. However, I argue that
managers must be acutely aware of which type of company they
happen to be managing. To espouse the ideas of a living company
while implementing the practices of an economic company could
do great damage, because the management practices which fit one
type of company are incompatible with the other. In the end, the
choice between these two companies will come down, not to the
rhetoric espoused by the executives, but to the way it manages its
information, its people and its finances.
It is in this light that we should consider the concept of cor-
porate growth. In an economic company, growth is an unequivocal
good. But a conservative financier is concerned about the speed and
quality of growth. Will it contribute to the company’s evolution, to
its development to be the best that it can be, fitting with its envi-
ronment? If not, then that growth will not be seen as successful.

Money as the expression of corporate reality

Just as most company manuals represent yesterday’s write-up of
the day before yesterday’s solution to last year’s problems, so the
 CONSERVATISM IN FINANCING 211

law tends to run similarly far behind reality. For example, there
are distinct differences between the way banks behaved 50 years
ago and the way they deal with corporate clients in distress today.
But bankruptcy laws are based on these long-deceased banking
practices of the previous half-century.
Worse still are the prevailing laws — and the social attitudes
behind them — concerning the role of managers. Corporate law in
many Western countries proclaims the investor, as the capital sup-
plier and asset owner, to be the carrier of ultimate power: the
power to decide the life or death of a company. Managers are
supposed to optimize capital before all other concerns, or else
they may be liable for damages. But these laws, which give top
priority to the rights of shareholders, are based on the assumption
that the human elements are mere extensions of the capital assets.
As a result, in concentrating our minds on the optimization of the
present capital asset base, we managers run a serious risk of short-
ening the lifespan of our companies.
This is one of today’s great dilemmas. And enlightened
chief executives understand it very well. They recognize the value
of community and of building trust. But they might break out of
a conversation about these values by saying that the next morn-
ing they have to address a meeting of financial analysts, or confer
with the corporate lawyer. In other words, they are caught
between two imperatives: the external realities created by the law
and the internal needs of managing with knowledge as the most
critical factor. Many CEOs cope by saying one thing in public
while they face a different reality inside the company.

New governance forms in a living company

Shareholders are suppliers of capital in the same way that banks
are suppliers of money. They have a legally different position and
 EVOLUTION 212

receive a slightly different form of remuneration. Instead of inter-

est payments they receive dividends, and they can sell their stock
at a profit (or loss). Although they are officially the ‘owners’ of
the company, their functional role (from the company’s point of
view) is merely that of suppliers of money. They provide cash in
return for equity.
To an economic company under today’s legislation, that
role as suppliers of money is all that shareholders need for them
to be considered as owners. After all, an economic company
exists in an environment where capital is the most critical asset,
and shareholders are the suppliers of that most critical asset.
Managers in an economic company know that their success
depends on showing results — fast. And they get very little
patience from shareholders or other outsiders: the law or political
regulators. They cannot talk freely in public about the need for
reinvesting in the company to develop new, long-term potential
or to build a cushion of cash to provide options for evolution.
They find it difficult to say, ‘Look, we will have our ups and
downs. But 10 or 20 years from now, you’ll have returns far
beyond anything you could extract from the company today.’
Even if some shareholders were sympathetic, the company
would be legally vulnerable. More seriously, it could throw itself
open to an attempt by raiders forcibly to take over the firm.
Many people who deal in investment aren’t interested in long-
term futures. They deal in present-day value. The discount fac-
tors in capital are so high that 20 years from now is practically
worthless to the shareholder.
A living company, however, cannot have the same relation-
ship with shareholders. To a living company, these suppliers of
money are much like the other stakeholders in the external envi-
ronment: unions, material suppliers, customers, shareholders,
local government bodies and the community around the plant.
 CONSERVATISM IN FINANCING 213

All of these are critically important forces in the outside environ-

ment. The company must remain in harmony with all of them.
But they are not members. They are not part of the company’s
persona. It is not necessarily in the best interests of the company
to obey them.
Yet they are critical parts of the company’s external envi-
ronment and the company should be engaged in constant conver-
sation with them. Such dialogues are, unfortunately, rare in the
current system. Conventionally, shareholders and managers do
not converse. Even in a takeover, the new owner of 51 per cent of
the stock does not hold an in-depth conversation with the man-
agers of the old regime. The new owner simply enters and dic-
tates. The majority of shares conveys that legal power, and the
persona of the living company cannot hold up against it. After all,
the new dictator has not been a member of the old company. He
has not entered the river at any point. He has not contributed or
gained the trust of others in the enterprise. He has simply bought
his way into the system. He will be an infection and the old sys-
tem will respond by resisting ... or by dying.
Under those conditions, many senior managers turn living
companies into economic companies. They are not wilfully
destructive. They are simply trying to do their best against the
background of a terrible dilemma. When shareholders and out-
side regulatory bodies ask about the senior managers’ results,
they do not ask about efforts to improve community. They do
not enquire into the company’s prospects for a long and prosper-
ous existence. They ask, ‘What is your return on capital
employed?’ ‘Aren’t you overcapitalized?’ and ‘What is your pro-
ductivity?’ In many cases, because there is no communication with
managers as members of a common entity, the shareholders will
apply any pressure they can to force a fast return on assets at the
expense of the company’s long-term development.
 EVOLUTION 214

To change this situation is beyond the power of any single

company. The shareholder-manager relationship may not express
corporate reality very well, but it is embedded in the law. As such,
the law is an anachronism. It is yesteryear’s write-up of the situ-
ation before the Second World War, when capital was a scarce
resource that deserved special protection and management atten-
tion to optimize its use.
In the intervening 50 years, capital has become much less
scarce than it was before, even during the postwar years. In addi-
tion, the personalities of the suppliers of capital have changed. The
link between the individual who makes the original saving and the
commercial institution which finally obtains this saved capital to
use in its wealth-creation process has become far less direct than it
was half a century ago. Much shareholder power today is held by
institutions such as large banks and pension funds, which pursue
their own institutional interests. These interests can quite easily be
detrimental to the company in which the shares are held.
The resulting tension between the business realities and the
legal fiction of corporate ownership is evident for everybody to
read in the daily newspapers. Some companies act according to the
legal fiction and sacrifice thousands of people to safeguard assets
or improve profitability. Others will focus on the business realities
and scuttle hundreds of millions of dollars to keep themselves
alive. In the choice between sacrificing assets or sacrificing people,
business no longer gives automatic preference to the capital factor
as the law allows or sometimes even demands. This is not because
these managers have become more socially aware or people con-
scious, but because it makes good business sense. A company
which pulls through a crisis with its human talent mostly in place
carries a real promise of a better outcome, both for its capital sup-
pliers and for its people. Yet managers who act in accordance with
this reality make themselves vulnerable to the law.
 CONSERVATISM IN FINANCING 215

National legislation will begin to reflect the emerging reali-

ty of the new era only after these issues have crystallized in a pub-
lic debate. That will take time, perhaps years. I write this with
regret because, in the meantime, there is much room for the abuse
of the dominant legal powers given to the capital suppliers. Many
more companies will be thrown wide open to parasitic and preda-
tory behaviour. This could, in fact, be a generic reason for the low
average life expectancy of companies.
Many readers of this book will be shareholders themselves
and all of us are citizens. In both these roles, we should be con-
cerned about the price of corporate demise. The disappearance of
a company is not without cost. A company’s untimely death is
hurtful and damaging to almost everyone concerned, including its
shareholders.

Why not let corporations die?

