Efficient Organization

Efficient Organization
A Governance Approach

M I K KO K E T O K I V I A N D J O SE P H T. M A HO N EY
 Oxford University Press is a department of the University of Oxford. It furthers
the University’s objective of excellence in research, scholarship, and education
by publishing worldwide. Oxford is a registered trade mark of Oxford University
Press in the UK and certain other countries.

Published in the United States of America by Oxford University Press

198 Madison Avenue, New York, NY 10016, United States of America.

© Oxford University Press 2023

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, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by license, or under terms agreed with the appropriate reproduction
rights organization. Inquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above.

You must not circulate this work in any other form

and you must impose this same condition on any acquirer.

Library of Congress Control Number: 2022920525

ISBN 978–​0–​19–​761029–​9 (pbk.)

ISBN 978–​0–​19–​761028–​2 (hbk.)

DOI: 10.1093/​oso/​9780197610282.001.0001

1 3 5 7 9 8 6 4 2
Paperback printed by Marquis, Canada
Hardback printed by Bridgeport National Bindery, Inc., United States of America
To the memory of Oliver Williamson
 Contents

Preface  ix
Acknowledgments  xv

PA RT I :   F U N DA M E N TA L S O F E F F IC I E N T
O R G A N I Z AT IO N
1. Introduction  3
2. The Efficiency Lens  30

PA RT I I :   G OV E R NA N C E W I T H I N A N D AC R O S S
O R G A N I Z AT IO N S
3. Contracting within and across Organizations  69
4. Stakeholder Analysis  99
5. Nonprofit and Public Organizations  131

PA RT I I I :   G OV E R NA N C E A N D T H E
O R G A N I Z AT IO NA L L I F E C YC L E
6. The Startup Organization  163
7. The Expanding Organization  187
8. The Institutionalized Organization  211

Epilogue  233
Glossary of Terms  237
References  273
Index  283
 Preface

If you own shares in a limited liability company and vote on the composition
of its board of directors at the annual meeting of the shareholders, you are
a designer. If you are one of the cofounders of a startup firm contemplating
how to split equity among the founding team, you are a member in a group
of designers. If you are the founder and the artistic director of a nonprofit
theater thinking of how to curate the upcoming season, you are a designer.
If you are a civil servant analyzing which parts of prison operations can be
contracted out to private firms without jeopardizing the integrity of the
prison facility, you are a designer. If you are a student of business administra-
tion, sociology, law, psychology, or political science interested in organiza-
tions, you may one day be a designer. This is a book written for the designer.
When contemplating design decisions, we invite all designers to embrace
the following simple premise: If there are two alternative ways of organizing,
choose the option that generates the least amount of waste. The challenge
must not be underestimated, because many forms of waste are difficult to
uncover. To be sure, the idea of wasting money and time seems salient. But
how about the idea that attention and communication can be wasted? Or, that
sometimes routines and procedures may develop into forms of “institutional-
ized waste,” that is, waste that has become taken for granted?
Our goal in this book is to enable designers to think of how to identify
and minimize all forms of organizational waste to maintain an efficient or-
ganization. Of all the alternative ways of organizing, the one that produces
less waste than the others is, in economic terms, comparatively efficient.
Importantly, all comparatively efficient alternatives will have flaws, but they
are still the ones the designer should choose, at least until a better alternative
becomes available.
We distinguish between two kinds of efficiency: myopic and sustainable.
The former focuses on short-​term gains through some form of exploita-
tion. For example, a powerful buyer improves its cost efficiency by squeezing
every penny out of its small suppliers by aggressively renegotiating contracts
at contract renewal; an employer demands that employees work over-
time without additional pay; those who possess information not available
x Preface

to others use the information asymmetry to their advantage. These are all
examples of myopic efficiency.
We are not interested in writing about the benefits of opportunistic be-
havior and short-​term gains. On the contrary, our intention is to enable
forward-​looking designers to organize in ways that promote sustainable ef-
ficiency: eliminating the excess, not the essential; nurturing cooperation,
not jeopardizing it; seeking mutual value creation, not selfish value capture.
We still subscribe to self-​interest, just not its strong-​form manifestations
of opportunism and exploitation. We are interested in myopic efficiency
only insofar as the honest designer can create governance structures that
discourage those inclined to act opportunistically. Nobel Laureate Oliver
Williamson, to whom we have dedicated this book, famously counseled that
“the world should not be organized to the advantage of the opportunistic
against those who are more inclined to keep their promises” (Swedberg 1990,
126). By embracing the objective of sustainable efficiency, our book joins
Williamson’s cause.
Economist Frank Knight (1941) was the first to describe an economically
efficient organization as one that succeeds in eliminating waste. Because we
want the word efficiency to invoke positive connotations, Knight’s descrip-
tion creates the appropriate association.
However, seeking to minimize even the excess may sometimes miss the
point of sustainable efficiency. It is well established that designers often con-
sider some forms of excess, or organizational slack, desirable even when
they seek to be sustainably efficient: (1) industrial firms keep inventories as
buffers for demand uncertainty; (2) much like a basketball team wants to
have several substitute players available “on the bench,” professional service
firms maintain a “healthy bench” by assigning no more than, say, 90 per-
cent of their professionals to client projects at any given time; (3) it is well
advised not to have organizational members process the maximal amount
of information but, rather, “give them some (cognitive) slack;” and so on.
Organizations maintain at least some slack for the same reason we all main-
tain a safe distance to the car in front of us while driving—​to buffer against
the unexpected.
Organization scholars Richard Cyert and James March ([1963] 1992) noted
that non-​zero slack makes organizations interesting. We not only concur but
also suggest that deciding when and where to introduce slack into an organ-
ization and how to govern it belong to the heart of governance and to the
designer’s agenda.
 Preface  xi

To be sure, our intention is not to suggest that efficiency should be the

designer’s only objective. We propose that organizational efficiency can use-
fully be thought of as analogous to a person’s blood pressure. In your annual
medical exam, your doctor seeks to obtain an accurate reading of your blood
pressure because your blood pressure is relevant, not because it is the most
important thing about your health. Then, if your blood pressure is uncom-
fortably high, the doctor will invite you to a conversation about a possible
intervention. At the risk of insulting your intelligence, let us state the ob-
vious: The doctor’s prescription does not mean that your life from that mo-
ment on should be about how to manage your blood pressure, let alone that
the reason for your existence is to maintain a low blood pressure.
Echoing the blood pressure analogy, we do not propose that the purpose
of an organization is to be efficient; we merely suggest that efficiency merits
attention. With those who embrace the premise that efficiency matters, we
engage in a conversation on the ramifications. With those who reject it, we
would still try to strike a conversation on the ways in which avoiding waste
might be relevant. This, too, is analogous with the doctor-​patient relation-
ship. Whether your blood pressure is of interest to you or not, your doctor
will try to persuade you to pay attention to it. If you choose to ignore the
issue, your doctor immediately withdraws all prescription. At the same time,
your decision to ignore your high blood pressure changes neither the doctor’s
conviction nor the fact that a high blood pressure can be hazardous to your
health.
Just as most physicians are convinced blood pressure is relevant to a
person’s health, we are convinced that sustainable efficiency is relevant to all
organizations. We also trust that just like most of us choose to give attention
to our blood pressures, most designers are sufficiently motivated to give at-
tention to efficiency. If the organization in question is the designer’s own, the
motivation arises directly from the fact that the designer is likely the primary
bearer of the costs of inefficient organizing. If the designer is in someone
else’s employ, the chances are there is a contractual or a fiduciary duty to act
in the best interest of the organization. Sustainable efficiency can often be
argued to be in the best interest of the organization.
To avoid confusion, it is crucial to emphasize from the beginning that an
organization need not seek profits for efficiency to become a relevant objec-
tive. At its foundation, the pursuit of efficiency is a quest for collaborative
value creation. This may involve not only adding economic value measured
in dollars, euros, or yen, but also treating patients suffering from depression,
xii Preface

teaching children new ways of thinking, or contributing to the rehabilitation

of prison inmates. All these may ultimately contribute to the creation of ec-
onomic value (and possibly profit), but economic value is at best a distant
outcome, not the designer’s immediate objective.
At the same time, both for-​profit and nonprofit organizations often have
economic slack in the form of net worth. How net worth is governed must
obviously be on the designer’s agenda. Therefore, an economic surplus does
ultimately become relevant to the designer, just not in the sense of how to
make it but how to govern it.
As a central point of terminology, the word organization in this book is not
used merely in reference to entities we colloquially think of as organizations.
Although organizational entities such as universities, firms, and legislatures
are relevant, they do not represent the essence of organization. In this book,
the word organization has less reference to entities and more to the principles
and practices by which those representing these entities organize coopera-
tive relationships with one another. In short, organization as entity is relevant
but organization as cooperation is essential. Accordingly, the efficiency ques-
tion pertains to efficiency not of entities but of cooperation: Is this relation-
ship organized efficiently? This question echoes a well-​established position
in organization economics that takes the relationship as the unit of analysis.
Since we are both academics trained in business economics, it should not
come as a surprise that we invoke economic theories and economic thinking
more broadly throughout this book. We have been greatly inspired and
influenced by numerous organization economists: Armen Alchian, Kenneth
Arrow, Jay Barney, Ronald Coase, John R. Commons, Harold Demsetz,
Eugene Fama, Sanford Grossman, Oliver Hart, Friedrich Hayek, Bengt
Holmström, Michael Jensen, Gary Libecap, Elinor Ostrom, Edith Penrose,
Michael Porter, and Oliver Williamson. Just like we the authors of this book,
all these economists have been interested primarily in efficient governance
and sustainable value creation through collaboration at various levels of anal-
ysis; surprisingly few economists focus on value distribution. Indeed, we sug-
gest that the question of value appropriation—​“who gets what?”—​requires
an extensive analysis of bargaining capabilities, power, influence, and organi-
zational politics. We leave these topics outside the scope of our book.
As interdisciplinary academics interested in practical problems of or-
ganization design and governance, we have also been strongly influenced
by social scientists more broadly: management and organization scholars,
psychologists, sociologists, legal scholars, and political scientists. We have
 Preface  xiii

found the works of the following individuals particularly insightful: Chester

Barnard, Lucian Bebchuk, Margaret Blair, Alfred Chandler, Robert Clark,
Richard Cyert, John DiIulio, Amy Edmondson, Jay Galbraith, Mark
Granovetter, Albert Hirschman, Daniel Kahneman, David Larcker, James
March, Richard Nelson, Douglass North, Mancur Olson, Philip Selznick,
Herbert Simon, Arthur Stinchcombe, Lynn Stout, James Thompson, James
Q. Wilson, and Sidney Winter, among scores of others.
As academics, we have both a professional and an ethical obligation to
give credit where credit is due. Throughout the chapters of this book, we ac-
knowledge scholarly work by citing the central contributions, as we would
in regular academic texts. Therefore, if this book exhibits the character-
istics of academic texts, this stems from the central duty of the academic.
Do not, however, let these characteristics mislead you to think that this is an
academic book written for an academic audience. By citing the relevant re-
search literature, we are merely fulfilling the dual duty of the academic: both
conveying the central ideas and identifying their authors.
Citing the research literature ultimately serves the reader as well. We en-
courage those readers who want to dig deeper into the topics to look at some
of the works cited, as they can be sources of further learning and insight. In a
book such as ours, we neither want nor can go into deep detail on every topic.
All the works cited in this book merit reading as they are highly complemen-
tary to our unavoidably limited exposition. We want to emphasize this espe-
cially to those readers who are students of organization at universities. The
vast majority of the academic articles cited in this book are available in elec-
tronic format and can be accessed through publisher’s web pages and digital
libraries.
The importance of citing the published literature brings us to the moti-
vation for writing this book. With each passing year as educators of man-
agement students, executives, entrepreneurs, and organizational leaders, we
become increasingly convinced that the numerous foundational messages
in the academic literature on organization economics can benefit the de-
signer. The problem is that these messages remain buried in the literature
published over the past hundred years in scores of specialized academic
texts. Furthermore, the key messages are not found in a particular book or
article as much as they are woven into the fabric of broader conversations and
debates, each spanning multiple decades and involving dozens of scholars
as participants. Understanding any individual message requires an under-
standing of the broader conversation, not just an individual book or article. If
xiv Preface

in this book we succeed in translating even a small fraction of these messages

into a language that is accessible to the designer, this book will have more
than served its purpose.