Some years ago, a Dutch television reporter asked me, after a
speech I had given on the living company, ‘Why is it so important
that Shell should survive?’
The question took me by surprise. To me it was natural that
companies should seek their own survival. I had seen nothing else
in my life. Companies struggle to keep going and to grow, for as
long as possible.
But the reporter was not asking about the company’s inten-
tions. He was questioning its value. What is so special about Shell,
or any other company, that it should continue to clutter the earth
with its presence? Why should it continue to exist when its utili-
ty for humanity diminishes or disappears?
The question takes on added relevance in an era when insti-
tutional shareholders and corporate raiders approve the buying of
companies and dismantling them into component parts. We scrap
 EVOLUTION 216

a car at the end of its useful life. So, why not scrap a company?
It is a good question — if, indeed, the company is an eco-
nomic company. Then it is a machine to produce a product, or to
produce money and profits. And machines can be scrapped when
their utility has ended.
But if it is a living company, then it is a persona, with a com-
munity of people embedded in it. And most people do not feel it
is ethically correct to scrap a persona or a community.
All living beings have a right to exist. At the purely biolog-
ical level, the question is not even asked. In nature, those who
survive have earned the right to live.
Companies, as it happens, have a terrific internal will to live.
We rarely hear of a company committing suicide. It can be done;
all that would be required is liquidating the assets and giving the
shares back to the shareholders. In liquidation, the solid body of
the company is literally converted into fluid capital. I believe this
deliberate act is far less frequent, per capita, than suicide among
people.
It is hard to imagine that corporate suicide can be done with
decency. Indeed, when the subject of liquidation is mentioned,
most boards refuse to discuss it. They know that the company is
essentially charged with the purpose of surviving.
It is true that sometimes partial suicide is contemplated — in
the name of giving the company back to the shareholders.
Exxon’s managers, during the early 1980s, boosted their share
price by spending up to $2.5 billion per year of the company’s
money buying back their own shares. They had more money in
the kitty than they knew how to invest profitably in their core
business. Perhaps the correct analogy is not a suicide attempt but
a slow dissipation. And yet Exxon is large and wealthy enough
that it can afford such episodes; other companies would find
themselves not merely weakened, but imperilled.
 CONSERVATISM IN FINANCING 217

It is far more likely that an organization will opt to stay

alive even when its purpose disappears. In the United Kingdom,
there was a very active anti-apartheid organization which held a
great meeting after the election of Nelson Mandela to the presi-
dency of South Africa. Clearly, apartheid was no longer a threat.
So the organization decided to combat international racism.
Anything to keep alive! Most companies seek life with equal pas-
sion and zeal.
In itself, however, that argument may not be enough to con-
vince people like the Dutch journalist. A more convincing argu-
ment arises when we look at the cost, to society and the rest of us,
of letting corporations die prematurely.

The price of a company’s death

Consider what happens when a company dies:

✦ The community of people bound up with the company is torn

apart. People lose jobs and are set adrift, without a work
community.
✦ The debt the company owes to its previous generations, who
gave themselves to its future, can no longer be fulfilled.
✦ And the company’s constituents — its customers and suppli-
ers — are bereaved.

This is a particularly acute problem in developing nations, where

private companies are often responsible for great portions of a
country’s infrastructure. In some African countries, Shell sup-
plied as much as one-third of the oil supply. The country’s social
fabric would disintegrate if we stopped. To be sure, some other
company would take over the function, but that would still
involve a painfully difficult transition period — particularly if the
 EVOLUTION 218

reasons for Shell’s death had nothing to do with its performance

in that country. The shock could, in itself, be difficult for the
country to recover from.
People do, in fact, mourn when a company dies. A prema-
ture death means unnecessary suffering in the environment of the
deceased company. A company’s demise causes disruptions, loss
of values and moral and physical danger — both to people and to
other companies. Witness the mill towns in New England, or the
consequences of dying industries in the UK Midlands. In the
midst of the Great Depression 60 years ago, the British writer
J.B. Priestley formulated the need for continuity in business as
follows in his English Journey:

The industry had to be ‘rationalized’; and the National

Shipbuilders’ Security Ltd. proceeded to buy up and then close
down what were known as ‘redundant’ yards ... Stockton and
the rest were useless as centres for new enterprises. They were
left to rot. And that perhaps would not have mattered very
much, for the bricks and mortar of these towns are not sacred,
if it were not for one fact. These places left to rot have people
living in them. Some of these people are rotting too.

Such people appear to have been overlooked; or perhaps

they were mistaken for bits of old apparatus, left to rust and
crumble away ... You may do a good stroke of work by
declaring the Stockton shipyards ‘redundant’, but you cannot
pretend that all the men who used to work in those yards are
merely ‘redundant’ too ... The planning did not take into
account the only item that really matters — the people.6

This type of community destruction is often described as an

inherent problem with capitalism, or as the dark side of multi-
national companies. But it is actually a phenomenon that takes
 CONSERVATISM IN FINANCING 219

place whenever living companies decay into mere economic com-

panies and then pass away. If far more companies lived somewhat
longer, we would see far fewer such scenes.
A company’s premature demise is almost certainly equally
damaging to its shareholders. The present-day value of an average
50 years of maximized profits is quite likely to be lower than the
present-day value of 200 years of moderate profits combined
with expansion of the company’s activities in all the areas for
which it is able to develop potential.
James Collins and Jerry Porras found strong evidence for
this in their research into visionary companies:

Visionary companies attain extraordinary long-term perfor-

mance. An investment of $1 in visionary company stock on
January 1, 1926 and reinvestment of all dividends, would have
grown to $6,356.- ... over fifteen times the general market.7

Consider, in that light, the argument that companies should make

their first priority the return on investment to shareholders.
Fifteen times more shareholder value is available for not putting
profits first.
A reduction in the corporate mortality rate would seem to
be advantageous for all parties: members, suppliers and contrac-
tors, the community and shareholders. If you are a manager, the
choice is yours: running a company as a maximizer of profits on
the one hand and existing for 30 or 40 years; or running the com-
pany to be professionally good at what it does and to be a good
citizen who stays in harmony with a changing world, and creat-
ing a legacy that may last decades longer and reward shareholders
more into the bargain.
11
Power: Nobody Should Have
Too Much
IT IS NOT ONLY THE SCARCITY FACTOR OF CAPITAL THAT HAS
changed in the 50 years since the end of the Second World War.
There is growing evidence that we should give more serious
thought to the way power is exercised inside a company. In the
immediate postwar years, this was never an issue.
In 1945 my home town, Rotterdam, was left with its heart
bombed out, its port systematically destroyed. The memory of
the Great Depression was still fresh in the mind. Even my gener-
ation, then only in its teens, remembered the breadlines of the
1930s. We remembered the collapse of once-great firms, which
left their former employees and labourers behind with damaged
self-respect and little hope for the future.
After the euphoria of the Liberation (which in the
Netherlands is still written with a capital L) there came, with the
Marshall Plan, a dogged mixture of realism and idealism. There
would be social equality to prevent the misery of the crisis years
from reemerging. The rebuilding of the town, the port and the
factories would create wealth for a ‘new beginning’ (as we called
it) after the depths of the war.
 POWER: NOBODY SHOULD HAVE TOO MUCH 221

The times created a great feeling of togetherness.

Everybody pulled their weight for the general good. If the coun-
try did well, if Rotterdam did well and if our companies did well,
we all knew we’d be doing fine. And it worked.
By the early 1950s, there was an atmosphere of enormous
hope. Everything could be done — no, it would be done. It was
great to play on a winning team. At the same time, there was not
much wealth. Life was simple and spartan by today’s standards.
We did not question the necessity of creating material wealth.
Anybody could see the need for it. There were constant
reminders in the lingering memory of the Great Depression, the
war damage and displaced persons everywhere. Material wealth
was necessary to repair the destruction and prevent it from ever
happening again.
Nor did we question what, in retrospect, seems like a para-
doxical premise: that the most effective way to produce this
wealth was to join together in large institutions. We knew that we
could create wealth only on a large scale. To be sure, we had all
seen how institutions could become vicious. Nobody was naive
about the potential dangers in political parties, states or armies.
At the level of the nation-state, a number of safeguards were built
into the political system that still make democracy feasible today
in Western Europe and Japan.
The most basic safeguard against having a political system
taken over by dictatorial power was a mandate that no existing
leaders could override the voters’ ability to replace them. As Karl
Popper has argued convincingly, ‘The essential element of
democracy is not necessarily the right to vote the leaders in, but
the capacity to get rid of the present leaders without a crisis.’
Yet we did not always have that capability at the level below
the nation-state. To be sure, we had many institutions with this
safeguard: trade unions, political parties, local governments, clubs
 EVOLUTION 222

and school committees. But in companies, where the employees

could not rid themselves of their managers, we did not bother
with the presence of democracy.
It might be that we relied on another democratic safeguard:
if we did not like it, we could quit. But my generation did not
exploit that basic freedom. When we joined a company, we joined
with the premise that we were in it for life. The safeguard of the
ability to quit did not seem to matter; it was overwhelmed by our
need to produce large amounts of material wealth quickly. We
were, in short, still oriented to living in a crisis.
The postwar generation understood that the individual
would be ineffective in the face of the massive task ahead. We
knew that we could repair the damage done by the war and raise
the standard of living only by joining together. United, we would
stand. Divided, we would fall. The whole would be more pro-
ductive than the sum of the parts.
Thus, in those postwar days in the Netherlands, most
young members of the workforce set out to join large companies
or governmental organizations. Shell, Unilever and Philips were
the favourite choices. They and many other companies like them
had existed and survived from the time of our fathers. Surely, they
would still exist at the time of our children.
Perhaps there was a risk of abuse of power in these compa-
nies all through the 1950s and 1960s. But it would take until 1968
for the weaknesses of most companies’ dominant organizational
principles, the abuses of their decision-making processes and the
resulting underutilization of their human resources to be aired in
public debate. In the meantime, we were content to subordinate
ourselves to the overall well-being of the companies which would
raise our standard of living.
We also accepted, unquestioningly, the organizational prin-
ciples which years of war had shown to be effective: top-level
 POWER: NOBODY SHOULD HAVE TOO MUCH 223

decision making in strongly disciplined hierarchies. Information

flowed upwards and commands downwards. Planning was cen-
tralized. The armies of the Allied nations provided a matchless
example for the way business should organize itself. These armies
had, in the end, prevailed, hadn’t they? Now, the faster growth in
material well-being which would result from becoming a member
of a large corporation seemed well worth the price of submitting
to strong central leadership vested in relatively few people. We
cared more about the quantity and distribution of what we would
produce than about the manner in which it would be produced.

An ethic of distributed power

As organized institutions, corporations are older than political
parties and trade unions; but they are far younger than the great
institutional forces of civilization: the family, the tribe, the king-
dom, religion, science and soldiering.
It’s no wonder, therefore, that many of the organizational
principles applied in companies smack strongly of copy-catting
from their older siblings. The military has been a particularly
strong source of inspiration. You can hear it in the continual calls
for ‘strategy’. Strategy, in the military, is the act of guiding and
steering a military army towards victory. Other phrases, such as
‘top-level decision making’ and ‘centralized planning’, also have a
romantic attraction that stems, perhaps, from their military her-
itage. Managers can see themselves as the Marlboro cowboy, rid-
ing the corporation like a horse and steering it into the sunset.
But should a CEO try to guide and steer a company as if it
were at war? Should a boardroom be compared with Napoleon’s
tent at Austerlitz? Like all crises, war offers little time for delib-
eration among the individuals who make up the organization.
The war situation demands centralized decision making; it should
 EVOLUTION 224

take very little time for commands to travel from issue to execu-
tion. Wartime also separates decision makers from executors.
Information travels upwards only and is not shared with anyone
outside the command line.
How appropriate are these characteristics as descriptions of a
business? They may apply to companies in a crisis state, but that
hardly represents a company at a well-developed stage of evolu-
tion. Although we sometimes describe business competition as a
fight, it is far from the crisis situation of a physical war.
More importantly, the centralization of power is inappro-
priate for the operation of a living company. It reduces the learn-
ing capacity of an organization. The alternative is to develop an
ethic of distributed power.
We have seen some examples of distributed power in other
sections of this book. Mitsui, as described in Chapter 6, main-
tained its cohesiveness despite being split into unrelated, smaller
organizations after the Second World War. Even when it came
back together, the smaller organizations retained authority over
their own decisions. They continued to see themselves, however,
as part of the larger Mitsui group, and they made their decisions
with the goal of contributing to its vitality.
The issue of power distribution finds an interesting illustra-
tion in the Royal Dutch/Shell Group. As noted in Chapter 9, the
group itself resulted from a cross-cultural merger — the 1907 merg-
er between Royal Dutch, a company incorporated in the
Netherlands (owning 60 per cent of the group’s shares), and Shell
Transport & Trading, incorporated in the UK (owning 40 per cent).
There are thus two parent companies, one Dutch and one
British; they continue to exist today, independent of each other
yet interlinked, domiciled in two countries, each with a different
legal system. The Dutch system is distinctly continental; it was
reshaped during the French occupation in Napoleonic times. The
 POWER: NOBODY SHOULD HAVE TOO MUCH 225

British jurisprudence system, accumulated since Saxon days, has

never had the benefit of a good spring clean. Consequently, in the
corporate law of these two countries there are some important
differences in the composition and role of corporate boards and
their members. The two parent companies must get on well with
each other, because there is no court of justice to which they can
refer disputes.
The full boards of both companies meet voluntarily once a
month, and they call their meetings the ‘conference’. No legally
valid decisions which are binding on both companies can be taken
during these meetings. The officers of both companies must hold
separate meetings during which they legalize, for each parent com-
pany, the decisions they have arrived at jointly.
From the top of the Shell Group down there is no tradi-
tional mechanism to resolve conflict. The group has no CEO. The
chairman of the managing directors is only primus inter pares,
first among peers. One way or another, the members of the
Committee of Managing Directors (CMD) and the two boards of
directors have to agree among themselves on solutions which are
acceptable to all. In practice, there will not always be real una-
nimity, but there is no good way to force through a decision to
which one or more of the members is actively opposed. The min-
imum required is a quasi-unanimity; otherwise the decision has to
be referred back to the next lower level. Quasi-unanimity does
not mean that everybody agrees with the proposal. It means that
no one is so violently opposed that he or she will show a veto
card. The chairman has no other power than his persuasion; he
has no casting vote or final decision.
In theory, decisions could be forced through shareholders’
votes. After all, Royal Dutch owns 60 per cent of the group. To
the best of my knowledge, however, there is no record of any
decision taken by vote at the board or CMD level. This testifies
 EVOLUTION 226

to the wisdom of many generations of directors. To force deci-

sions by majority vote in such a delicate structure, without any
legal recourse, would have broken up the marriage long ago. But
it withstood the competitive pressures of the 1920s. It withstood
the Second World War, when Royal Dutch all but disappeared as
a working entity during the German occupation of the
Netherlands. It even withstood government pressures during the
oil boycott in 1973, when the Netherlands was completely cut off
by the Arab suppliers and the UK suffered diminished supplies.
Had the group’s board voted to favour either the UK or the
Netherlands with its supply of embargoed oil, the Dutch domi-
nance of voting power could easily have created serious cracks in
the international group unity.
As it happens, most conflicts and tensions do not come
directly from the outside world and enter at board level. Conflicts
mostly percolate upwards through the system. The coordinators,
one level below the CMD, learn very quickly that it is not propi-
tious for their careers to channel potential conflicts at their level
to the CMD. It is far better to reach an amicable settlement with
their colleagues themselves. In their turn, coordinators are not
favourably disposed towards subordinates who do not solve their
own conflicts. And so on further down the line.
Consequently, at many levels down the hierarchy, a lot of
people are required to participate in most decisions — and certain-
ly the decisions which will require change. Each person has a
quasi-veto power. It is not easy in Shell to exclude people from a
decision if they are to be involved in implementation. After the
Second World War, this quality was further reinforced by the
introduction of a matrix organization. As the popular definition
puts it, a matrix is ‘an organization in which nobody can take any
decision on his or her own, but anybody on his own can stop a
decision being taken’.
 POWER: NOBODY SHOULD HAVE TOO MUCH 227

For all of these reasons, ever since the signing of the merg-
er agreement 90 years ago, the downward pressure on conflict
resolution has been a fixture of Shell life. Neither internally nor
externally is there a desk at which the buck stops. The buck stops
at thousands of desks, each at an appropriate level. The
Economist, in a lead article, once expressed this idea as a generic
principle: ‘The important thing about power is that nobody
should have too much of it.’

The implications of distributed power

The arguments against distributed power are well known. They
represent well-nigh irresistible temptations to many managers.

✦ It takes forever to take a decision. We do not have that sort of

time. The world and the competition are moving forward; we can-
not stay behind. A wide distribution of power can be incredibly
frustrating, but it means that the number of minds which are
actively engaged in the decision-making process is considerably
increased. There is no convincing evidence that it leads to slower
action (although it certainly takes longer to come to conclusions).
It may well lead to better action and it may enhance the organi-
zation’s ability to learn. In fact, there is much circumstantial evi-
dence that institutions are more successful, survive much longer
and thrive better if they have effectively given power to minori-
ties to veto or delay majority decisions which go against their
interests or better judgement.

✦ A cohesive decision requires an ultimate seat of power so that

the result has unity. This objection is based on a fallacy. Many
managers assume that the crucial point of a decision is the
moment when the chief executive has a brilliant thought — or has
 EVOLUTION 228

been convinced by somebody else’s brilliant thought.

But that moment of conclusion is only incidental to the real
impact of the decision — the way it is implemented. In the world
of business, only action counts. In other words, implementation
is an integral part of the decision and not something separate
which happens afterwards. When a decision is made, leaving out
the people whose cooperation is necessary may speed up the
moment of arriving at a new conclusion, but it will lose any time
so gained through a slow and almost certainly unintelligent
implementation.