In Madrid and in Champaign, on June 3rd, 2022

Mikko Ketokivi
Joseph T. Mahoney
 Acknowledgments

We dedicate this book to the memory of organization economics giant and

Nobel Laureate Oliver Williamson, whose writings have been the single most
important influence on the thoughts expressed in this book. This dedication
is particularly fitting also because our collaboration as coauthors was sparked
by a common interest in Dr. Williamson’s theory of transaction costs. That
we were introduced to one another by someone neither of us knew well and
whom one of us never even met is as mysterious as it is precious.
We are indebted to scores of practitioners whom we have come to know
personally and professionally during our careers: in the classroom, in our re-
search projects, through our consulting, or in casual conversations. We thank
especially the following executives, managers and organizational leaders,
who either generously devoted their time to comment on our chapter drafts,
provided illustrative examples, or otherwise engaged us in conversations
on the ideas presented in this book: Benjamin Barraza, Deema Bibi, Adrian
Blockus, Nigel Brashaw, Salvador Cerón de la Torre, Bastian Gerhard, Brett
Hinds, Belinda Holdsworth, Monty Hoxie, Yasalde Jiménez, Sean Joyce, Jussi
Kaulio, Ian Kesler, Sheree Kesler, Marja-​Liisa Ketola, Patrick Küttner, Sofian
Lamali, Markku Lappalainen, Matías Lira, Rauf Mammadov, Shaffi Mather,
Carlo Meloni, Jaakko Nevanlinna, Lee Newman, Jeremy Nurse, Alfred Ortiz,
Alejandro Pacheco, Suzanne Peters, Salma Qarnain, Efrain Rosemberg, Eila
Sailas, Jarmo Salminen, Yasmina Suleyman, Bram van Olffen, Francisco
Vazquez, and Joachim von Goetz.
We also thank our colleagues and coauthors in the academic community
for providing insightful feedback either on this book or on our work on the
economics of organization more generally: Nick Argyres, Janet Bercovitz,
Akhil Bhardwaj, Lyda Bigelow, Dan Breznitz, Phil Bromiley, Tyson Browning,
Joep Cornelissen, Felipe Csaszar, Suzanne de Treville, Luis Garicano, Ranjay
Gulati, Martin Kenney, Peter Klein, Yasemin Kor, Juha-​Antti Lamberg, Mike
Manning, Saku Mantere, Scott Masten, Kyle Mayer, Jackson Nickerson,
Joanne Oxley, Mirva Peltoniemi, Christine Quinn Trank, Petri Rouvinen,
Fabrizio Salvador, Roger Schmenner, Brian Silverman, Kalyan Singhal,
Deepak Somaya, Andy Van de Ven, Jouni Virtaharju, and Pekka Ylä-​Anttila.
xvi Acknowledgments

We are also grateful to all the anonymous peer reviewers who have given
us constructive feedback and invaluable guidance during countless journal
review processes. We recognize the double-​blind review process as the bed-
rock institution of the academic community—​it is always about the message,
not the messenger. As in all our texts, the standard disclaimer applies: All
errors, oversights, and omissions remain our responsibility.
Every page of this book echoes the central premise that securing long-​term
contractual relationships requires a discriminating approach to contracting,
conscious foresight, and safeguards to serve as buffers against uncertainty. In
precious contrast to the always uncertain and sometimes hazardous world
of organizations, the very presence of Eva De Francisco and Jeanne Marie
Connell in our lives serves as a constant reminder that in the truly mean-
ingful relationships, this central premise can be tossed to the winds without
hesitation.
 PART I
FUNDA ME N TA L S OF E F F IC I E NT
ORGA N IZ ATION

The first part of the book consists of two chapters: c­ hapter 1 (“Introduction”)
and ­chapter 2 (“The Efficiency Lens”). The purpose of these two chapters is
to lay the conceptual and intellectual foundation for the chapters that follow.
Laying the foundation involves an explication of both what this book is and
is not about. All approaches to the complex topic of organizations—​whether
economic, sociological, political, or psychological—​ constitute a way of
seeing. Our way of seeing the organization is through the lens of efficiency.
Because the word efficiency tends to invoke negative connotations,
such as powerful buyers pressuring their smaller suppliers by aggressively
renegotiating contracts, employers making employees work more hours
without additional pay, and so on, we make the crucial distinction between
myopic efficiency and sustainable efficiency. The objective of the first part of
the book is to establish the latter as our primary focus: eliminating the excess,
not the essential; nurturing cooperation, not jeopardizing it; and enabling
mutual value creation, not unfettered self-​interest.
In c­ hapter 1, we clarify our general approach to organization design and
governance by explicating five foundational premises, all based on various
economic approaches to organization. Most importantly, our focus is on the
deliberate and rational choices organization designers should make to en-
sure sustainable efficiency and value creation. All organizations have a job
to do, and the designer’s task is to ensure this job is performed in a way that
minimizes waste in all its forms. Chapter 1 also serves as a general introduc-
tion to the entire book. To this end, an overview of the book’s structure and
its chapters is provided at the end of ­chapter 1.
Chapter 2 introduces the Efficiency Lens, an analytical tool that the de-
signer can use to seek efficient organization. The Efficiency Lens consists
of three central elements of management, oversight, and risk, their
2  Fundamentals of Efficient Organization

interrelationships, and their dynamics. We apply the Efficiency Lens

throughout the chapters of this book.
Finally, to support the first part of the book in particular, there is an ex-
tensive Glossary of Terms at the end of the book. In the glossary, we define
and describe the central concepts used throughout the book. In order for the
book’s key messages to become salient to the reader, consistent use of termi-
nology is crucial.
 1
Introduction

You stand on the corner of Calle Alberto Aguilera and Calle Acuerdo in
Conde Duque, in the heart of Madrid, looking for a taxi to take to a birthday
party in La Moraleja, a twenty-​minute taxi ride away. As you see a free taxi
approaching, you wave your hand to let the driver know that you need trans-
portation. The driver pulls over, you hop in the back seat, and give the driver
your destination. The driver nods, turns on the taximeter, and you are on
your way. In a world of a billion transactions, there is one more.
Even though the chances are you will not spend much time thinking about
the specifics of the transaction, let us consider for a moment organization
economist Oliver Williamson’s (1999, 321) recommendation that “even that
which is obvious can sometimes benefit from explication.” What exactly
happens during the twenty-​minute contractual relationship between you
and the taxi service provider? What are all the factors that are required to
make the transaction work smoothly?
Let us go back to the start of the journey. As you entered the taxi, you made
a point of instructing the driver to avoid Paseo de la Castellana, because there
is a football game at the Santiago Bernabéu stadium. With eighty thousand
football enthusiasts piling up into the stadium, northbound Castellana is
going to be in a deadlock. You know the driver is already well aware that
there is a game in town (the pre-​game broadcast is playing on the radio), but
your point was not so much to give the driver instructions as it was to signal
knowledge; flashing your street smarts at the driver has benefits. The driver,
being no stranger to such signaling, realizes your intent but takes no offense;
the driver understands that it is reasonable for a customer to worry about
being taken, more than just literally, for a ride. Indeed, it has happened to you
a few times, and you have picked up a few tricks and have learned to adapt.
In addition to learning over time, technology helps. Most of the time, you
use the taxi app on your smartphone, which gives you a guaranteed max-
imum rate if you enter the destination address when you place the order.
Moreover, in case the service is not to your expectations, it takes you all
of ten seconds to use the app to engage in ex post monitoring and submit a

Efficient Organization. Mikko Ketokivi and Joseph T. Mahoney, Oxford University Press. © Oxford University Press 2023.
DOI: 10.1093/​oso/​9780197610282.003.0001
4  Fundamentals of Efficient Organization

performance evaluation of the driver at the end of the journey. Another pre-
caution at your disposal is ex ante screening—​you may limit the choice of
taxis to those with the highest, five-​star rating.
On reaching the destination, you pay the driver €20 for the fare, which is
smack dab in the middle of the price range the app gave you as you placed
the order. You ensure you do not leave anything behind and given your pos-
itive experience, reward the driver with a five-​star rating on the app. Both
transacting parties have received what they expected from the transaction,
supply has met demand—​the market clears—​and in a world of a billion
transactions, there is one fewer.
From start to finish, there were no resources wasted in the transaction: no
surprises, no haggling, and no deception. Furthermore, although both ex-
change parties acted out of self-​interest (which is what makes the economy
run in the first place), neither behaved in a self-​serving, opportunistic way.
It is not always this smooth, but this is an accurate description of your ex-
perience 98 percent of the time. In fact, you do not even remember the last
time you had a problem, and the biggest problem you have ever had in the
past eight years was once paying roughly €7 more than what you had ex-
pected. Nothing to lose your sleep over. You gave the driver a one-​star rating,
and after blacklisting him, your paths never crossed again. When specificity
in the transaction is low (when you are not dependent on any specific taxi
driver), you can easily and costlessly walk away from an unpleasant en-
counter. Therefore, avoiding recurring transactional hazards is straightfor-
ward. Symmetrically, honest taxi drivers benefit from not having to pick up
the same troublemakers time after time. Customers are well aware that taxi
drivers rate customers as well, and just as you have access to their ratings,
they have access to yours. In this sense, there is no information asymmetry.
Let us suggest that the story does not end at getting where you need. We
might also wonder whether it would make economic sense for you to own a
car. Conducting the make-​or-​buy analysis is not complicated. If you lived in
downtown Madrid and owned a car, you would pay at least €250 per month
for parking, €80 per month for insurance, and another €120 for gasoline.
Finally, as far as the financial investment is concerned, your capital equip-
ment would operate at a less-​than-​10-​percent capacity utilization rate (less
than two hours a day, on average), spending the vast majority of its time sit-
ting idle, depreciating at a rate of €300 per month. It is difficult to think of a
more abysmal target for a €30,000 capital investment, particularly since there
is an abundant and diverse supply of reliable transportation services in the
 Introduction  5

city. Why insource (“make”) transportation when you can easily outsource
(“buy”) it?
Having considered the issue in its entirety, spending €500 per month to
have a professional drive you around in Madrid sounds less like a luxury and
more like economic common sense. This comparatively lower cost is fur-
ther reduced significantly if you use the subway instead of taking the taxi.
The point is that even when a comparative analysis involves the most expen-
sive, most convenient, and most flexible transportation outsourcing option,
having someone else drive you around is more economical than driving
yourself. In addition, when you consider hybrid governance options such as
a private lease, you find that their total cost is roughly the same as the cost of
ownership.
Finally, as you set aside pure self-​interest to consider the collective interest
and, in particular, the negative externalities involved, you cannot avoid the
conclusion that while the jam-​packed metropolis of over six million people
needs many things, yet another private vehicle, to be operated at a very low
capacity utilization rate, is perhaps not one of them. You might therefore
choose not to contribute to the roughly 10 percent annual growth rate in the
number of registered passenger vehicles in Madrid.
For the taxi driver, the world looks very different. Immediately after
terminating the transaction with you, the driver is not only free but strongly
incentivized to enter into a new transaction with another customer as quickly
as possible. The driver wants to make the most out of a capital investment
that involves not only the vehicle but also the taxi license, which costs around
a €100,000. A casual survey of taxi drivers reveals that an individual driver
will work anywhere between eight and fourteen hours each day for five or six
days a week.
Given the sizable capital investment, the driver has a high-​powered incen-
tive to provide a reliable service and to maintain a five-​star rating. Another
casual survey of taxi drivers reveals that they value and seek to maintain their
five-​star rating. Furthermore, is it just a coincidence that concurrently with
the introduction of the rating system, vehicle cleanliness improved signif-
icantly? Or that additional services such as phone chargers, breath mints,
and hand sanitizer are now offered to passengers? Might these changes have
something to do with the fact that your taxi app lets you provide a star rating
not only to the driver but also the vehicle?
In this example of a mutually credible transaction, supply met demand,
and the contract was completed without unnecessary ex ante or ex post
6  Fundamentals of Efficient Organization

transaction costs. Both exchange parties not only entered into the transac-
tion voluntarily but were also able to allocate their time and efforts to what
they preferred and to what they were skilled at doing. You chose to outsource
driving to a professional and spent the twenty-​minute ride sending two quick
emails and calling the restaurant in La Moraleja to ensure that everything
was ready for the birthday celebration.
In sum, the transaction was efficient and mutually beneficial in all relevant
respects, even though you probably did not spend a single second thinking
about it. This unearths yet another benefit of efficient contracting: It enables
efficient allocation of not only our time and money but also our attention.
Indeed, sometimes the very act of having to think about something can be a
form of waste.

Efficient Organization

The transaction associated with the taxi ride is an example of efficient organ-
ization. Efficiency is a relevant concern at all levels of analysis: individuals,
organizations, industries, communities, even entire societies. Who would
object, at any of these levels, to being efficient?
This book is about efficient organization, where the word organization is
used in the broadest possible sense. It refers to any form of voluntary coop-
eration, be it between individuals or broader entities such as firms or other
collectives. A massive multinational corporation is an organization, but so
is the taxi ride example. The punchline of the taxi ride example was that a
comparative economic analysis reveals that it is more efficient for a person
residing in downtown Madrid to outsource private transportation to a taxi
service than to own a private vehicle.
Because boundary conditions are always essential, let us be explicit about
what the efficiency view does not cover. In the taxi ride example, we ab-
stracted out questions such as the following: How meaningful is owning, or
not owning, a car to you? Do you enjoy, or not enjoy, driving? How skilled
a driver are you? How well do you know the city? How highly do you value
convenience? Do you have enough money to buy a car? These questions may
obviously be of interest to some of us, but because they are not about eco-
nomic efficiency, they are outside the scope of this book.
The last question in particular merits attention. In the efficiency per-
spective, and in many economic approaches more generally, various wealth
 Introduction  7

effects (Milgrom and Roberts 1992, 35–​39) are abstracted out of the analysis.
Although this may seem strange at first, a brief reflection should reveal that
whether owning a car is economically more efficient than using taxi services
does not depend on the wealth of the transacting parties. To be sure, wealthier
individuals are both more likely to use taxi services and less likely to be taxi
drivers. But we were not interested in the specific individuals involved in the
exchange, our aim was to determine which private transportation alternative
created less economic waste. Therefore, even though you and the taxi driver
were the two protagonists in the example, the story was ultimately not about
you or the taxi driver, it was about the comparative long-​term cost of alterna-
tive forms of transportation. Accordingly, we are not psychologists who “try
to get inside the heads” of individual decision makers in an attempt to under-
stand the choices they make.
At the same time, no matter what one’s personal preferences are, explic-
itly analyzing the efficiency implications of one’s choices can be useful. That
the fixed cost of owning a mid-​sized passenger vehicle in downtown Madrid
is more than €600 per month came as a surprise to a colleague of ours who
owns a private vehicle in Madrid. One reason the total cost may not be sa-
lient is because it is scattered over various monthly, semiannual, and annual
payments in different cost categories. The total cost becomes salient only if
one engages in an explicit calculation. Herein resides a learning opportu-
nity: Becoming better educated about the total cost of car ownership might
end up affecting one’s personal preferences.

On the Use of Economic Jargon

Philosopher Paul Feyerabend noted in his autobiography (Feyerabend

1995) that philosophers cannot understand one another until they have
translated everything into Latin. In light of Feyerabend’s remark, what are we
to think of the use of terms such as high-​powered incentives, ex ante screening,
ex post monitoring, hybrid governance, specificity, and negative externalities?
Why speak Latin in a book written primarily for nonacademics?
In the taxi ride example, the terms borrowed from economic theories of
organization appeared italicized; there were quite a few of them. The use of
unduly sophisticated language is exactly what is wrong with academics, you
may think. Were these really essential terms to describe a taxi ride? We sug-
gest that the answer is both no and yes. The answer is an emphatic no in the
8  Fundamentals of Efficient Organization

sense that specialized terminology is not required to describe the process of

taking a taxi to a birthday party. From the descriptive point of view, speaking
Latin is as irritating as it is unnecessary.
But suppose we want to understand and analyze why a private individual
obtaining a personal transportation service from a private entrepreneur
was organized the way it was. The particular taxi ride aside, why were there
15,974 taxis in Madrid in 2020? Why is there a market for this type of private
transportation? In what sense is it economically rational? Does it help reduce
waste? Are there comparatively more efficient alternatives? After our brief
economic analysis, we concluded that owning a private vehicle in down-
town Madrid surely seems like an economically poor choice. Consequently,
we are not surprised at the thousands upon thousands of taxi cabs shuttling
the streets of Madrid. The emergence of not only ride-​sharing but also car-​
sharing arrangements is equally unsurprising. In Madrid, rental cars are
currently available at about twenty eurocents per minute by car-​sharing serv-
ices such as ShareNow. These outcomes are predictable consequences of the
simple economic reality that owning a private vehicle in downtown Madrid
is economically wasteful.
We suggest that when the objective is to analyze and to understand, spe-
cialized terminology becomes useful. No matter how strange the terms may
at first glance appear, in economic conversations and analyses they have
specific meanings and definitions. Economists in particular and academics
more generally are skilled at using terms with precision. Indeed, that is how
we have been trained, and the reason we “speak Latin” is because we want to
ensure the use of terminology is consistent.
If we used colloquial terminology, we would face the problem of different
people using the same term in different meanings. Just ask a dozen executives
how they define profit, and you will get at least a half dozen different answers.
Some focus primarily on return on equity, others on return on assets, return
on capital employed, or economic value added. We must be careful with the
assumption that terminology can be taken as commonly understood. The
upside of using specialized terminology is that it gives you, the reader, useful
pause and compels you to think about what the concept means and why it is
introduced.
Consider the notion that the taxi driver has a high-​powered incentive to
offer taxi services to as many customers as possible during a work shift. As
strange as the term may seem at first glance, it succinctly captures an idea that
would take considerably longer to describe in colloquial terms. Moreover,
 Introduction  9

suppose we now formally introduce the opposite, low-​powered incentive. The

term is likely immediately understandable to the reader without additional
clarification. To make the term and the distinction salient, we might present
hourly wages as an example of a low-​powered incentive; an incentive is low-​
powered when the payoff does not depend on how the work is performed.
Piece rate is an example of a comparatively high-​powered incentive; under
the piece-​rate system, the employees’ compensation is tied to how they per-
form the work.
Once we have introduced high-​and low-​powered incentives (two ends
of a continuum), we can now introduce the general term incentive intensity.
And as we will show in the chapters that follow, incentive intensity is a central
design variable.
Let us look at some of the important nuances of incentive intensity that
may not be immediately salient. First, in applying a high-​powered incentive
it is not the effort but the outcome that is rewarded. Second, incentive inten-
sity is not about whether the absolute levels of compensation are high or not,
it only determines how strongly compensation is tied to outcomes. Indeed,
the reason taxi drivers in Madrid may work up to fourteen hours a day is
because a regular eight-​hour shift is woefully inadequate to provide a suffi-
ciently high absolute level of income. In other words, the high-​powered in-
centive motivates drivers to work more, but the fact that absolute levels of
pay are low forces them to work a lot more. COVID-​19 made the situation
downright unbearable. One driver told us that in April 2020, he picked up
a total of five passengers during a fourteen-​hour shift, that is, roughly one
customer every three hours. After deducting all fixed and variable expenses,
a Madrid taxi driver makes no more than a few euros per trip. Even with a
high-​powered incentive, the absolute level of compensation may be seriously
inadequate.
Once we have understood incentive intensity, we can speak efficiently
about incentives without risking confusion. In the case of the taxi driver, the
driver is motivated to find as many clients as possible during the work shift.
This high motivation is due to a built-​in high-​powered incentive embedded
in the governance structure of taxi services. No extra incentives or motiva-
tional pep talk are required, and no additional resources are required to mon-
itor the driver’s behavior during the work shift. The effect of a high-​powered
incentive is relentless self-​regulation of the driver’s effort, which constitutes
a highly efficient mechanism in a decentralized and fragmented organiza-
tion that consists of over fifteen thousand independent entrepreneurs. In this
10  Fundamentals of Efficient Organization

case, being one’s own boss is highly efficient. In some other contexts, it may
not be.
Finally, clear definitions and selective use of specialized terminology are
also useful whenever it is important to avoid misleading connotations that
colloquial terminology sometimes creates. For instance, it is crucial to un-
derstand that in this book, efficiency pertains to the way in which tasks are or-
ganized, and, specifically, whether in performing these tasks, time, resources,
or effort is wasted. An efficient organization minimizes waste.
Importantly, minimizing waste is about productivity. Whether being ef-
ficient translates into an economic surplus (profit) is an altogether different
question, as is the question whether the organization even seeks a profit in
the first place. Insofar as efficiency is concerned, profit-​seeking is not nearly
as relevant as we may think. In fact, a for-​profit and a nonprofit organization
may often seek efficiency in surprisingly similar ways.
In this book, we use economic terminology, but we do it sparingly, be-
cause the aim is not to teach the reader economics, and we do not want to
write a book that requires a companion economics dictionary to decipher.
We use economic terminology with the primary objective of being able to
analyze various decision situations in a rigorous and consistent way. We
want to ensure that all economic terms are not only carefully defined but
also accessible. To this end, and to make this book self-​sufficient in terms of
its terminology, we have provided an extensive glossary of key terms at the
end of the book.

Five Foundational Premises

The dilemma we face in writing about something as complex as organiza-

tions is that all approaches aimed at understanding organization—​be they
economic, sociological, political, or psychological—​offer a limited point of
view. Therefore, any individual perspective abstracts out much of what might
be relevant for real-​life organizations. Indeed, nothing is outside the scope of
real-​life organizations.
Unlike problems that call for solutions, dilemmas call for choices. On the
one hand, a holistic approach to organization design sounds appealing, be-
cause it would do better justice to the complexity and ambiguity of organ-
izations. At the same time, choosing an eclectic approach runs a risk that
is best captured by financial economist Michael Jensen’s (2001, 9) critique
 Introduction  11

of the stakeholder literature: “Without the clarity of mission provided by a

single-​valued objective function, companies embracing stakeholder theory
will experience managerial confusion, conflict, inefficiency, and perhaps
even competitive failure [ . . . ] [W]‌hen there are many masters, all end up
being shortchanged.”
We agree that without a clear objective, it is difficult to present coherent
arguments about organization design and governance. Because we want this
book to exhibit a “clarity of mission” (to use Jensen’s words), we have chosen
not to write an eclectic book that takes different points of view along the way
and, predictably enough, arrives at on-​the-​one-​hand/​on-​the-​other-​hand-​
types of conclusions. Although we are in general sympathetic to giving voice
to different perspectives on organization design and governance, we focus in
this book on economic efficiency.
But what exactly does the focus on efficiency entail? What are the
consequences for analysis? Having chosen the focus of efficient organizing
and voluntary cooperation, we can delineate five specific premises on which
this book rests.

First Premise: Organizations as Deliberately Designed

Even though we find various evolutionary approaches useful and enthusiasti-

cally embrace emergence as essential to many organizational phenomena, we
choose to approach organizations primarily as deliberately designed entities.
Throughout this book, we use the term designer in reference to all those who
make deliberate design decisions.
To propose that analyzing organizations as deliberately designed is useful
should not be too controversial. For example, in the annual meeting of the
shareholders of a limited liability company, we can think of the shareholders
as the designers whose task is to make the (incremental) design decision of
whether the composition of the board of directors should be changed. In the
founding discussions and negotiations of an industrial startup, the prospec-
tive founding partners are the designers who make the (foundational) design
decision of how many shareholders the firm will have, how equity will be
split, and whether both cash and in-​kind contributions will be accepted in
the purchase of shares.
To clarify, this premise does not suggest that emergent phenomena cannot
give rise to design decision situations—​they can, and they often do. For
12  Fundamentals of Efficient Organization

example, a buyer-​supplier relationship may start as a simple transactional re-

lationship where the buyer chooses one supplier from many alternatives; the
identity of the supplier does not matter, in a manner of speaking. However,
the relationship may transform over time into one where the specific supplier
becomes a preferred supplier; identities start to matter. In the organization
economics literature, this process is called the fundamental transformation
(Williamson 1985, 61). The term is fitting, because there is indeed something
fundamental about how the relationship transforms into one where the iden-
tity of a particular supplier (or buyer) starts to matter. As a thought exercise,
just think how fundamentally the market for taxi services would transform if
you as the consumer had to use the same taxicab every time and a given taxi
driver could serve only a predetermined set of individual customers.
When we say our focus is on deliberate design, we mean that insofar as the
fundamental transformation is concerned, we are not interested in why and
how it occurs; the process is likely highly complex and nuanced. We focus on
analyzing its governance implications, the kinds of design decision situations
it presents, and the kinds of adaptation challenges it creates for the designer.
In sum, emergent phenomena may lead to situations where deliberate design
decisions are required.
The deliberate design focus, particularly when considered in con-
junction with the efficiency premise, also leads us to focus on remedi-
able problems. Specifically, when we discuss various organization design
decisions, we seek to identify solutions that are not only feasible but
have the potential of solving efficiency problems—​this is the essence of
remediability. This approach is important for at least two reasons. One
is that it is sometimes difficult to see how a proposed solution is fea-
sible in the first place. Feasibility might be jeopardized by government
regulations, legal requirements, and corporate cultures; sometimes even
the organization’s information systems (e.g., Goold and Campbell 2002b).
The comparatively efficient alternative may not always be feasible from
an organizational perspective. For example, it is often not possible for an
industrial firm to produce a component in-​house (“make”); instead, the
firm must purchase it from a supplier (“buy”). A case in point, although
national railways such as Indian Railways may own and operate locomo-
tive assembly plants where locomotive assembly and maintenance are
performed, it would be simply infeasible for Indian Railways to produce
internally many of the components and subsystems required in final as-
sembly (Bhardwaj and Ketokivi 2021).
 Introduction  13

The other, more subtle reason is that sometimes even feasible solutions
ultimately fail to solve problems and improve efficiency. Evidence regarding
outsourcing decisions offers a good example. Many companies find, after the
fact, that an outsourcing decision turned out to be associated with hidden
costs that ultimately made outsourcing less efficient than in-​house produc-
tion. This outcome runs counter to the idea of remediability, which holds
that implementing a design decision must result in net gains. The intended
efficiency gains of a design decision must offset all undesirable (and possibly
inadvertent) efficiency losses that the decision might cause in other parts
of the organization (Williamson 1975, 79). An industrial firm outsourcing
the production of a component to a specialized parts supplier may find that
enforcing the buyer-​supplier contract on a daily basis has led to additional
administrative costs that offset any production productivity benefits.
Focusing on net gains underscores the importance of understanding or-
ganization design and governance decisions in their entirety—​this is one of
the central governance principles in the organization economics literature
(e.g., Williamson 1996). In ­chapters 4 and 5, we examine, among other issues,
whether the board of directors should be opened to employee participation.
In the spirit of investigating net gains, we point out that whereas employee
participation on boards may work toward addressing potential problems in
the governance of employer-​employee relationships, it may create tension
and friction elsewhere in the organization.