✦ You cannot oblige people to produce results except through

authority. In the end, the lure of power — the need to be wanted
and the need to feel in control — deters many senior executives
from approving the distribution of power, thus diminishing the
institutional learning capacity.
Top managers will typically phrase their qualms in terms
like, ‘I want to sleep well at night.’ This is only possible, they
seem to feel, if they have some sort of certainty: nothing unex-
pected and unwelcome can happen in the company unless they
can know about it and prevent it first. Or they may feel that the
need to maintain cohesion in the company requires an open path
(a controlling path) from the top. How else can the chief execu-
tive take responsibility for the company’s ultimate results?
The usual resolution of this dilemma between control and
freedom is, once again, to err on the side of caution — to create a
wide-open path of delegation all the way to and from the top of
the hierarchy. This tendency plays into the hands of the employ-
ees who, at any given moment, do not feel capable of dealing with
a business situation. They send their problems and conflicts to the
next higher level of authority. This phenomenon is the obscure
shadow side of the more famous problem of delegation of author-
 POWER: NOBODY SHOULD HAVE TOO MUCH 229

ity. Game playing, incompetence or laziness at the lower levels

meet the bosses’ desire to control, pride at being asked, or the
illusion that only senior managers know the answer.
If both the lower and the higher levels in the hierarchy give
in to this temptation, the end result is that fewer people with less
factual knowledge participate in decision making. Once again,
corporate learning suffers.

Upward impediments
As I noted earlier, relinquishing control can be very frightening.
It goes against managers’ personalities, their training and the
incentives that they have received all their lives. Now, as I sug-
gested in Chapter 8, they must balance control and tolerance in a
different way. Rather than steering, they must set a context. ‘Go
ahead,’ they must say to their subordinates. ‘You are free to fail
within the context of the learning that we have done together. I
will watch.’
Even when managers accept the idea that power must be
distributed, many managers have little experience of the reality.
They don’t know how to ensure that distribution of power will
actually take place.
There is a widespread illusion, for instance, that one can
distribute power in a company, and engage more brains, simply
by delegating authority. It does not matter whether you just tell
your subordinates to ‘get on with it’ or give them statutory pow-
ers to spend money without reference. If you give them legal
power to take decisions, they will take the decisions — or so peo-
ple believe.
Some employees will be happy to take decisions. Some are
eager to take decisions which are not theirs to take. But a top
manager cannot count on decisions being taken consistently at
 EVOLUTION 230

the lowest level, where they should be taken, unless impediments

are created to upward delegation of difficulties and conflicts.
In other words, make it difficult to move conflict up the
hierarchy. Set in motion policies, implicitly or explicitly stating
that people can ask the next higher levels for advice but cannot
ask them to make decisions. This policy is counterintuitive to
many top managers. It flies in the face of their misguided belief
that they are a paterfamilias — put in place to solve the difficulties
of the family and smooth the path for the children.
I learned about designing upward impediments as an oper-
ating company manager in Brazil. In Latin America, the prevail-
ing model of corporate culture is the family. The father is, indeed,
the ultimate power and wisdom of the family. Therefore, the
weaker members of the family are conditioned to go and cry at
‘pater’s’ knees. Things are rarely done without the superior’s
agreement. And because Brazilians love to play that role, it is very
difficult to delegate authority. The senior managers of the compa-
ny found themselves bedeviled with trivia, but unwilling to give
up the burden.
We tried unsuccessfully for a long time to deal with this
problem. Finally, we set arbitrary levels of decision-making scale.
If a decision exceeded a certain number of cruzeiros worth of oil
sales, or a certain level of personnel management, we allowed it to
be delegated upwards to the central office. We chose the levels
based on a ratio of roughly 90 per cent to 10 per cent. In other
words, only one-tenth of the problems should be kicked up to the
central office. Then we made sure that the central office did not
have the resources to deal with more problems than those, so that
anyone who sent up more than one-tenth of their problems would
have to wait an inordinately long time for results. We also made
sure that the local offices did have the resources to deal with their
90 per cent of the problems. We redistributed people back from
 POWER: NOBODY SHOULD HAVE TOO MUCH 231

the central office in Rio de Janeiro to the field offices around the
rest of Brazil, making sure that they were given better titles and
salaries in the process. This was counterintuitive; in the past, pro-
motions had nearly always meant a move to the central head-
quarters. We supplemented these promotions with in-depth
training so that local managers would gain the same experience, or
more, with such important skills as cash flow management.
There were immediate consequences. Within one year, Shell
Brazil’s profit jumped 60 per cent and stayed at the higher level.
But the long-term consequences were more significant still. When
the country fell into economic difficulties, the working capital sit-
uation in the company began to totter. Other oil companies found
themselves stultified and torpedoed by changing credit terms for
their supplies. But Shell Brazil adapted with enormous flexibility
and speed. Back at the central office we needed only a small team;
people at local offices had all the power they needed to deal effec-
tively with the situation.

The living company needs new governance

In Chapter 10, I argued that shareholders should give up their
ultimate power over the life and death of companies, because it
optimizes the wrong production factor — capital — to the detri-
ment of the shareholders themselves. In this chapter, I have exam-
ined incidental but growing evidence that management should
similarly relinquish some of its own power. If management gives
in to the irresistible temptation to concentrate that power at the
top, too few brains are engaged in institutional learning.
Where then should power be seated in the knowledge-creat-
ing company of tomorrow, and to whom should it be distributed?
When phrased in this way, the question of corporate gover-
nance becomes reminiscent of the debate that took place in the
 EVOLUTION 232

Western world at the time of the French Revolution and the sub-
sequent conceptualization of the US Constitution. It was the era
of the decline of absolute monarchy and the evolution of the
democratic form of government. I believe that we would find a
great deal of inspiration for the forms that corporate governance
should take by returning to the debate that took place when we
developed the governance structure of our Western nation-states.
The developers of the US constitution, for instance, are well
known for setting out to devise a system that would provide con-
tinuity without the need for the absolute powers of a monarchy. At
the same time, in the emerging nation-states of Europe, people had
become painfully aware of the dangers for society as a whole of
power which is too highly concentrated. The results were declara-
tions of basic principles and rights and constructs such as the Trias
Politica (the separation of legislative, executive and judicial power
and its distribution over three independent power centres, with
checks and balances to prevent one from overtaking the others).
Ever since, these three powers have been tugging and pulling in an
ever-shifting balance.
Various nations may differ in the exact nature of their dis-
tribution of rights, principles and power. For example, the US
concentrates more power in the hands of its chief executive than
would be possible in the Scandinavian countries. And no one is
ever completely happy with the distribution of power in their
own particular democracy — that is, as Winston Churchill point-
ed out, until they consider the alternatives.1 But the basic concept
of continuity without absolute power has become a common
touchstone for all developed nations. In a system of distributed
power, there are safeguards to ensure that no one interest will pre-
vail. In a system of governance based on checks and balances,
there are ways to get rid of a bad leader without plunging the
community into a crisis.
 POWER: NOBODY SHOULD HAVE TOO MUCH 233

We need a similar touchstone for corporations. We need a

system of corporate governance that provides continuity, with all
the requirements that nurture a living company and a human
community, without absolute power concentrated in the hands of
either shareholders or management.
To develop this system of corporate governance in our age
of knowledge, we need to open the debate about power and gov-
ernance. As matters stand today, companies may too easily suffer
the consequences of ultimate power given to one basic interest
group, the shareholders, whereas the governance structure gives
ample opportunity to an almost medieval exercise of absolute
power by management.
It is no wonder that, under these conditions, companies can
become fiefdoms for the few, to be exploited like a machine.
Under these conditions, no living company will be able to thrive,
if its success depends on freedom, space and mutual trust between
members.
And the role of commercial institutions is an important one:
to provide humanity with the material goods necessary for a
decent living. More than ever, success in this undertaking is
dependent on the extent to which these companies will be able to
create knowledge, not in the head of the individual, but know-
ledge on which the company as a whole can act. This is blinding-
ly clear in the brain-rich, asset-poor institutions which have
shown such spectacular growth over the last 20 to 30 years: the
law firms, auditor partnerships, software companies and organi-
zations like VISA. But even the old type of asset-rich company,
such as oil and steel firms, now need much more knowledge
embedded in their actions than was the case some 20 years ago.
The Baron de Montesquieu, in writing about the principle
of the Trias Politica, pointed out that separated and distributed
power meant ‘freedom’. The inverse — concentrated power in one
 EVOLUTION 234

hand — meant, he wrote, that ‘All was lost.’