Second Premise: Start at the Main Problem

Should the chief executive officer (CEO) also chair the board of directors?
Should a component be outsourced or produced in-​house? Should country
organizations within a multinational corporation be assigned profit-​and-​
loss (P&L) responsibility? Addressing various organization design and gov-
ernance questions requires the formulation of a more specific problem that
the designer seeks to address. Problems are never a given; instead, an issue
becomes a problem only after the decision makers have explicitly agreed
on how it should be framed. In this book, we use the term main problem
in reference to problem framing. Before evaluating potential governance
alternatives, the designer must define the main problem.
As an example of formulating the main problem, consider the question
whether the CEO should chair the board of directors—​this is sometimes
14  Fundamentals of Efficient Organization

dubbed CEO duality structure (Rechner and Dalton 1991). In some contexts,
CEO duality makes sense, in others the opposite, CEO independence struc-
ture, is a better option. In the CEO independence structure, the CEO may
be a board member but cannot chair the board. But what exactly is the gov-
ernance problem that the designer seeks to address with CEO duality (or
independence)?
Both CEO duality structure and CEO independence structure are
organization-​specific responses to a specific representation of the main
problem. Those advocating the CEO independence structure are more likely
to frame the problem as an agency problem (Ross 1973; Jensen and Meckling
1976): How does the organization ensure the CEO (the agent) acts in the best
interest of the organization (the principal)?
If the main problem is the CEO potentially not acting in the best interest
of the organization, the CEO independence structure works toward suffi-
cient separation of powers. The problem of insufficient separation is further
exacerbated if the other board members are not independent of the CEO.
Even an outside board member may satisfy all the requirements of being for-
mally independent of the firm but simultaneously be in one way or another
beholden to the CEO. A recurring concern that several proxy advisory firms
have expressed regarding Tesla, for instance, is that Tesla’s board members
have been formally independent of the firm but not truly independent of
CEO Elon Musk’s influence. Indeed, Tesla’s board has faced several lawsuits
where shareholders have claimed the board has not acted in the best interest
of the company. In 2018, Elon Musk was forced to step down as Tesla’s chair-
person of the board but remained Tesla’s CEO. Clearly, there were concerns
that the CEO duality structure was not functioning well.
However, the agency problem is not the only possible problem formula-
tion in the context of board composition. For example, if the organization
operates in a highly uncertain and dynamic environment where many im-
portant decisions must be made rapidly, unity of command may provide
benefits over separation of powers. If all decisions contemplated by top man-
agement must be vetted and approved by an independent board of directors
chaired by a person whose primary occupation is elsewhere and who spends
roughly one day a month attending to director duties, efficient decision-​
making may be significantly hampered. In such contexts, CEO duality struc-
ture may offer the comparatively efficient governance alternative.
Considering the possibility that those adopting CEO duality may
be addressing a different main problem than those adopting CEO
 Introduction  15

independence, we do not find it surprising that researchers have found “little

or no evidence” (Larcker and Tayan 2015, 134) of the performance impact
of CEO duality versus independence. Because CEO duality and CEO inde-
pendence can both be reasoned decisions given the specifics of the situa-
tion, comparing firms with CEO duality to those with CEO independence
is ultimately an “apples-​to-​oranges” comparison. To make the comparison
“apples-​to-​apples,” one must analyze a sample of companies that define the
main problem in a similar way.1
In defining the main problem, it is important to keep in mind that the
issue at hand is one of governance, and as such, foundational to the organ-
ization. Foundational decisions must be made on foundational grounds.
For example, suppose the argument for CEO duality is made on the grounds
that because the founding CEO is a skilled individual with unwavering in-
tegrity, the concern for an agency problem is minimal. We counsel against
such an ad hoc justification. The decision must be made in a forward-​looking
manner, not merely in light of the organization’s current situation. It is my-
opic to base a governance decision on a transient issue, such as the founding
CEO’s characteristics and reputation. It is only a matter of time when the
founding CEO leaves, and a successor will be appointed. It would be my-
opic to assume that the successor, or the successor’s successor, will present no
agency concerns. Governance decisions must be made with conscious fore-
sight to avoid myopia.

Third Premise: Viability Builds on Collaborative

Value Creation

The processes by which value is created are distinct from the processes by
which it is captured or appropriated. In colloquial terms, the distinction is
between “baking the pie” and “dividing the pie.” This book is about “baking
the pie,” that is, ensuring that the organization creates value by securing
the cooperation of its most important constituencies. Accordingly, we ask

1 The ubiquitous, vexing dilemma that governance researchers face is that causal effects are notori-
ously difficult to uncover with empirical analysis. The researcher cannot conduct experiments where
governance choices are randomly assigned to organizations in the sample the same way individuals
are randomly divided into treatment and control groups in medical experiments. Just how trust-
worthy would a medical experiment be if individuals could choose whether they receive the treat-
ment or the placebo? In the context of governance, alternative “treatments” are always a matter of
choice. But this is not to be viewed as a shortcoming, that they are matters of choice is the whole point!
16  Fundamentals of Efficient Organization

questions such as the following: How do contracting parties enter into

relationships? When is the relationship credible in the eyes of the contracting
parties? What kinds of safeguards are implemented to avoid contractual
hazards? How are conflicts resolved in a way that does not jeopardize the re-
lationship? These questions are aimed at evaluating the extent to which col-
laborative, value-​creating relationships are organized efficiently. When the
focus shifts from the entities that organize to the processes of organizing, the
relationship becomes the central unit of analysis.2
In some contexts, the transacting parties seek profits; in others they do
not. We propose that insofar as efficient organization of relationships is con-
cerned, profit-​seeking is an ancillary concern. Whether avoiding waste is de-
sirable or not should not depend on whether any of the transacting parties
are interested in producing a surplus. To be sure, the desire to organize effi-
ciently may be driven by the profit motive, but because value capture is about
how much value a specific party is able to appropriate, it misguidedly shifts
attention from the relationship and cooperation to the outcomes for just one
of the participants to the exchange.
As much as possible, we want to keep the focus on securing and
maintaining exchange relationships, because it paves the way toward un-
derstanding the viability of organizations. This objective is consistent with
many economic theories of organization that emphasize survival, not profit-
ability. A case in point, two central architects of agency theory, Eugene Fama
and Michael Jensen (1983a, 301 [emphasis added]) noted that their central
concern was “with the survival of organizations in which important decision
agents do not bear a substantial share of the wealth effects of their decisions.”
This quote is instructive not only because it establishes survival as the central
concern but also because it points to the separation of decision-​making and
risk-​bearing as the central governance challenge. Whether the organization,
or those involved in it, seeks profits or other types of private benefits is a sec-
ondary concern.
Some organizational forms are more viable than others. Consider as an
example a limited liability company where the same individuals populate

2 Organization economists often take the transaction as the unit of analysis (Williamson 1985).
In our view, this is the economist’s way of expressing that the focus is on relationships; the links of
a dyad, not its nodes. Commons (1934, 4) described the objective succinctly: “[E]‌conomic organi-
zation [ . . . ] has the purpose of harmonizing relations between parties who are otherwise in actual
or potential conflict.” Along the same lines, Williamson (1996, 365) noted that “the main purpose
(which is not to say the only purpose) of economic organization is to infuse integrity into contractual
relations.”
 Introduction  17

both the top management team and the board of directors but at the same
time, own a very small percentage of shares. In other words, the individuals
who make the most important decisions also exercise oversight over their
own decisions, without bearing significant risk. Who is going to invest in
such a company? To the extent the firm is dependent on raising equity to fi-
nance its operations and investments, separating risk-​bearing from decision-​
making without simultaneously separating decision-​making from oversight
jeopardizes viability. To make the organization viable, one must either pop-
ulate the board primarily with independent board members or require those
populating both the top management team and the board of directors to in-
vest significantly in equity. The appointment of independent board members
is standard practice in all publicly traded corporations. Top management and
the board investing in equity can, in turn, be found in most small startups
where the same individuals run the business, exercise oversight, and bear in-
vestment risk.
In economic terminology, this book and its key messages reside on the
value creation or preappropriation side. Profit is a postappropriation measure
of performance, because it is what is left over after all contractual obligations
have been met and the associated revenue appropriated. It would be awk-
ward to think of organization design and governance in such “leftover terms.”
Instead, we suggest that the attention be directed, first and foremost, to how
organizations create value. Avoiding waste is readily consistent with this
objective.3

Fourth Premise: Stakeholdership as Reciprocity and Risk

It is common to think of shareholders as the owners of the firm. Similarly,

shareholders are often considered the only relevant principal and the
organization’s sole stakeholder because they are the only constituency with
rights to the organization’s residual.
We find thinking of shareholders as the only constituency with a residual
interest both unnecessary and misleading. It is unnecessary because al-
though shareholders may be the most salient constituency with a residual

3 Think of all the economic value that is appropriated from revenue before it is turned into
profit: employee salaries, pension payments, suppliers’ invoices, insurance, depreciation, and taxes.
After all these contractually and legally binding appropriations have been made, the economic sur-
plus emerges as a postappropriation measure of performance.
18  Fundamentals of Efficient Organization

interest, there may be others. It is further misleading to assume that the

agency problem is limited to organizations that have shareholders. In order
for an agency problem to occur, all that is required is a principal-​agent rela-
tionship of some kind. In this vein, Fama and Jensen (1983a, 314) noted that
the misapplication of CEO duality structure has consequences that reach
well beyond shareholders; it threatens the very survival of the organization.
Definitions of stakeholder are as numerous as they are diverse. In this
book, we start at the Cambridge Dictionary definition, which covers all the
aspects relevant to the efficiency argument. Specifically, a stakeholder is “a
person such as an employee, customer, or citizen who (1) is involved with the
organization, (2) has responsibilities towards it, and (3) [has] an interest in
its success.”4 This definition effectively distinguishes stakeholders from other
constituencies.
To make the distinction between constituency and stakeholder clear, con-
sider the following relationships:

1. A is B’s supplier;
2. A is unilaterally dependent on B;
3. A has the ability to influence B’s actions;
4. A is able to extract value from B;
5. A is B’s beneficiary; and
6. A creates value for B.

In all these six cases, the three conditions of the definition of a stakeholder
relationship may or may not be satisfied. For example, buyers may develop
stakeholder relationships with some of their suppliers over time, but merely
having a buyer-​supplier relationship is not sufficient to transform a constitu-
ency into a stakeholder. Similarly, a constituency that creates value for the or-
ganization may have a simple transactional relationship with no discernible
stakeholder characteristics that warrant the designer’s attention. Finally, even
though a beneficiary may seem like an obvious stakeholder, beneficiaries
tend not to have significant responsibilities toward the organization whose
benefits they enjoy, or at least, these responsibilities pale in comparison with
the benefits. Children being the beneficiaries of trusts set up by their parents
is a good example.

4 www.dic​tion​ary.cambri​dge.org/​dic​tion​ary/​engl​ish/​stak​ehol​der.
 Introduction  19

The Cambridge definition offers a useful starting point. However, we pro-

pose the definition should incorporate reciprocity. Specifically, we should
not ask whether A is B’s stakeholder but, rather, whether A and B are one
another’s stakeholders. Furthermore, because stakeholder relationships can
involve both individuals and organizations, the definition should incor-
porate a more general term, such as party. Finally, since parties who have
responsibilities toward one another are by definition involved with one an-
other, the notion of being involved with one another can be eliminated.
In summary, we modify the Cambridge definition as follows: Two parties
are one another’s stakeholders if they have responsibilities toward one an-
other and are interested in one another’s success. We submit that this def-
inition avoids the problem of overpermissiveness that often leads to the
conclusion that every constituency is a stakeholder and that everyone is
equally important. But if everyone is important, then no one is, and as far as
stakeholder issues are concerned, all our work would still be ahead of us.
Prioritization of stakeholders is challenging but essential. We discuss this
topic further in detail in ­chapter 4 where we examine how communities or-
ganized the delivery of COVID-​19 vaccinations. It would have been ethi-
cally untenable to base vaccine delivery on the principle that everyone had
an equal right to receive the vaccine. Instead, those with more at stake would
have to be prioritized.
We propose that the inability to prioritize stakeholders would have similar
consequences as the inability to prioritize recipients of COVID-​19 vaccine
would have. Specifically, a designer’s failure to incorporate the irrefutable
fact that some constituencies have more at stake than others will jeopardize
the organization’s credibility in the eyes of those who have comparatively
more at stake. If those with more at stake are critical in value creation (which
often is the case), the inability or unwillingness to prioritize stakeholders im-
minently threatens organizational viability.

Fifth Premise: Focus on Private Ordering

We distinguish between two kinds of context. One refers to the legal and insti-
tutional context in which the organization operates. Consider two organiza-
tions: DuPont de Nemours, Inc., a US-​based publicly traded limited liability
company, incorporated in the state of Delaware, and traded on the New York
Stock Exchange (NYSE), and the Mondragón Corporación Cooperativa, a
20  Fundamentals of Efficient Organization

massive cooperative organization based in the Basque region of Spain. The two
organizations face drastically different legal and institutional environments,
and consequently, their designers must respond to the demands of their re-
spective environments. In short, the institutional context matters.
Focusing on the institutional context directs the designer’s attention to
the “rules of the game.” For example, the three central sets of rules DuPont
must incorporate into its governance decisions are the applicable US federal
laws, the applicable laws of the state of Delaware, and the NYSE Corporate
Governance Guide. The institutional context is something to which the or-
ganization must adapt as a matter of compliance.
Similarly, all companies incorporated in Spain are subject to the same
legal requirements. Perhaps largely for this reason, the legal requirements are
often general, leaving much to the designer’s discretion. For example, Title
IV, chapter I, article 210 of the Spanish Corporate Enterprises Act stipulates
that “company administration may be entrusted to a sole director, several
directors acting jointly or severally, or a board of directors.” Furthermore,
title IV, chapter VI, article 242 requires that if company administration is
entrusted to a board of directors, “the board shall have no [fewer] than three
members.”
As the Spanish Corporate Enterprises Act effectively illustrates, the insti-
tutional context is a source of very general guidelines. For example, consider
the question whether to delegate administration to a board of directors and
the subsequent decision of how many board members to have. Both are es-
sentially matters of discretion and private choice, or as the economist would
put it, matters of private ordering (Williamson 1996). This book is about pri-
vate ordering and, accordingly, the ways in which the organizational context
matters.
At the founding of the organization, the relevant private-​ordering choices
are commonly embedded in the organization’s founding documents: arti-
cles of incorporation, corporate bylaws, and shareholders’ agreements. These
documents are best thought of as contracts in which the designer seeks to ad-
dress relevant local and idiosyncratic design questions. Indeed, courts often
treat these documents as binding contracts: “[T]‌he governing documents of
the corporation—​the charter and bylaws—​operate and bind both managers
and shareholders as if they had negotiated their terms and signed them, like a
common law contract” (Fisch 2018, 377).
It is useful to make the distinction between the institutional and the con-
tractual pillars of efficient organizing. Efficient organization arises from
 Introduction  21

compliance with the institutional rules on the one hand and local contractual
adaptation on the other. Although both pillars are relevant and require the
designer’s attention, we focus in this book primarily on the contractual pillar,
that is, the choices that designers make within the boundary conditions set
by the institutional pillar. Insofar as the institutional pillar is concerned, the
only prescription we give to the designer in this book is the following: Ensure
compliance with the requirements and the boundary conditions set by the
institutional environment. Accordingly, questions such as whether a US-​
based corporation should incorporate in the state of Delaware or another
state (see Bebchuk and Cohen 2003) are beyond the scope of this book.
The premise in this book, and in many organization design and govern-
ance conversations more generally, is that governance challenges must be
addressed by individual organizations in their own, idiosyncratic contexts.
Focus on private ordering suggests that governance is not so much a legis-
lative and policy issue as it is a private, organization-​specific matter. This
framing implies that the central focus should be on how the organization
seeks to address the local governance problems by making informed choices,
which appropriately shifts attention to and emphasizes the responsibilities of
the designer.
To further justify why private ordering merits attention, let us examine
conflict resolution in interorganizational exchange. Why not just rely on the
institutional pillar and resolve conflicts in the court of law? Although the an-
swer may be obvious, there are some nuances that merit attention.
The courts have an indispensable role in supporting conflict resolution in
interorganizational exchange. However, relying on them has three general
disadvantages: Litigation is time-​consuming, expensive, and has potentially
a nonexpert presiding over the dispute. The first two should be self-​evident.
As to nonexperts presiding, the assumption that in the case of, say, a buyer-​
supplier conflict, the court system will efficiently appoint a judge with the
requisite substantive expertise in interfirm business relationships and supply
chain structures is wishful thinking. Unless the contracting parties get lucky,
they may find themselves in the courtroom with a judge who is not an ex-
pert on the substance of the dispute, only the laws that apply to the dispute.
Indeed, both legal and economic scholars have suggested that courts some-
times rely on “oversimplified economics” and “unfounded or disproven
assumptions” (Leslie 2014, 939), even to the point of suggesting that legal
experts at the highest levels of the judiciary may at times be “deeply confused”
(Williamson 2002a, 9) about the economics of organization. This critique is
22  Fundamentals of Efficient Organization

not to blame judges but simply to acknowledge that they are experts on law,
not economics or business practice.
Instead of referring to the courts (the institutional pillar) for dispute
resolution, relying on private ordering (the contractual pillar) may offer
the comparatively efficient option. For instance, the contracting parties
could stipulate in the contract the following dispute resolution mech-
anism: If a conflict escalates to a point where negotiation attempts fail and
a third party is required, the contracting parties commit to mediation or ar-
bitration instead of litigation. In economic terms, instead of assuming that
the institutional environment will provide comparatively efficient conflict
resolution, the designer chooses to build in various internal quasijudicial
functions (Williamson 1975, 30) by which conflicts can be addressed more
efficiently: faster, cheaper, and with substantive experts presiding over the
dispute.
The designer must also understand that in some cases invoking the institu-
tional pillar will be not only comparatively inefficient but also infeasible. For
example, if one division of an industrial firm buys components from another
division that belongs to the same parent corporation, the two cannot settle
their disputes in court—​the corporation would effectively be suing itself.
The private-​ordering approach places not the law or the institutional en-
vironment but various contracts at center stage. Here, the word contract
should be understood as a broad term that encompasses all aspects of private
ordering—​both implicit and explicit—​by which two or more parties agree
on duties and responsibilities. Indeed, for the purposes of our argument, we
find resonance in the sentiment that “the notion of contract is so broad as to
include virtually all voluntary social arrangements” (Blair and Stout 1999,
254). This broad definition also underscores the idea that lawyers may not
always be the preferred experts to address contractual issues. In this vein, or-
ganization scholars Nicholas Argyres and Kyle Mayer (2007) suggested that
whereas lawyers may possess superior capabilities to address matters of over-
sight and control, managers may be better equipped to handle issues such as
resource allocation and contingency planning.

Overview of the Chapters

This book comprises eight interrelated chapters organized into three

parts. Part I, “Fundamentals of Efficient Organization,” consists of the
 Introduction  23

“Introduction” (­chapter 1) and “The Efficiency Lens” (­chapter 2). In these two
chapters, we introduce the key terminology and the analytical framework,
the Efficiency Lens, applied throughout the book. The extensive “Glossary of
Terms” at the end of the book serves as a companion to Part I. The main pur-
pose of Part I is to establish the relevance of efficient organization.
Part II, “Governance within and across Organizations,” contains three
substantive chapters: “Contracting within and across Organizations”
(­chapter 3), “Stakeholder Analysis” (­chapter 4), and “Nonprofit and Public
Organizations” (­chapter 5). In these chapters, we discuss some of the cen-
tral themes associated with a governance-​based approach to organization.
Because contracts, and contracting more generally, are central to govern-
ance, the focus is specifically on contractual relationships.
In Part III, “Governance and the Organizational Life Cycle,” we ex-
amine the governance challenges that organizations face at different stages
of their lifecycles: “The Startup Organization” (­chapter 6), “The Expanding
Organization” (­ chapter 7), and “The Institutionalized Organization”
(­chapter 8).
Chapters 3 through 8 form a “matrix” in that the topics discussed in
­chapters 6 through 8 cut across ­chapters 3 through 5 (see fig. 1.1). For ex-
ample, stakeholder issues are relevant in all organizations, no matter at what
stage in their lifecycle they may be. At the same time, the stakeholder issues
that a small industrial startup faces are different from those faced by large,
publicly-​traded corporations. In the following, we briefly introduce the con-
tent of c­ hapters 2 through 8.

Chapter 2: The Efficiency Lens

In ­chapter 2, we introduce a general framework and analytical tool we apply

throughout the book. The Efficiency Lens consists of three elements—​
management, oversight, and risk—​ and their interrelationships and dy-
namics. We find the word lens descriptively accurate, because adopting the
efficiency view indeed constitutes a way of looking at the organization. The
main purpose of the Efficiency Lens is to bring various sources of organi-
zational waste into focus by analyzing how the work gets organized, who
exercises oversight, and who bears risk. The overarching objective is to en-
sure that the organization establishes and maintains credibility in the eyes
of those who create value for the organization. Not being able to establish
24  Fundamentals of Efficient Organization

Figure 1.1  The structure of the book

credibility has numerous inefficiency consequences, such as excessive em-

ployee turnover, higher cost of capital, underinvestment in R&D, and various
contractual disputes, to name just a few common examples.
The Efficiency Lens also helps us give useful definitions and ways of
describing both the content and the dynamics of various organizational
forms. For example, we can use the Efficiency Lens to define a nonprofit
 Introduction  25

organization as the organizational form with no residual claimant, and to

describe the modern corporation as an organization where management,
oversight, and risk are materially separated from one another. We can also
use the Efficiency Lens to describe governance changes that take place in
a startup firm as it grows and heads toward an initial public offering. The
Efficiency Lens constitutes the primary interpretative tool throughout all the
chapters.

Chapter 3: Contracting within and across Organizations

Contracts and contracting are at the heart of private ordering. In

­chapter 3, we turn our attention to contracting and the associated costs of
transacting. Instead of treating intra-​and interorganizational transacting
separately, we combine both into one chapter; the designer’s choice is
often to organize an exchange relationship either within the organization
or across organizations. For example, in deciding how to safeguard intel-
lectual property, a high-​technology firm may consider either contracting
with an external law firm or hiring an in-​house counsel. It is useful to con-
trast the alternatives to determine which is more efficient. In ­chapter 3, the
objective of efficient organizing becomes particularly salient: Alternative
forms of contracting can often be explicitly compared in terms of their
total cost.
Jointly discussing contracting within and across organizations makes
sense also because many intraorganizational contracts have characteristics
similar to interorganizational contracts. In their insightful examination of
internal transactions, management scholar Robert Eccles and sociologist
Harrison White (1988) noted that markets often exhibit authority properties
found within firms and that multidivisional firms may contain pricing
mechanisms found in markets.

Chapter 4: Stakeholder Analysis

We have found many stakeholder conversations to lack analytical rigor.

Consequently, in our exposition of stakeholder issues in ­chapter 4, we focus
specifically on stakeholder analysis: What are the criteria by which a con-
stituency becomes considered a stakeholder? How should stakeholders be
26  Fundamentals of Efficient Organization

prioritized? How should the fact that a constituency is considered a stake-

holder be incorporated into governance decisions?
Our approach to stakeholder analysis is based on risk in general and re-
sidual risk in particular. Accordingly, we define as stakeholders those
constituencies who, by virtue of becoming involved with the organization,
have voluntarily put something at stake. The shareholder of the limited lia-
bility company is the obvious stakeholder, but it is often an oversimplification
to consider the shareholder the only stakeholder. For example, many con-
temporary employment relationships allocate more risk to employees, offer
lower job security, and incorporate variable compensation as a significant de-
terminant of wages (e.g., Rousseau and Shperling 2003, 2004). Consequently,
at least some employees, or employee groups, may have a legitimate residual
interest in the organization, just like shareholders. Organizations may also
develop stakeholder relationships with suppliers and customers that make
relation-​specific investments.
In ­chapter 4, we propose that a systematic stakeholder analysis be
conducted to rigorously examine all the relationships the organization has
with its constituencies. Further, promoting a constituency to stakeholder
status must have implications for how the relationship is governed, lest
stakeholder management regress to mere rhetoric. Managing stakeholder
relationships calls for the implementation of safeguards that would be redun-
dant (and wasteful) in transactional, arm’s-​length relationships.

Chapter 5: Nonprofit and Public Organizations

Even though all chapters in this book apply both to for-​profit and nonprofit
contexts, we dedicate one entire chapter to governance issues outside the
conventional corporate context. In ­chapter 5, we examine the implications
of an organization not having a residual claimant who expects a return on
investment. Or perhaps more accurately, what are the implications of not
having any other residual claimants than the organization itself?
Note that the distinction between for-​profits and nonprofits is not about
whether the organization creates an economic surplus. In reality, both for-​
profits and nonprofits make a surplus and have net worth. A central gov-
ernance question in nonprofits is how a potential surplus is governed in the
absence of a residual claimant. Ultimately, managing the surplus is just as
relevant to nonprofits as it is to for-​profits.
 Introduction  27

In ­chapter 5, we also examine the public/​ private distinction. What

does it mean for an organization to be public, and what are the govern-
ance implications? Because most public organizations are more accurately
described as public-​private partnerships, the question is less about what
distinguishes a public organization from a private one and more on what
kinds of roles public and private actors can assume in these partnerships.
Chapter 5 also offers an important challenge to the analysis of net gains and
efficient contracting. This analysis becomes salient as we examine contexts in
which the relationships between the organization and its constituencies are
not essentially contractual, and where participation in the organization is not
necessarily voluntary. In contexts such as psychiatric care, primary educa-
tion, and incarceration, the designer’s primary attention must be to ensure
organizational integrity, making efficiency, at best, a secondary concern. The
discussions and the examples in ­chapter 5 provide an important boundary
condition for both efficiency thinking and the contractual approach to
organizations.

Chapter 6: The Startup Organization

Sociologist Arthur Stinchcombe (1965) famously argued that organizations

“are frozen at birth” in the sense that initial conditions have long-​lasting
consequences. In ­chapter 6, we examine the design choices made at the
founding of the organization: What are the central founding decisions that
establish the initial organization design and governance conditions? In this
chapter, the importance of understanding private ordering becomes pro-
nounced. The assumption that laws, regulations, and other institutions offer
adequate guidance and protection for the designer of a startup is misguided.
In ­chapter 6, we discuss, among other topics, the function of the shareholders’
agreement in startup companies.
The general message in ­chapter 6 evolves around the importance of consid-
ering organization design and governance as ex ante problems: How does the
designer make decisions regarding organization design and governance in a
forward-​looking manner? Which potential future problems can be folded,
either wholly or partially, into an ex ante contract? Which problems, in turn,
are not ex ante contractible, and how should the designer think of them at
the moment of the organization’s founding? What kinds of safeguards can be
implemented in the case of noncontractible issues?
28  Fundamentals of Efficient Organization

Chapter 7: The Expanding Organization

As the organization grows, the scale and the scope of its activities expand,
the need for oversight becomes more elaborate, and many constituencies de-
velop stakeholder relationships with the organization, effectively becoming
risk-​bearers. In this chapter, we propose that the governance challenges
expanding organizations face are ultimately caused not by the expansion it-
self but the fact that management, oversight, and risk—​the three elements of
the Efficiency Lens (­chapter 2)—​gradually separate from one another. For
example, although those who make the most important decisions tend also
to populate the board of directors in a startup organization, an expanding
firm must start appointing independent members to its board of directors,
effectively separating management from oversight. Similarly, as the organiza-
tion expands, not everyone who bears risk can be afforded an oversight role,
effectively separating oversight from risk. Not only the expansion of man-
agement, oversight, and risk but also their separation requires the designer’s
attention.
As management, oversight, and risk become separated, the expanding
organization may also become vulnerable to takeovers. In c­hapter 7, we
examine the ways in which expanding organizations can seek protection
against unwanted takeovers. We also investigate the conditions under which
various antitakeover provisions are recommended and whom they ulti-
mately protect. Importantly, whether a takeover is wanted or unwanted is
not straightforward; it may be welcomed by some constituencies but strongly
resisted by others.

Chapter 8: The Institutionalized Organization

Sociologist and legal scholar Philip Selznick (1957, 16–​17) noted that the
institutionalization of an activity, structure, or routine involves “infusion of
value beyond the technical requirements of the task at hand.” Such “infu-
sion of value” may have both desirable and undesirable consequences. In this
chapter, we examine the undesirable consequences of institutionalization.
To the extent that organization design and governance choices become in-
stitutionalized and are no longer being relentlessly questioned, analyzed, and
modified, institutionalization may lead to various efficiency distortions.
 Introduction  29

In ­chapter 8, we examine the reasons why governance choices tend to

become taken for granted or otherwise persist over time without being
questioned or challenged. We focus in particular on the persistence of the
internal organization, that is, the principles and practices that become insti-
tutionalized within organizations. In our discussion of institutionalization,
we further focus primarily on the inadvertent sources of persistence. For
example, once implemented, the division of tasks by function—​functional
structure—​tends to persist over time. One reason for persistence is that func-
tional managers start to protect the interests of their own functions and
actively resist change. Although we acknowledge the role of active resist-
ance, we direct the designer’s attention to the more elusive source of persist-
ence: the inability to identify comparatively efficient alternatives because the
current structure has become taken for granted and “infused with value” be-
yond the purpose it was intended to serve.