Concentrated power means no freedom. No freedom
means little knowledge creation and, worse, little knowledge
propagation. No propagation means little institutional learning
and, thus, no effective action if the world changes. One of the
main driving forces of a company is the development of its poten-
tial. Can we create a form of governance which maximizes the
potential of our membership and thereby reduce the corporate
mortality rate? Or is all of that potential lost?
Epilogue:
The Company of the Future
IN THE GLOBAL VILLAGE THAT IS EMERGING TODAY, ECONOMIC
companies risk being economic losers. A world of shrinking
political barriers will need more and more living companies.
This is because, paradoxically, the horizons of individual
businesses are growing while political horizons shrink. There is a
great temptation for companies to move out of their regional or
national niche into a wider and therefore more unfamiliar envi-
ronment. Even those that resist the temptation run the risk that
the outside world will invade their home turf.
Over time, fewer and fewer companies will live and work in
an environment over which they have much control. More and
more companies will be growing potatoes in the Andes (to use
my metaphor from Chapter 8) rather than in a glasshouse. With
their habitat shrinking, economic companies might become an
endangered species — pushed back into isolated niches and legally
protected national parks.
What then does a healthy company of the future look like?
How do we recognize when we are on the right track towards a
living company? And if a company does not look very healthy,
what could the concerned manager do to restore it?
Everything that this type of company does is rooted in the
two main hypotheses of this book:
 THE LIVING COMPANY 236

✦ The company is a living being.

✦ The decisions for action made by this living being result from
a learning process.

I have used the four key factors of the Shell study of long-lived
companies — learning, persona, ecology and evolution — as my
own way of defining what happens in living companies day to
day. But managers in living companies around the world will be
inventing for themselves more and more of the practice of man-
aging a living company. Over the next several years, it will
become increasingly apparent when companies are ‘living’. They
will have characteristics such as the following.
A living company will have members, both people and
institutions, who subscribe to a set of common values and who
believe that the goals of the company allow them and help them
to achieve their own individual goals. Both the company and its
members have basic driving forces: they want to survive and, once
the conditions for survival exist, they want to attain and expand
their potential. The underlying contract between the company
and its members (both individuals and other institutions) is that
the members will be helped to reach their potential. It is under-
stood that this, at the same time, is in the company’s self-interest.
The self-interest of the company stems from its understanding
that the members’ potential helps create the corporate potential.
The nature of this underlying contract creates trust, which
results in levels of productivity that cannot be emulated by disci-
pline and hierarchical control. Trust also allows space and toler-
ance, both inside the hierarchy and towards the outside world.
These are basic conditions for the high levels of organizational
learning that will, on occasion, be very necessary.
The company has a will; thus, it makes choices. As a result
of making choices, it may diverge from the conditions or the val-
 EPILOGUE: THE COMPANY OF THE FUTURE 237

ues in its environment. Continued disharmony with its world will

lead to a crisis, possibly a fatal one.
To avoid a crisis and to gain a better perception of a diverg-
ing environment, the company must be open to the outside
world. It develops memories of the future to guide its decision
making. This means that there is tolerance for the entry of new
individuals and ideas. However, members know who is ‘us’ and
who is ‘not us’. The organization builds up its own corporate
immune system to protect itself against parasites.
The organization is a community whose membership is
variable, not only through individuals entering and leaving over
time, but also by enlargement or reduction of the whole over
time. This community has an identity — a persona — which con-
tinues to exist even when its composition changes.
Sometimes members are forced out or converted into sup-
pliers or contractors (a money relationship) — as when their value
system is not harmonious with that of the company. This shift is
healthy, because harmonized value systems are a basic require-
ment for corporate cohesion. Sometimes reduction in membership
takes place, because an inner group of members redefines ‘us’ and
enlarges the definition of ‘not us’. This may not be healthy,
because of the shock to the trust levels of the remaining members.
The human members of a healthy company are mobile,
both in the different jobs they perform during their careers and in
the places where they perform those jobs. They network, they
meet and they communicate across the whole organization. There
is mutual trust that people will act fairly and that the leaders are
as honest as one could expect from human beings. People know
their trade. Power is distributed; there are checks and balances in
the power system and the present leaders understand that they are
just one generation out of many still to come.
In addition to members, the company will have physical
 THE LIVING COMPANY 238

(capital) assets which it uses for one or more economic activities to

earn a living. Once survival is assured, the economic activity is used
as the basis from which the community develops its potential. The
company’s evolution is thus governed by financial conservatism.
While it is engaged in a particular activity at a particular
place, the work community is surrounded by suppliers (of materi-
als, capital, labour and intelligence), by customers, by the regional
or national community and by other stakeholders who are all part
of its world and with whom it must maintain a state of harmony.
If survival is at stake, the community will divest itself of
assets and try to change the content or nature of its economic
activity before it divests itself of people.
In setting out all of these characteristics, I have tried to por-
tray not just the economic aspects of a living company but the
psychological, sociological and anthropological aspects. All of
these aspects complement, rather than fight, each other.
And in portraying them all, I hope that I have made it clear
that the priorities of the management of a living company cannot
be exclusively expressed in economic terms.
If corporate health falters, the priority should be to mobi-
lize human potential, to restore or maintain trust and civic behav-
iour and to increase professionalism and good citizenship.
Everything starts from that basis. If companies can meet
those conditions, I believe that average corporate life expectancy
will begin to rise, and that all humanity will benefit as a result.
This is not to say that companies should live forever. In the
corporate species, however, the gap between average and maxi-
mum life expectancy is still so wide that we could conclude that
too many companies suffer an untimely death. A reduction in the
corporate mortality rate would seem to be advantageous for all
parties: members, suppliers and contractors, the community and
shareholders.
Notes and References

PROLOGUE: THE LIFESPAN OF A COMPANY

1 Royal Dutch/Shell Group Planning PL/1, Corporate Change: A Look
at How Long-Established Companies Change, September 1983. This
private study is not available to the public; I have quoted extensively
from it in this book, however. The facts and figures on the first two
pages of this chapter are all taken from this study. In all, 30 companies
were studies, for 27 of which case histories were prepared. The compa-
nies were Anglo American Corporation, Booker McConnell, British
American Tobacco, Daimaru, DuPont, East India Companies, Anthony
Gibbs, W.R. Grace, Hudson’s Bay Company, IBM, Kennecott, Kodak,
Kounike, 3M, Mitsubishi, Mitsui, Pilkington, Rolls-Royce, Rubber
Culture, SKF, Siemens, Société Générale, Suez Canal Company,
Sumitomo, Suzuki, Unilever and Vestey.
2 Ellen de Rooij, A brief desk research study into the average life
expectancy of companies in a number of countries, Stratix Consulting
Group, Amsterdam, August 1996.
3 Corporate Change, Appendix V, p. 25.
4 James C. Collins and Jerry I. Porras, Built to Last: Successful Habits of
Visionary Companies, New York: HarperCollins, 1994, p. 9.

CHAPTER 1: THE SHIFT FROM CAPITALISM TO A

KNOWLEDGE SOCIETY
1 See, for example, Fernand Braudel, The Wheels of Commerce, vol. 2 of
Civilization and Capitalism, 15th-18th Century, trans. Sian Reynolds,
Berkeley: University of California Press, 1992, p. 466ff; and Henri
Pirenne, Les périodes de l’histoire social du capitalisme, 1922, Brussels.
2 The Wheels of Commerce, p. 52.
 THE LIVING COMPANY 240

3 The extremes of this world are recounted by Robert L. Heilbroner in

The Worldly Philosophers: The Lives, Times, and Ideas of the Great
Economic Thinkers, New York: Simon & Schuster, 1953, 1986. See
Chapter 8, ‘The World of Thorstein Veblen’.
4 See, for example, Peter Drucker, The New Realities, New York: Harper
& Row, 1989, p. 178ff; or Ikujiro Nonaka and Hirotaka Takeuchi, The
Knowledge-Creating Company, Oxford: Oxford University Press, 1995.
5 I have taken this definition from one of my old handbooks, by Professor
Dr Wilhelm Röpke, Die Lehre von der Wirtschaft, Zurich: Eugen
Rentsch Verlag, 1946; but different versions of the same definition
appear in countless handbooks on economics in many languages.
6 Jean Piaget, The Psychology of Intelligence, London: Routledge &
Kegan Paul, 1986, pp. 8-9 and 103.