Epilogue

We end the book by a brief reprise of the book’s central messages, along
with an important reminder of the central boundary conditions in the ap-
plicability of comparative efficiency analysis to governance decisions. Even
though we promote efficiency thinking throughout the chapters, we find it
important to remind all designers of the boundaries of its applicability.
 2
The Efficiency Lens

In thinking of how to approach the subject matter of organization design and

governance, we are faced with a fundamental challenge: How do we do jus-
tice to the vast heterogeneity of organizations? How can we write about or-
ganizations and analyze them without arriving at the predictable conclusion
that every organization is unique and that there simply is no one right way to
organize? All one has to do to conclude that every organization is unique is to
glance at a chart that depicts the organization’s structure.
At the same time, there are a number of general characteristics featured
in all organizations. Insofar as governance is concerned, we suggest that
there are three: management, oversight, and risk (see fig. 2.1). The framework
introduced in this chapter—​the Efficiency Lens—​combines these three,
their dynamics, and their interrelationships. The Efficiency Lens is based on
roughly one hundred years of organizational research and at least as many
years of practical experience and recorded company history. The Efficiency
Lens is not based on any specific theory of management or of economics but,
instead, incorporates ideas from several theories and ways of approaching
organizations from an efficiency perspective.
The main premise in the Efficiency Lens is that in seeking efficient
organizing, the designer should adopt three objectives: (1) organizing and
coordinating individual activities in the organization efficiently, (2) ensuring
that individual activities align with the interests of the entire organization,
and (3) safeguarding the interests of those who have something at stake in
the organization.
Let us briefly explore the metaphor of a lens by pointing to three distinct
events that occur when we observe an object through a lens. All three reveal
something essential about the utility of the Efficiency Lens.
The first is that a lens can bring something into focus so that it can be seen
more clearly. As the name suggests, the Efficiency Lens brings economic ef-
ficiency into focus. In this sense, we can think of the Efficiency Lens as if it
were a set of eyeglasses.

Efficient Organization. Mikko Ketokivi and Joseph T. Mahoney, Oxford University Press. © Oxford University Press 2023.
DOI: 10.1093/​oso/​9780197610282.003.0002
 The Efficiency Lens  31

Figure 2.1  The three elements of the Efficiency Lens

The second is that a lens can magnify, thus exposing the detailed character-
istics of the object of interest. Magnifying is useful as you read the fine print
of an insurance policy, for example. In contrast with eyeglasses, when a lens
magnifies, it does not bring something into focus as much as it emphasizes
something with the purpose of enabling a careful examination of its detail. In
this sense, we can think of the Efficiency Lens as if it were a magnifying glass.
Finally, a lens is a useful metaphor because it emphasizes the active role
and expertise of the observer. Both a microbiologist and a layperson can view
the structure of a coronavirus through the set of lenses of a microscope. To
the layperson, the image may appear novel and fascinating, but aside from
perhaps learning to appreciate why the word corona (crown) was used to
name the virus, there is not much more that can be learned. In stark contrast,
a microbiologist or a virologist will find the image informative and useful.
Therefore, the utility of a lens is fundamentally dependent on user exper-
tise; it is useless to apply a lens without some understanding of what one is
looking at. In this sense, we can think of the Efficiency Lens as if it were a
microscope.

Building the Efficiency Lens

In constructing the Efficiency Lens, we face a dilemma. On the one hand, we

want to use the established terms management, oversight, and risk; inventing
new concepts would lead to confusion. On the other hand, because our
aim is to provide a general lens applicable to all organizations, we use these
terms in meanings that are both broader and more malleable than they are
in many conventional treatments on organization design and governance.
Consequently, we invite all readers “to reboot their conceptual hard drives.”
32  Fundamentals of Efficient Organization

Rebooting the Conceptual Hard Drive

Let us reexamine some of the taken-​for-​granted meanings of organizational

terminology. This reexamination is particularly important in the context of
organization design and governance, where the terminology has a feature that
is as frustrating as it is fascinating: Over time, the original neutral meanings
of many concepts have acquired conspicuously negative connotations.
What is the first thought that comes to mind as you read the words cen-
tralization, control, hierarchy, authority, and bureaucracy? All five were orig-
inally introduced as neutral, descriptive terms, but for some reason, all have
been permeated with negative associations. These negative associations
have become so widespread that they have made their way into dictionaries.
A quick glance at a few online dictionaries reveals that although some of
the definitions of bureaucracy are neutral (as in “following predetermined
rules”), others are evaluative (as in “complicated rules”), even pejorative (as
in “excessive red tape”). We counsel the reader to heed the intended neutral
definitions.
Why this infusion with negative connotations occurs is a mystery to us.
Perhaps its origins are in a recurring, enduring problem we encounter in
the popular press on organizations—​an obsession with novelty. The rhet-
oric is familiar enough: Established ways of organizing are outdated as they
no longer meet the requirements of contemporary organizations and busi-
ness environments. Centralization, hierarchy, and bureaucracy belong to the
dustbin of history; today’s organizations are all about replacing hierarchies
with flexible lateral structures, self-​management, and informal networks.
Alas, a rigorous, dispassionate analysis of organizations reveals that both
hierarchies (e.g., Leavitt 2005) and bureaucracy (e.g., Adler and Borys
1996) are essential to all organizations, and that there is a time and a place for
centralization (e.g., Alonso, Dessein, and Matouschek 2008). The question is,
“When, where, and how?”
As an example, consider centralization and, specifically, the geographic
centralization of activities. Some technology firms have concluded that
centralizing research and development (R&D) into one large, centralized
R&D center is preferred; others decentralize their R&D activities to smaller,
specialized, and geographically dispersed R&D units. There is no general
prescription or best practice, instead, the degree of centralization constitutes
a design decision that, just like all design decisions, must be linked to the
main problem the designer seeks to address. In their empirical examination
 The Efficiency Lens  33

of organizing R&D, organization scholars Nicholas Argyres and Brian

Silverman (2004, 929, 930) found that centralized R&D tends to “generate
innovations that have larger and broader impact on subsequent technolog-
ical evolution,” whereas decentralized R&D “encourages more proximate
(‘capabilities-​deepening’) research.” Therefore, there is nothing inherently
superior about decentralization (or centralization), everything depends on
what the designer is trying to achieve. The same logic applies to all other
concepts.
As another example, let us revisit management, one of the three central
concepts of the Efficiency Lens. Here, we find the writings of Chester Barnard
particularly insightful. Barnard was not an academic but an executive who
authored a number of prescient and influential works in the early twentieth
century, most notably his 1938 book The Functions of the Executive (Barnard
1938). We invoke Barnard’s work here not only because of his unique and
highly influential impact on both the practice of management and research
on organizations but also because his approach helps us convey the broader
meaning of management.
Above all else, Barnard emphasized that the main function of the
organization’s top leadership—​“the executive”—​was to secure the contin-
uing, voluntary cooperation of all those who create value in the organization.
With this objective in mind, Barnard’s approach to management authority
was based not on top-​down command but rather on bottom-​up con-
sent: “Management’s authority, Barnard realized, rested on its ability to per-
suade, rather than to command” (Mahoney 2002, 164). Moreover, Barnard
explicitly warned about the dangers of thinking of organizational authority
in terms of directing the work of others: “You put a [person] in charge of
an organization and your worst difficulty is that [he or she] thinks [he or
she] has to tell everybody what to do; and that’s almost fatal if it’s carried far
enough” (Wolf 1973, 30).
To be clear, the purpose of invoking Barnard’s ideas is not to promote
persuasion over command and control; this would amount to uncritically
replacing one orthodoxy with another. We refer to Barnard’s work because
he offers a valuable contrast by presenting the broadest possible approach to
what management can be. Specifically, coordination of work and exercise of
authority can be based on command and control, but they can also be based
on persuasion.
Which should it be, then, persuasion or command? Just as in the case of
centralization, how to organize management is a design choice. Persuasion
34  Fundamentals of Efficient Organization

may sound intuitively and emotionally more appealing than command, but
organization design is not about what sounds good or seems fashionable; it
is about what a rigorous analysis of alternatives reveals to be comparatively
efficient. Returning to the taxi ride example in ­chapter 1, owning (or not
owning) a car because it feels intuitively or emotionally appealing misses the
point of efficient organizing entirely.
The purpose of “rebooting our conceptual hard drives” is to purge our
minds of misleading and negative connotations that are unfortunately
common in our thinking about organizations. To the end of avoiding neg-
ative associations, we invite the reader not only to think of all the concepts
that follow in neutral terms but also all contrasts in terms of potentially vi-
able alternatives. Just as there are no “positive” and “negative” concepts in
this book, we do not compare “good” and “bad” governance options to one
another. The point of a comparative efficiency analysis is to compare feasible
alternatives to one another.
Negative connotations are dangerous precisely because they may lead the
designer to dismiss an option that may in fact offer the comparatively efficient
governance alternative. Students in our business school seminars are caught
by surprise when they find out in the classroom discussion on the technology
giant Cisco Systems that at the foundation of its organization resides a decid-
edly old-​fashioned, functional structure (Gulati 2010). At the same time, a
rigorous analysis reveals that as old-​fashioned as it may sound, the functional
organization makes perfect sense not only for Cisco but also to many others.

Management

The job of automobile assembly plants is to perform the final assembly of

automobiles; the job of a law firm specializing in intellectual property is to
enable its clients to manage, protect, and trade and license their patents; the
job of a cardiovascular care unit of a hospital is to perform heart surgeries. All
organizations have a primary job that needs to get done, and how this work
is organized and coordinated is an essential part of governance. The manage-
ment aspect of governance directs attention to how value-​creating activities
are divided and coordinated within and across organizations. Furthermore,
the Efficiency Lens directs attention specifically to ways in which man-
agement should be organized to avoid waste: by avoiding excessive layers,
 The Efficiency Lens  35

redundancies, excessively complex cross-​unit linkages, poorly defined ac-

countability, and so on (e.g., Goold and Campbell 2002b).
In the Efficiency Lens, management does not refer to any specific indi-
vidual, or individuals, and is not to be interpreted as synonymous with man-
ager or managers. Instead, management refers more generally to managerial
work that is relevant in all organizations: planning and scheduling, task and re-
source allocation, coordination, and the like. Whether this work is performed
by individuals whose role in the formal structure is that of a manager is an
altogether different issue, and a matter of choice. Some organizations choose
to employ managers to perform managerial work; others choose not to assign
managerial work to a separate class of employees called managers. In addi-
tion, in many organizations routinized managerial work such as planning and
scheduling may also be automated to a significant degree, that is, embedded
in various enterprise resource planning (ERP) systems.
If the idea of organizing without managers sounds utopian, we might
note that professional managers have not existed all that long in the history
of organizations. Indeed, the formal position of a manager in organizations
did not emerge until the 1800s in the context of the industrial revolution.
At the same time, management historians Wolfgang Pindur, Sandra Rogers,
and Pan Suk Kim (1995, 59) noted that the use of various management
techniques, such as systematic written records of business transactions, can
be traced back to as early as 3000 BCE.
The Efficiency Lens does not take the formal position of manager as a
given; instead, it is presented as one alternative of organizing management.
Given the increasing recent interest in self-​managing organizations, the idea
of not having formal managerial positions in the organization should not
seem too strange (Lee and Edmondson 2017). Simply put, yes, managers do
perform managerial work, but managerial work need not be performed by a
person whose business card has the word manager on it.1

1 Lee and Edmondson (2017, 46) make the point succinctly in their discussion of self-​managing
organizations (SMOs): “Eliminating ‘managers’ as a formal role does not mean self-​managing or-
ganizations are devoid of managerial work. The work of monitoring progress towards organizational
goals, allocating resources or projects, designing tasks and organizational structures, and providing
feedback to individuals remain vital to effectiveness in SMOs. However, in SMOs, these authorities
are formally distributed to individuals in a way that is not permanent, unbounded, or vested in hi-
erarchical rank.” Hamel (2011) made similar arguments in his Harvard Business Review article pro-
vocatively entitled “First, Let’s Fire All the Managers.” There is nothing utopian about the idea of
an organization where employees are in charge of managing their own work. However, we recom-
mend that the designers remain relentlessly comparative: The question, “Does this organization need
to hire managers, and if yes, for what purpose?” requires an explicit comparative evaluation of the
alternatives.
36  Fundamentals of Efficient Organization

Of course, many contemporary organizations choose to structure themselves

in a way that makes managers central to management. Insofar as individual
organizational members are concerned, we include in management not only
executives and top management but more broadly everyone involved in the pla-
nning and coordination of value-​adding activities. In a manufacturing firm, this
includes everyone from C-​suite executives and middle managers to production
supervisors and team leaders. In a hospital, managerial positions are held by
hospital administrators and administrative staff as well as those directly pro-
viding care to patients (i.e., doctors, nurses, and other clinical staff). To make
matters simple, let us suggest that organizational members are a part of man-
agement when (1) they have an employment relationship with the organization,
and (2) their job description includes at least some planning and coordination
responsibilities.
The notion of management is neither limited to top management nor about
a person at one level of the organization directing the activities of those at
lower organizational levels in a top-​down manner. At the same time, the defi-
nition does not exclude a top-​down approach either. Sometimes planning and
coordination are most efficiently performed in a centralized manner by the
organization’s top managers. A hierarchical top-​down organization may con-
stitute the comparatively efficient organizing alternative in stable organizational
environments where the organization’s task is well defined and unchanging, and
where innovation is not a central concern (Burns and Stalker 1961).
Are there still stable environments in today’s dynamic world? Of course
there are. For example, the demand for public transportation in a given city
on a given day is reasonably predictable, and this predictability provides
the most important input to those who plan the daily routes and schedules
of buses, trams, and trains. We might also point out that the talk about
increasing uncertainty, hypercompetition, and other destabilizing factors
tends not to survive critical empirical scrutiny. We propose that thinking that
we live in special times is a variation on the obsession-​with-​novelty theme.
We invite those who think the world is more dynamic and unpredictable
today than it was in the past to look at the evidence.2