CHAPTER 2: THE MEMORY OF THE FUTURE

1 See, for example, Milton Moskoqitz, Michael Katz and Robert
Levering, Everybody’s Business: An Almanac, New York: Harper &
Row, 1980, pp. 603-10.
2 Corporate Change, p. 6.
3 Corporate Change, p. 9.
4 Sven Rydberg, The Great Copper Mountain: The Stora Story,
Hedemora: Gidlunds, 1988, p. 50. This book was published on the occa-
sion of the 700-year anniversary of the enterprise.
5 David Ingvar, ‘Memory of the Future: An Essay on the Temporal
Organization of Conscious Awareness’, Human Neurobiology, 1985(4):
127-36.

CHAPTER 3: TOOLS FOR FORESIGHT

1 This story was adapted from one often told by Pierre Wack, who talked
of the mayor of Dresden. I prefer to use Rotterdam, the city of my birth.
2 Independent, 24 October 1992.
3 Daniel Yergin, The Prize: The Epic Quest for Oil, Money and Power,
New York: Simon & Schuster, 1991.
 NOTES AND REFERENCES 241

4 Peter Schwartz, The Art of the Long View: Planning for the Future in an
Uncertain World, New York: Doubleday/Currency, 1991, pp. 72-90.
5 Art Kleiner, The Age of Heretics, Heroes, Outlaws, and the Forerunners
of Corporate Change, London: Nicholas Brealey Publishing, 1996, pp.
162-3.
6 Joseph Campbell, The Hero with a Thousand Faces, Princeton, NJ:
Princeton University Press, 1979.
7 Peter Schwartz, The Art of the Long View: Planning for the Future in an
Uncertain World, New York: Doubleday/Currency, 1991.
8 Kees van der Heijden, Scenarios: The Art of Strategic Conversation,
New York: John Wiley, 1996.
9 Art Kleiner, The Age of Heretics, Heroes, Outlaws, and the Forerunners
of Corporate Change, London: Nicholas Brealey Publishing, 1996.
10 Pierre Wack, ‘Scenarios: Uncharted Waters Ahead’, Harvard Business
Review, Sept-Oct 1985: 72-89. Reprinted in Scenarios: The Gentle Art
of Reperceiving, ‘Strategic Planning in Shell Series No. 1’, Shell
International Petroleum Company Limited, Group Planning, London,
February 1986.

CHAPTER 4: DECISION MAKING AS A LEARNING

ACTIVITY
1 John Holt, How Children Fail and How Children Learn, London:
Pitman, 1964 and 1967; Harmondsworth: Penguin, 1970.
2 There is a variety of sources available on the cycle of learning, all with
somewhat different terminology. The cycle described here is based on
Jean Piaget’s model of learning and cognitive development. Piaget
labelled his phases ‘Active Egocentricism (Acting), Concrete
Phenomenalism (Perceiving), Internalized Reflection (Embedding), and
Abstract Constructionism (Concluding)’. See Jean Piaget, Genetic
Epistemology, New York: Columbia University Press, 1970. The most authoritative
source on comparative learning cycle theory is David Kolb, Experiential Learning
(Experience as the Source of Learning and Development), Englewood Cliffs, NJ:
Prentice-Hall, 1984. Kolb synthesized and expanded on theoretical work by Piaget,
American educational philosopher John Dewey, organizational
 THE LIVING COMPANY 242

psychology pioneer Kurt Lewin and others. British management writer

Charles Handy adapted the idea of a ‘learning wheel’ for business read-
ers in his book The Age of Unreason, London: Century Hutchinson,
1989. It was given practical, day-to-day application by Richard Ross,
Bryan Smith and Charlotte Roberts in ‘The Wheel of Learning’, in Peter
Senge, Art Kleiner, Richard Ross, Charlotte Roberts and Bryan Smith,
The Fifth Discipline Fieldbook, London: Nicholas Brealey Publishing,
1994, p. 59.
3 Jean Piaget, The Psychology of Intelligence, London: Routledge &
Kegan Paul, 1986.
4 This heading is influenced by Donald N. Michael, Learning to Plan, and
Planning to Learn, San Francisco: Jossey-Bass, 1974, 1996. This book,
ahead of its time, helped create the understanding that learning has an
important place in the life of a company.
5 D.W. Winnicott, Playing and Reality, London: Tavistock, 1971;
Harmondsworth: Penguin Education, 1980; John Holt, How Children
Learn, New York: Dell, 1967; Seymour Papert, Mindstorms: Children,
Computers, and Powerful Ideas, New York: Basic Books, 1980.
6 Schwartz tells this story in The Art of the Long View, p. 91.
7 Stella and iThink are trademarks, © 1990 High Performance Systems,
Inc., Hanover, NH.
8 Some of the first microworlds which led to successful group learning
experiments (representing a biotechnology startup firm, the Shell petrol
retailing business in the Netherlands and the natural gas business after
the oil-price collapse of 1986) were developed with the help of John
Morecroft of London Business School; David Lane, then of Shell
International and now at the London School of Economics; David
Kreutzer of GKA, Inc. (formerly Gould-Kreutzer Associates); and
Jenny Kemeny of Innovation Associates. John Morecroft also wrote a groundbreaking
article on the role of computer models as maps and microworlds for
experimentation and learning. A recent version of that article has appeared under
the title ‘Executive Knowledge, Models and Learning’ in John Morecroft and John
Sterman, Modeling for Learning Organizations, Portland, OR: Productivity Press,
1994.
 NOTES AND REFERENCES 243

9 This technique has been developed by Tony Hodgson of IDON in

Scotland.
10 Useful results were also achieved with systems developed by Peter
Checkland, Jonathan Rosenhead and Colin Eden.
11 Peter Senge, Art Kleiner, Charlotte Roberts, Richard Ross and Bryan
Smith, The Fifth Discipline Fieldbook, New York: Doubleday/
Currency, 1994.

CHAPTER 5: ONLY LIVING BEINGS LEARN

1 Since 1974 this attitude towards large corporations has changed
somewhat. See ‘Everybody’s Favourite Monster’, Economist, March
1993.
2 R.B. McLeod, ‘Obituary for William Stern’, Psychological Review,
45(5), September 1938.
3 William Stern, Person und Sache, Zweiter Band: Die menschliche
Persönlichkeit, 2nd edn, Leipzig: Verlag von Johann Ambrosius Barth,
1919, pp. 6, 9 and 40ff.
4 Francisco Varela and Antonio Continho, ‘Somebody Thinks — The
Body Thinks: Why and How the Immune System Is Cognitive’, J.
Brockman (ed.) The Reality Club, Vol. 2, New York: Phoenix Press,
1988.
5 Simon Schama, The Embarrassment of Riches, Berkeley: University of
California Press, 1988, p. 334.
6 Ibid., p. 345.
7 Stern, Person und Sache, ‘III. Die Aufnahme der Fremdzwecke in den
Selbstzweck (Introzeption)’, p. 55ff.

CHAPTER 6: MANAGING FOR PROFIT OR LONGEVITY:

IS THERE A CHOICE?
1 Corporate Change, p. 10. The remark is attributed to Lord Cole, chair-
man of Unilever.
2 Stern, Person und Sache, ‘II. Das System der Fremdzwecke
(Heterotelie)’, p. 49.
 THE LIVING COMPANY 244

3 Robert Putnam, Making Democracy Work: Civic Traditions in Modern

Italy, Princeton, NJ: Princeton University Press, 1993, p. 165.
4 Joe Jaworski, Synchronicity: The Inner Path of Leadership, San
Francisco: Berrett-Koehler, 1996, p. 131ff.

CHAPTER 7: FLOCKING
1 Jeff S. Wyles, Joseph G. Kimbel and Allan C. Wilson, ‘Birds, Behavior
and Anatomical Evolution’, Proceedings of the National Academy of
Sciences, July 1993.
2 From an interview with Bram Roza, head of Group Training, Royal
Dutch/Shell, in the Dutch language magazine Shell Venster,
January/February 1994.

CHAPTER 8: THE TOLERANT COMPANY

1 Corporate Change, p. 12.
2 Milton Moskowitz, Robert Levering and Michael Katz, Everybody’s
Business: A Field Guide to the 400 Leading Companies in America, New
York: Doubleday/Currency, 1990, p. 529.
3 Corporate Change, p. 9.
4 See Charles Hampden-Turner, Charting the Corporate Mind, New
York: Free Press, 1990, for a fuller development of this theme.
5 Henry Mintzberg, The Rise and Fall of Strategic Planning, New York:
Free Press, 1994, pp. 98-99 and 119-21.
6 Mintzberg, The Rise and Fall of Strategic Planning, pp. 98-99, and The
Nature of Managerial Work, New York: Harper & Row, 1973.
7 The poet is Antonio Machado.
8 Alison Smith, ‘Empty Room at the Top — Leeds Permanent’s Long
Quest for a Chief Executive’, Financial Times, 5 August 1994, p. 9.
9 Ibid.
10 Ibid.
11 Rachel Bodle, ‘Everyone a Rainmaker’, Insight, 8(1), Jan-Mar 1994, p.
23.
 NOTES AND REFERENCES 245

CHAPTER 9: THE CORPORATE IMMUNE SYSTEM

1 Francisco Varela and Antonio Continho, ‘Somebody Thinks — The
Body Thinks: Why and How the Immune System Is Cognitive’, J.
Brockman (ed.) The Reality Club, Vol. 2, New York: Phoenix Press,
1988.
2 Michael Porter, ‘From Competitive Advantage to Corporate Strategy’,
Harvard Business Review, May-Jun 1987, pp. 43-59; Economisch-
Statistische Berichten, 11 March 1988.
3 Richard Dawkins, ‘Universal Parasitism and the Co-evolution of
Extended Phenotypes’, Whole Earth Review, Spring 1989, p. 90.
4 In the case of British chemical company ICI, the demerger of the life sci-
ence interests to form a new and independent company called Zeneca
may well have been partly prompted by a feeling within the pharma-
ceutical division that it was contributing more to ICI than it was gain-
ing from its parent.

CHAPTER 10: CONSERVATISM IN FINANCING

1 Richard Onians, ‘Making Small Fortunes: Success Factors in Starting a
Business’, talk at the Royal Society of Arts in London, 11 January 1995;
published in RSA Journal, CXLIII(5459), May 1995, p. 22.
2 Ibid., p. 25.
3 Ibid., p. 26.
4 Collins and Porras, Built to Last, p. 189.
5 Ibid., p. 8.
6 J.B. Priestley, English Journey, London: Mandarin, 1994, p. 345.
7 Collins and Porras, Built to Last, p. 4.

CHAPTER 11: POWER: NOBODY SHOULD HAVE TOO

MUCH
1 Winston Churchill, speech in the House of Commons, 11 November
1947.
Index
3M 176 cohesion 13, 16, 17, 121-2, 127, 128-9,
Ackoff, Russell 15 132, 135, 146, 174, 182, 189, 224,
acquisitions and mergers 194-7 228, 237
adaptation 16, 32, 44, 76, 161, 172, 210 Colgate 15
Collins, James and Jerry Porras 15-16,
BASF 137 21, 138-9, 206, 208, 219
Beck, Peter 58 Collyns, Napier 58, 93, 254
Benetton 151 commodities trading 147-8
Billiton 196 community, work 4, 9, 13, 16, 17, 23,
blue tit, evolution 161-2, 167 29, 123, 125, 126, 133, 137, 139,
Bodle, Rachel 188-9 146, 148, 152, 153, 156, 165, 211,
Booker McConnell 172-3, 174 216, 217, 233,
Braudel, Fernand 22, 23 238
competitive advantage 25, 67, 188
Campbell, Joseph 62 computer modelling 86-8, 89
capital, controlling access to 23-25 conflict 226
centralization 182, 223-4 conservativism in financing 14, 16, 205-
change: anticipating need for 31; 7, 238
capability to 27; effects of 30-31; context, setting 188, 229
management 16, 27, 36, 43, 92 continuity 4, 142, 146, 152, 233
Chilean potato, parable of 177-9 control 3, 168-9, 170, 180, 182, 183,
cognitive mapping 90 184, 210, 228, 229
cognitive psychology 40, 44
 INDEX 247

core competencies 175-6 5, 139, 144, 153, 156, 194, 207-8,

crisis 38-40, 222, 223-4, 237 210, 212, 216, 235
cycle of continuous learning 73, 188 economics 101
ecosystems 156
Daimaru 11-12 education 71, 80
Dawkins, Richard 197-8, 199-200 entity, corporate 99-100, 213
de Rooij, Ellen 8 environment 35, 37
death, corporate 9, 215-19, 234, 238 equifinality 145
decentralization 16, 128, 172, 173, 174, ethics 113, 114, 115
182 European companies 8, 12, 15
decision making: 68-9, 70, 71, 72, 77, evolution 3, 16, 32, 160, 167, 197, 202,
80, 83, 92, 93, 187, 222-3, 225, 227, 209-10, 212, 236
229, 230, 237; as learning process exit rules 152-3, 153-4
72, 75; by teams 166; speed of 77, exploration and production 42-4
88, 92 Exxon 19, 155, 216
Deutsche Bank 137
Deutsche Shell 128 fear 79, 167
diversification 10-11, 13, 58, 128, 172, financing, conservatism in 14, 16, 205-7,
174, 175-6 238
diversity 128, 129, 140, 174-5, 179, 181, flocking 144, 162-4, 165, 166-7, 168-9
189 Ford 15
driving forces, of organizations 61, 63, forecasting, financial 51
208, 234, 236 foresight 37, 40, 44, 49, 59, 66
DuPont 11, 12, 31-2, 174 Forrester, Jay 85-6, 91
DYNAMO 86 freedom 167-9, 182, 228, 233
future, making sense of 44, 45
ecology 16, 32, 158, 169, 171, 180, 203,
236 General Electric 143
economic company 25-7, 28, 122, 123- General Motors 14, 15, 19, 31, 32, 109
 THE LIVING COMPANY 248

genetic clock 159-60 intolerance 180-1

governance, corporate 132, 211-15, introception 111, 112-13, 117, 119, 120,
231-4 127, 132
group dynamics 90 intruders 193-4, 197, 199-200
growth: corporate 16, 17, 76, 210; iThink 86
managing for 43
Japanese companies 8, 11-12, 15
Hanover Insurance Company 90 Jaworski, Joe 150
Hawken, Paul 177
health, corporate 1, 13, 14, 152, 192-3, Kahn, Herman 57, 58, 67
200, 207, 235, 238 Kleiner, Art 60, 67, 254
Hewlett-Packard 206 knowledge 19, 22, 24, 27, 28, 72, 189,
Hoechst 137 211, 233
Holt, John 70, 80 Kodak 11
homo economicus 101, 105
Hudson Institute 57 law, corporate 211, 225
Hudson’s Bay Company 11, 12 leadership 71-2, 152, 188-9, 223
learning: 47, 75, 80, 89, 93, 111, 135,
IBM 14, 19 156, 181, 188, 191, 194, 196-7,
identity, corporate 3, 11, 13, 16, 17, 32, 236; accelerating 70, 80, 88, 93,
34, 40, 93, 128, 135, 173, 210, 237 158; among animals 158-61; by
immune system, corporate 190-94, 199, accommodation 27, 75-7, 79, 169,
237 181; by assimilation 75-6, 79, 181;
implementation of decisions 88, 227 by experience 78; corporate: 4, 16,
information overload 46, 47 17, 22, 27, 29, 30, 32, 36-7, 39, 69,
Ingvar, David 44-8, 49, 52, 76 70, 71, 72, 88, 95, 112, 152, 167,
innovation 160, 161, 162, 167, 168-9, 189, 193, 202, 203, 224, 227-9, 231,
176 236; cycle of continuous 73, 188;
 INDEX 249

intergenerational 160; organization matrix organisation 226

19, 28, 29, 77; role in evolution McCutcheon, Ian 28
159; through play 80-83, 84-5, 88, McLeod, R.B. 103
89, 92, 93, 95 McMaster, Michael 113
life expectancy, corporate 1, 7-9, 14, 17, memories of the future 45-8, 52, 53, 56,
18, 21, 29, 110, 215, 238 61, 76, 185, 237
lifetime employment 149 memory, institutional 42
living being, company as 2-5, 17, 29, 96, mental maps 40, 68, 87
122, 137 mental models 73-5, 89, 91, 92
living company: characteristics of 32, microworlds 84-5, 86, 92
235-8; implicit contract of 146-7, Mintzberg, Henry 183-4
148-9, 150, 210; motivation in 145 mission statement 133, 209
Lockheed 163, 176 Mitsubishi 11
LOGO 84, 91 Mitsui 11-12, 134-7, 174, 224
longevity, corporate: 12, 21, 126, 173, Mobil Oil 100
202 mobility 161, 162, 163, 165, 166, 237
longlived companies:11, 31, 77, 126, money: as measure of success 207-9;
133, 172, 176, 206, 208; borrowing 204-5, 206; role in
characteristics of 12-14; Shell study evolution of; company 200, 202-4;
of 9-11, 12-15, 16, 19, 21, 31, 132, monocropping 179, 181
172, 176, 177, 181, 205, 208 mortality rate, companies 203
loyalty 199 Motorola 15, 176