2 McNamara, Vaaler, and Devers (2003) conducted an empirical analysis using the Compustat
Industry Segment Database from 1978 to 1997. Their conclusion was that “managers today face
markets no more dynamic and opportunities to gain and sustain competitive advantage no more
challenging than in the past.” McNamara et al. (2003, 261).
 The Efficiency Lens  37

The Role of Technology in Management

Humans are not the only ones to perform work and to add value.
Consequently, management is not limited to human beings. Using welding
robots in car assembly and artificial intelligence to automate formalized tasks
in a law firm are salient examples of technology creating value. Furthermore,
just as in the case of employees, we suggest that thinking of technological
assets as entities that have a contractual relationship with the organization
is useful. Specifically, some productive assets are the property of the organi-
zation and appear on its balance sheet. Some of these assets are further em-
ployed both in direct and indirect productive uses. For example, just like
consumers may choose to borrow against the equity of the house in which
they live, an industrial firm may use general-​purpose production equipment
as collateral in debt financing.
It should be equally clear that an organization can make productive use of
assets it does not carry on its balance sheet. For example, it may not make
economic sense for a global car rental company to own the entire fleet of
hundreds of thousands of vehicles it rents out to customers. Instead, various
leasing arrangements likely provide economically more efficient contracting
alternatives. Indeed, one foundational question of governance is choosing
which assets to own and which to borrow: Which components should a man-
ufacturer produce in-​house, and which should it outsource? Does it make
sense for a hotel chain to own the buildings that host its activities? How about
the land on which these buildings are located? Does it make sense for a busi-
ness school to rely on visiting faculty to teach its classes? Should oil companies
carry a fleet of oil tankers on their balance sheet? These are examples of make-​
or-​buy decisions, which constitute an essential part of management.
Fully realizing that using the word asset to refer not only to firms and
technologies but also to humans may sound awkward; nonetheless, we
will do so, because in an economic analysis of organizations, it is useful to
think of assets as all the entities, human or otherwise, involved in value cre-
ation. This said, nothing in our use of the word asset in reference to humans
suggests that they should be treated the same way as technology; we merely
want to point to the advantages of seeing and analyzing the organization as
an entity that has contractual relationships with other firms, value-​creating
technologies, and human beings.
38  Fundamentals of Efficient Organization

Oversight

How should a multinational manufacturing firm conduct analyses regarding

production location decisions? How is the decision to go public by an initial
public offering (IPO) made in a high-​technology startup? How should the
police department structure the compensation of its police officers? What is
the process by which a theater company or a symphony orchestra chooses its
program for the upcoming season?
Some questions that organizations face are more foundational than pla-
nning and coordinating specific activities (i.e., management) in that they
pertain to the general rules, principles, and structures that guide decision
making in specific decision situations. It is useful to treat these more general
questions separately from questions of management. In the Efficiency Lens,
these general questions belong to the domain of oversight.
In many conversations, what we call here oversight is equated with gov-
ernance. However, we consider such a definition of governance unneces-
sarily narrow. We propose that although oversight is essential to governance,
it is not the essence of governance. We think of governance more broadly in
terms of the entire Efficiency Lens: as a triple that encompasses management,
oversight, and risk, and how the three relate to one another over time. We
hope that the following chapters will show the utility of this broader concep-
tualization. Most importantly, we cannot write about efficient organization
without considering all three factors.
To arrive at a sufficiently broad understanding of oversight, we link its task
to three governance issues: (1) the best interest of the organization, (2) re-
sidual income, and (3) residual rights of control.

The Best Interest of the Organization

In 2016, Tesla paid $2.6 billion to acquire SolarCity, a company that sells
solar energy generation systems. The acquisition created controversy among
a number of union pension funds and asset managers whose organizations
held stock in Tesla. In 2017, a number of dissatisfied shareholders filed a law-
suit against Tesla’s board of directors, alleging that the SolarCity acquisition
amounted to a bailout, and that it was Tesla’s CEO Elon Musk, his friends,
and his family who had benefited from the acquisition. At the time of the ac-
quisition, Mr. Musk chaired SolarCity’s board of directors and was its largest
 The Efficiency Lens  39

shareholder. Further, SolarCity’s founders Peter and Lyndon Rive are Mr.
Musk’s first cousins. Tesla’s entire board of directors was named in the law-
suit, but by the time the trial started in 2021, everyone but Mr. Musk had
settled the lawsuit out of court.
We do not know whether the lawsuit has merit and take no sides on the
substance of the dispute. The trial is ongoing, and it is now the task of the
trial judge to decide whether the acquisition was fair to Tesla’s shareholders
or not. Our point is that the controversy surrounding Tesla effectively raises
a foundational governance question: How can the designer ensure that when
the organization delegates decisions to its members, the decisions that are
made serve not the local interests of those who are trusted to make the de-
cision but, more broadly, the best interest of the entire organization? This is
one of the questions that belongs to the domain of oversight. Shareholders
filing a lawsuit against the entire board of directors suggests that at least some
stakeholders think that there is something seriously wrong with oversight.
There are many reasons why local decisions and broader organizational
interests do not align. One possibility is that the decision is made with the
explicit intent of serving interests other than those of the organization; the
decision maker willfully and deliberately ignores or violates broader organ-
izational interests. Such self-​serving behavior lies at the heart of the agency
problem (see c­ hapter 1).3
Although misalignment of the principal’s and the agent’s priorities may be
the direct result of opportunistic behavior by the agent, self-​serving behaviors
have in our view received disproportionate amounts of attention. There are
other, comparatively benign possibilities that merit the designer’s attention.
Specifically, misalignment of local decisions and broader interests can also
stem from the agent being more risk averse than the principal. Consequently,
the agent’s decision may be more conservative than what the principal would
prefer. But reluctance to take risks is not necessarily a manifestation of self-​
serving behavior or opportunism; the agent may simply be more preoccupied
with the survival of the organization than the principal. Particularly in
situations where the principal is equated with shareholders who are free to
sell their shares in the open market, we might go so far as to suggest that the

3 In the context of the limited liability company, we commonly think of the shareholders as the
principal and managers as the agents. Consequently, the best interest of the organization translates to
shareholder wealth. However, this is only one possible formulation of the agency problem.
40  Fundamentals of Efficient Organization

fact that the agent’s preferences and the principal’s preferences regarding the
organization’s survival do not coincide “is not a bug but a feature.”
Other, even less deliberate sources of misalignment link to decision
makers’ limited cognitions. All of us tend to approach decision situations by
incorporating primarily the local context in which we operate and the una-
voidably limited and specialized expertise and experience we have. We do
the best we can, but the limits to cognition, experience, and information—​in
short, bounded rationality—​get the best of us. Bounded rationality can lead
to misalignment, conflict, and disagreements of an altogether benign variety.
One often hears bounded rationality used as a catch-​all phrase to describe
all the possible limitations of human decision makers. However, under-
standing the nuances of bounded rationality has profound implications for
governance. Specifically, many conflicts to which opportunism is ascribed
may be merely honest disagreements due to different individuals having ac-
cess to different kinds of information and different bases of expertise. To be
sure, not only opportunistic behavior but also honest disagreements can be
sources of inefficiency in organizations. Organization economists Armen
Alchian and Susan Woodward (1988, 66) elaborate: “Even when both parties
recognize the genuine goodwill of the other, different but honest perceptions
can lead to disputes that are costly to resolve.” They also counsel against im-
mediately interpreting the use of various safeguards such as monitoring
as attempts to preempt opportunism: “[M]‌ any business arrangements
interpreted as responses to potential ‘dishonest’ opportunism are equally ap-
propriate for avoiding costly disputes between honest, ethical people who
disagree about what event transpired and what adjustment would have been
agreed to initially had the event been anticipated” (Alchian and Woodward,
1988, 66).
The problem with failing to understand how bounded rationality operates
in decision making is that it may lead to excessively negative readings of a
conflict situation and, consequently, an unnecessarily adversarial approach
to its resolution. The purpose of checks and balances in organizations or
formal contracts between buyers and suppliers is not merely, or even pri-
marily, to curb opportunistic behaviors; they are also a manifestation of the
fact that organizations are populated by human beings with severely limited
cognitions.
The more benign distortions may occur even in situations where we are
explicitly encouraged and genuinely motivated to approach decisions from
an organization-​wide perspective. It should not come as a surprise that
 The Efficiency Lens  41

a sales manager “sees” a given organizational problem differently than a

production manager, even when the two are explicitly instructed to ignore
their specific positions and adopt an organization-​wide perspective.4 As
psychologists such as Nobel Laureate Daniel Kahneman (e.g., Kahneman
2011; Kahneman, Slovic, and Tversky 1982) and scores of others have
shown, cognitive biases are endemic not just to organizations but to
humans more generally. As hard as we try to make rational decisions, our
rationality is bounded, and our access to information limited. Insofar as
designing organizations is concerned, it is wishful thinking to assume that
functional managers can simply shed their functional backgrounds and ex-
pertise and approach an issue from an organization-​wide perspective. The
resultant, unavoidable position bias can lead to locally efficient but glob-
ally (organizationally) inefficient decisions and solutions (Ketokivi and
Castañer 2004).
Keeping the problems stemming from limited cognitions separate from
those driven by deliberate opportunistic behaviors is crucial because the
two have different antecedents and remedies. In the case of opportunism,
the problem is intentional and motivational, and in the case of limitations,
inadvertent and cognitive. Governance actions required to curb opportun-
istic behavior differ from the actions required to address problems arising
from cognitive biases. Whereas boards can incentivize top managers to make
risky decisions (this is what stock options are designed to do), incentives are
generally inefficient when it comes to problems arising from cognitive lim-
itations. However, every designer must take measures to counter various
sources of misalignment between individual decisions and the organization’s
best interest. Ensuring alignment is one of the main tasks of oversight.
Just like management, oversight involves individuals. The most salient
part of the organization charged with oversight is a board of some kind.
Boards come in different forms and with different names, depending on
the context. For-​profit limited liability companies have boards of directors,
universities have boards of regents, the Spanish Railway Foundation and the

4 In a classic study of managerial perceptions, organization scholars DeWitt Dearborn and

Herbert Simon (1958) had a group of executives from different functional departments of the same
manufacturing company read the Castengo Steel Company case used in instruction in business
schools in the 1950s. Before discussing the case, the executives were asked to write a brief statement
of the problem they thought Castengo faced. The executives were further explicitly instructed to ap-
proach the Castengo case from the point of view of the company instead of their own department or
function. Yet, most sales executives saw sales as the main problem, production executives saw the
problem as one of internal organization, and so forth.
42  Fundamentals of Efficient Organization

Rockefeller Foundation have boards of trustees, the Federal Reserve System

has a board of governors, and so on. All these boards have a job to do, but
this job must be distinguished from the job of management. Board members
typically do not have employment contracts with the organization but, in-
stead, fixed-​term appointments with little job security. In many limited lia-
bility companies, board members may be dismissed without cause at every
annual shareholders’ meeting. It is often useful to think of those serving on
the board not as members but as stewards of the organization; we also find
the word trustee descriptively accurate. The task of a steward or a trustee is
to ensure that the members of the organization act in the best interest of the
entire organization and all its stakeholders, whoever they happen to be in the
specific situation. In short, board membership is a position of trust, not ad-
vocacy. We return to the fiduciary (as opposed to contractual) role of boards
in ­chapters 4 and 5.
Just like management consists of both human and nonhuman assets, over-
sight involves both individuals and groups making decisions (at board, com-
mittee, and shareholders’ meetings), as well as various rules, guidelines, and
procedures. Some of these rules and procedures are provided by the insti-
tutional environment, most notably, the relevant laws. However, others are
provided by the organization itself as matters of private ordering: articles
of association, bylaws, and scores of both formal and informal contracts.
Consistent with the Fifth Premise in ­chapter 1 (“Focus on Private Ordering”),
we focus on the organization-​specific instruments available to designers to
use at their discretion. Thus, in examining the governance of startup firms
(­chapter 6), for example, we focus less on what the law stipulates and more on
what the founders can and should do to ensure efficient governance.
Alignment of local actions within the organization with broader organ-
izational interests is important but not the only function of oversight. We
propose that decisions regarding oversight must also address how a potential
organizational residual (e.g., an economic surplus) is governed. To this end,
the designer must understand the role of residual claims and residual rights
of control in oversight.

Residual Claims

In their groundbreaking book The Behavioral Theory of the Firm, organiza-

tion scholars Richard Cyert and James March ([1963] 1992, 42) suggested
 The Efficiency Lens  43

that organizations become particularly interesting when they exhibit at least

some slack. In this context, slack refers to some form of excess that is not im-
mediately required to keep the organization functioning.
Imagine an organization that always consumes all the resources available
to it just to be able to function. Such an organization would be in many ways
simpler than one with slack. What an organization should do with the re-
sources it does not need for immediate use is indeed a central question for
oversight—​in short, slack must be governed. An organization with zero slack
is simpler because governing slack is unnecessary.
Consider the most salient form of organizational slack: An organization
creates revenue, and after all contractual and legal obligations have been met,
there is something left over. In the corporate context, this leftover is called
profit, but since our focus here is more broadly on organizations, we use the
more general term surplus. Nonprofit organizations such as universities often
produce an economic surplus, and since it would be confusing to speak of “a
nonprofit making a profit,” we suggest surplus is a better term.
What kinds of rules guide the decisions regarding the governance of the
surplus? In some organizations, someone can make an explicit claim for the
residual; these are called residual claimants. Residual claims and claimants
are most commonly discussed in the context of for-​profit organizations in-
corporated as limited liability companies where shareholders are residual
claimants.
Some organizations have residual claimants, others do not, or, more accu-
rately, in other organizations the only residual claimant is the organization
itself. In such organizations, a potential economic surplus will be used solely
for the benefit of the organization. Fama and Jensen (1983a, 318) proposed
this as a straightforward definition of a nonprofit organization, which we
adopt in this book as well. Specifically, a nonprofit organization is one that
has no residual claimants other than the organization itself. Therefore, the
criterion of a nonprofit is not whether or not the organization produces an ec-
onomic surplus but whether someone is entitled to appropriate it in the form
of private benefits. As a general rule, nonprofit organizations—​particularly
those that receive a favorable tax treatment—​are not allowed to grant private
benefits. We return to nonprofit organizations, with examples, in c­ hapter 5.
Organization economists Armen Alchian and Harold Demsetz (1972,
789) insightfully asked: “[W]‌ hy should stockholders be regarded as
‘owners’ in any sense distinct from other financial investors?” To jus-
tify their position, Alchian and Demsetz maintained that the primary
44  Fundamentals of Efficient Organization

difference between shareholders and bondholders, as investors, was that

the former were more optimistic about the future of the firm. We partly
agree with Alchian and Demsetz’s position but suggest that shareholders
are indeed owners of the corporation in an important sense: They have
property rights over residual income. At the heart of ownership is re-
sidual income and residual control (Milgrom and Roberts 1990), which
introduces the third aspect of oversight: residual rights of control.