machine for making money, company as negotiation 78

2-5, 208 Newland, Ted 58, 64, 67
management by objectives 55 non-members of organization 192, 213,
management of change 16 237
mapping 85, 89
Maslow, Abraham 145 oil crises 10, 42, 67, 96-9
 THE LIVING COMPANY 250

oil industry 42-4, 52, 56, 58, 88, 96-9, Porter, Michael 195
151-2 potential: development of 18, 76, 126,
Onians, Dick 203-5 142-3, 144, 149, 163, 185, 208, 210,
openness 175, 181, 190, 193, 202, 237 212, 234, 236; human 28, 143-4
outsiders 150-2, 195 power: 220, 222, 224, 231; distributed
224, 227-9, 232, 237
pain, corporate 38, 44 prediction 44-53, 67
Papert, Seymour 80, 84, 91 pride, corporate 19
parasites 197-8, 199-200, 237 Procter & Gamble 15, 19
perception 30, 37, 45, 47, 48, 73, 88, profitability 14, 21-22, 27, 127
189; myths about lack of 37-48 propagation, social 161, 162, 163, 166
persona: 16, 32, 49, 99-100, 104-5, 107- psychology 102-4
11, 112-14, 115, 120, 121, 122, 127, puddle company 125, 139, 144, 153,
132, 171, 192, 203, 213, 216, 236, 156
237; characteristics of 104; ladder purpose, corporate 3, 18, 19, 35, 123,
of 107-9 124, 125, 126, 150, 156, 183, 208-9,
Personalismus 102, 103 210
Petrobras 96-7, 110 Putnam, Robert 146
Philips 14, 222
Piaget, Jean 27, 75, 76, 79, 85, 102, 181 Rambo style of management 83
Pirenne, Henri 22 Rand Corporation 57
planning: 52, 53, 60, 66, 68, 70, 186, recruitment 137-40
189, 223; as learning 93, 187; rejection mechanisms 194
bottom-up 54; financial 53-55 relationships 32, 126, 150, 212-3
Popper, Karl 221 remuneration, executive 132
Porras, Jerry and James: Collins 15-16, resistance: mechanisms 194-5, 197;
21, 138-9, 206, 208, 219
 INDEX 251

to change 38 longlived companies 9-11, 12-15,

Richmond, Barry 86 16, 19, 21, 31, 127-8, 132, 172, 176,
river company 125-6, 127, 131, 139, 181, 205, 208; Year 2000 study 57-8
145, 153, 155 signals, of change 40, 41-2, 47, 49, 60,
robin, evolution of 161, 167 76, 77
Rotterdam, story of mayor of 49-50 simulation 74, 89, 90, 92
Royal Dutch/Shell: 1, 8, 19, 26, 28, 37, soft mapping 90-91
49, 55, 56, 57, 62, 63, 64, 67, 68, songbirds, evolution of 159-60
70, 71, 77, 82, 84, 87, 92, 98, 99- Sony 15
100, 150, 151, 152, 174-5, 195-6, South Africa 115, 236
208-9, 222, 224-7: recruitment in steering 184, 186-7, 223, 229
139; training and development in Stella 86, 88
144, 163-4 Stern, William 100-109, 112-3, 116, 120,
Rubin, Harriet 254 131, 199, 208
stewardship 152, 207
scenario planning 56-69, 87, 189 Stone, Nan 254
Schama, Simon 114 Stora 7-8, 9, 33-4, 44, 174
Schwartz, Peter 59, 63, 67, 84, 158 strategic planning 183-4
self-interest 29, 32, 131, 198, 208, 236 strategy 184, 186, 187, 188, 189, 223
self-knowledge 135 Stratix Group 8
Senge, Peter M. 1-6, 90 success, measures of 207-9
sensitivity to environment 12, 16, 29, Sumitomo 8, 9, 11-12, 173, 174
30, 32-3, 48, 208 Suzuki 11
Shell see also Royal Dutch/Shell: Angola system dynamics 85-8, 89, 91
117-18; Brazil 96-7, 106, 110, 111,
115-21, 115-21, 128, 230-1; Tavistock Institute 80, 81
Ethiopia 118-19; South Africa 116- teams 95, 166-7
17, 119-20, 146-7; study into
 THE LIVING COMPANY 252

Tercentenarians Club 12, 19 119, 120, 121, 126, 127, 128, 131,
territoriality 162, 166-7, 169 132, 133, 135, 136-7, 138, 139, 140,
thinking about the unthinkable 57, 58, 149, 173, 176, 200, 236
59, 66 van der Heijden, Kees 66
time paths 45, 46, 47, 49, 52, 78 van Wachem, Lo 10-11
tolerance: 13-14, 16, 127, 169, 170-71, Varela, Francisco 108, 111, 190-1
174, 175, 179, 180, 182, 183, 184, visionary companies 15, 139, 219
187, 189, 191, 194, 202, 208, 229,
236, 237; and decentralization 172- W.R. Grace 11, 173, 174
7; dangers of 190 Wack, Pierre 58, 59, 60, 63, 64, 67, 85
training and development 163, 164, 165 Wal-Mart 15
transitional objects 81, 84 Walker, Patrick 51
trust 146-7, 149, 154-5, 169, 211, 233, war games 82
236 wealth, sources of 22
Weick, Karl 145
UK companies 12 Welch, Jack 143
uncertainty, reducing 53 Williams, Marjorie 254
Unified Planning Machinery 55-6, 57 Wilson, Allan 158-62
Unilever 13, 222 Winnicott, D.W. 80, 81
unthinkable, thinking about 57, 58, 59,
66 Yergin, Daniel 52
US companies 9, 11, 15
Zenith 15
values, corporate 113, 114, 115, 116,
Acknowledgements
THE MATERIAL FOR THIS BOOK HAS ACCUMULATED OVER MANY
years in a constant dialogue. This dialogue was an integral
part of my being together with my colleagues at Shell — who come
from many countries and from many and varied backgrounds and
with whom one works and competes in the demanding milieu of
a multinational company. There was also the dialogue with the
many people I have met in the world of international business.
More specifically, there was intense dialogue with my colleagues
in Group Planning, many of whom are mentioned by name in this
book because of their specific contributions to the body of
thought on the living, learning company. And around Shell
Group Planning there was and is that network of remarkable peo-
ple, from musicians and film directors to academics and consul-
tants, who provided such a source of inspiration. All together,
they supplied me with the material of which experience is made.
Together we lived through the events which became the growth
buds of questions about companies — their purpose, their very
nature and what that means for their managers. The questions led
to a search for answers and it is this search which forms the back-
bone of this book. In the first month of my first year at universi-
ty, the professor in philosophy looked sternly at us novices and
said, ‘Remember, you will never have an original thought in your
life — every thought, every idea will already have been thought
long ago by somebody else.’ He was right: I owe a great debt to
all my interlocutors in that long dialogue.
 THE LIVING COMPANY 254

The writing of this book would never have occurred were it

not for Harriet Rubin, a graduate in poetry and a publisher with
a sharp intuition, who realized that there could be a book long
before anyone else, with the exception of Napier Collyns. Over
the years, both never ceased to encourage me finally to write that
book. Still, it took Nan Stone, then senior editor of the Harvard
Business Review, a week of her annual leave to coach me through
the material and to help me arrive at the conclusion that, after all,
it might be worth a try. Her prompting and my growing experi-
ence of lecturing in many countries to wide-ranging audiences
finally crystallized and honed the ideas which had germinated in
the rich environment of Shell’s planning coordination.
The book’s inception produced a manuscript which
Marjorie Williams, who saw an early draft, characterized as ‘a
book about a journey, whereas I would like to satisfy my curios-
ity about the country of destination’. Fortunately, Napier
Collyns and Nan Stone succeeded in convincing Art Kleiner, his-
torian and author, to take the time to reshape the manuscript
around its main theme of ‘the living company’. Therefore, I sup-
pose, there is justice in the fact that Marjorie became the book’s
final coach.
Arie de Geus
February 1997