Residual Rights of Control

The most elusive aspect of oversight relates to residual rights of control.

Whereas residual income and residual claims refer to the economic surplus
that is left over after all contractual obligations have been met, the notion of
residual rights of control refers to all the considerations that remain unspec-
ified in contracts. This shifts attention from deciding how to govern residual
income to decisions about how the organization uses its assets; its materials,
its technologies, its employees’ time, and so on. Many aspects of how assets
are to be used can be written into formal contracts, but long-​term contracts in
particular are often materially incomplete; they cannot possibly cover every
detail, scenario, and contingency. The premise that contracts are often mate-
rially incomplete stands at the foundation of many economic approaches to
governance.
Consider the scenario where a technology firm retains the services of
a law firm in an intellectual property dispute with a competitor. The tech-
nology firm and the law firm enter into a materially incomplete, open-​
ended contract where the law firm agrees to represent the client. The
contract is incomplete because many fundamental aspects of the contract,
such as contract duration and total price remain unknown and unspeci-
fied. Who retains the decision rights over issues not covered in the con-
tract? Here, the attention turns to the discretion and ownership aspects of
oversight.
There is a long and rich tradition in the economics literature on what it
means to own something. The main point in this literature is that the essence
of ownership lies not so much in the legal conception of having title to an
asset as it does in the organizational and practical implications of ownership.
 The Efficiency Lens  45

This view takes us to the notion of control rights, residual rights of control in
particular. In the following sections, we discuss two examples.

Residual Rights of Control in Specialized Industrial Production

Consider a production line that can produce in a single eight-​hour shift
a total of 150 ice hockey sticks made of carbon fiber and epoxy resin.
Suppose further that the production line is highly specialized in that its
production technology can be used only to produce said hockey sticks,
nothing else. Then consider the design decision of whether the firm that
sells the sticks should invest in the production technology and make the
sticks in-​house or outsource production to a technology supplier. In this
example, what are the residual rights of control, and how are they relevant
to the designer?
Suppose the hockey stick firm decides to focus on marketing and brand
management and outsources stick production to an external supplier. In
the buyer-​supplier contract, the contracting parties specify contract dura-
tion, volumes, prices, delivery schedules, warranties, termination clauses,
and other pertinent, contractible issues. At the same time, a lot will remain
unspecified, making the contract materially incomplete. For example, it is
highly unlikely that the contract will stipulate the geographic location where
production takes place, when the sticks will be produced, who supplies the
raw materials, how production equipment maintenance is to be arranged,
and what kinds of employment contracts the factory employees will have.
These unspecified issues can usefully be considered a residual, which we
submit the designer should consider analogously with economic surplus
when thinking of oversight. However, this residual does not take the form of
an economic surplus but, rather, the rights to make decisions. The contrac-
tual party that gets to make the decisions regarding the issues not specified
in the contract has the residual rights of control. From an organization de-
sign and governance point of view, this is the primary implication of owner-
ship: The owner retains the residual rights of control.
Why not focus on ownership in intuitive and conventional terms? “I
paid for it, I have title to it, I do as I please with it.” Such thinking can be
misdirected. The reasoning is as follows.
The expensive and highly specialized production line is designed to pro-
duce composite ice hockey sticks, nothing else. Thus, even though the owner
46  Fundamentals of Efficient Organization

of the production line can do with the production equipment “as it pleases,”
it really has no other option but “to please itself ” with the production of ice
hockey sticks made of carbon fiber and epoxy resin.5 But this is exactly the
same as what the buyer would have “to please itself with” if it decided to pur-
chase the production technology and produce the sticks in-​house. Therefore,
no matter who holds title to the specialized technology, it will be used to pro-
duce ice hockey sticks.
Particularly in contexts where assets are highly specialized, the notion of
“doing as one pleases” loses much of its relevance. We suggest that the rel-
evant difference between in-​house production and outsourcing has to do
with residual rights of control. Again, both the buyer and the seller would
use the specialized production equipment for the purposes of producing ice
hockey sticks; there is no material difference here. In contrast, whether the
residual rights of control remain with the buyer or the supplier has important
implications.
Perhaps the most consequential question with regard to residual rights
of control concerns production capacity: How many production lines will
the factory host and in what geographic location, or locations? This ques-
tion looks very different to the supplier and the buyer, and in the case of ice
hockey stick production, likely leads to the conclusion that outsourcing will
be comparatively efficient. Why?
In the ice hockey equipment business, those selling the equipment to
consumers—​ Bauer, CCM, Warrior, True Hockey—​ are marketing and
brand management firms with little in-​house production. Both production
and product design are outsourced to large, specialized technology firms
in Southeast Asia. Furthermore, these technology firms seek to take advan-
tage of both economies of scale and scope, and in their contracts with par-
ticular buyers, they reserve the right to use their technology to supply other
buyers as well. Within their residual rights of control are also factory location
decisions, supplier selection, workers’ employment contracts, planning and
scheduling of production, and R&D investments.
The buyers, in turn, understand that economies of scale and scope on the
supplier side work to the buyers’ advantage as well, even if it means that the
same manufacturers also supply the competitors. Outsourcing production to

5 There is also another, albeit less crucial, point about the firm “doing as it pleases” with the assets
it owns. If an asset has been offered as collateral for a loan, the firm generally cannot sell the asset
without the creditor’s blessing. This, too, can be understood in terms of residual rights of control: The
creditor maintains significant residual rights of control in decisions pertaining to the sale of the asset.
 The Efficiency Lens  47

specialized technology firms that consolidate the production needs of mul-

tiple customers into one manufacturing firm is comparatively efficient, be-
cause economies of scale in production are such that an individual brand
owner producing only to one’s own needs would be inefficient. Large scale
has the further advantage of making investments in specialized assets and
automation attractive. If the small size of the production unit preempts such
investments, inefficiency results from underinvestment in technology.
We can also think of residual rights of control from the point of view of
factory management incentives. Suppose the buyer contemplates the op-
tion of in-​house production by vertical integration whereby it would pur-
chase a formerly separate and independent production plant, including all
its employment contracts. Vertical integration may create a disincentive, per-
haps even an explicit constraint, for plant management to seek clients other
than the firm that owns the internal production unit. Perhaps the internal
production unit would serve only an internal sales unit to which the sticks
would be transferred using an arbitrary transfer price. After vertical integra-
tion, the production unit would likely be run as a cost center. Consequently,
its profit would become arbitrary because it would depend on an arbitrary
transfer price. Making the production plant a cost center instead of a profit
center seems like a prudent alternative. But the trade-​off is that if instead of
assigning the plant profit-​and-​loss responsibility one only assigns cost re-
sponsibility (a budget), the plant management’s incentives shift from value
creation, innovation, and business expansion to cost control. Such incentive
distortions may lead to substantial inefficiency. We return to the more gen-
eral topic of efficiency distortions in ­chapter 8 where we discuss the govern-
ance challenges of established organizations.

Residual Rights of Control in Higher Education

A similar analysis can be applied in the examination of contracting with
employees. In fact, the benefits of conceptualizing ownership through re-
sidual rights of control become particularly salient when the focus shifts
from physical assets to work performed by humans. To be sure, the idea that
a consulting firm or a university “owns its employees” is awkward. In con-
trast, the notion that the organization retains residual rights of control over
its employees is salient.
Consider the question of whether a university should use internal fac-
ulty (employment contracts) or visiting and adjunct faculty (buyer-​supplier
contracts) to organize its teaching. Most universities use a combination of
48  Fundamentals of Efficient Organization

both, but there is considerable variability in degrees; some universities rely

exclusively on internal faculty, but in others, visiting faculty dominates.
There are many factors that account for this heterogeneity, but we suggest
that one can gain insight by examining residual rights of control.
The contractual arrangement with visiting faculty is governed as ei-
ther a one-​off or a recurring contractor-​contractee relationship. In the
formal contract, all activities to be performed by the contractee can flex-
ibly be tailored to fit the specific situation. For example, the contracting
parties may agree that the visiting faculty member will teach an elective
that consists of fifteen eighty-​minute sessions on organization design and
governance in the Master of Business Administration Program in the Fall
2022 semester.
The visiting faculty member is contractually bound to perform the spe-
cific activities stipulated in the contract but simultaneously retains many
residual rights of control, such as what teaching techniques to use, whether
to hold a final exam or have the students complete a group project, and
how much weight to give to in-​class participation in performance evalua-
tion. In higher education, it is generally a good idea to leave residual rights
of control regarding activities inside the classroom to the faculty member.
Other residual rights of control, such as how many students will be allowed
to enroll and where the class will be taught, are left to the employer.
The situation of the internal faculty member is different in a number of
relevant ways. One is that whereas internal faculty members enjoy some of
the same residual rights of control inside the classroom as the visiting faculty
member, internal faculty members may be required to teach some courses
using a common syllabus used by other professors as well. For example, a
professor of operations management teaching a core MBA seminar on op-
erations management may be required to coordinate syllabus content with
other professors teaching the same core seminar, which materially limits the
professor’s discretion.
Another difference is found in the substance of the formal contract.
Internal faculty members’ employment contracts are written in compara-
tively general terms, leaving significantly broader residual rights of control
to the employer. For example, the faculty member’s employment contract
may stipulate a teaching load of one hundred sessions every academic year,
without specifying what courses the faculty member is to teach, to whom,
and when. Similarly, the contract may require 240 hours of service per aca-
demic year, leaving the details of what this service entails to the employer’s
 The Efficiency Lens  49

discretion. A professor’s service could involve serving on faculty committees

or as department chair, advising graduate students, applying for external
grants, and the like. Some of the service could also be converted to teaching.
Further specification in this case is generally the employer’s prerogative; in
short, the employer has residual rights of control.
That the university retains the primary residual rights of control does not
mean that it should unilaterally dictate how internal faculty members are to
allocate their time. If the university is interested in securing the long-​term
cooperation of particularly those faculty members it does not want to lose, it
should be open to ex post negotiation and incorporate the faculty members’
preferences into all decisions. For example, employment contracts seldom
address teaching in the evenings and over the weekends, and it would be ill
advised for the university to present unilateral requirements. A better alter-
native is to provide sufficient financial incentives to secure the cooperation of
an adequate number of faculty members willing to devote their evenings or
weekends to teaching.
Even though unilaterally dictating what the internal faculty member
is to teach and when would likely be myopic, the central governance prin-
ciple is that the university, not the faculty member, exercises oversight over
the allocation of the faculty member’s time. Accordingly, in the case of con-
flicting preferences, the university should be more comfortable challenging
the internal faculty member’s preferences than vice versa. Faculty members,
at least those well versed in the ways of efficient governance, should feel
compelled to accept the premise that in their employment relationship, the
employer is entitled to exercise residual control, at least as long as it is in the
employee’s zone of acceptance. Hopefully, both contractual parties will be in-
terested in seeking an efficient solution that establishes and maintains mu-
tual credibility.
The upside of using visiting faculty is that no long-​term commitments are
required. This advantage has allowed universities to adjust to the COVID-​
19 pandemic. Specifically, whereas the university can selectively renew only
those visiting faculty contracts required at any given time, it cannot termi-
nate the contracts of internal faculty without substantive and substantial
cause. In fact, in some jurisdictions the bar is set so high that unless the uni-
versity can establish that the pandemic threatens the very survival of the or-
ganization, internal faculty members may contest as illegal any termination
or suspension of employment if the main justification is that the employees’
contributions are not needed at the particular time. This reasoning applies to
50  Fundamentals of Efficient Organization

private organizations as well: The employer must establish a legal (not merely
an economic) basis for termination.
The downside of using visiting faculty is that unlike in the case of internal
faculty, visiting faculty cannot flexibly be assigned to different organizational
tasks if the situation calls for it. The visiting faculty members make an offer of
how their time will be used, and once the university and the visiting faculty
member sign the contract, the university has no further discretion. Asking
the visiting professor at the last minute to teach another class is both unrea-
sonable and infeasible; canceling the class at the last minute likely constitutes
a breach of contract. With internal faculty members, in contrast, the univer-
sity retains comparatively broader residual decision rights regarding the use
of the internal faculty member’s time even after the contract has been signed.
Why do we see such diversity in the extent to which universities use
internal versus visiting faculty? It is undoubtedly partly a matter of uni-
versity reputation, as leaning heavily toward using visiting faculty can be
viewed as illegitimate. Simply put, a reputable business school has its own
faculty, it does not borrow someone else’s. However, we suggest that the
diversity can also stem in part from different universities thinking dif-
ferently about the extent to which they seek to maintain residual rights
of control. Even though the use of visiting faculty may confer many
advantages, a business school that allows itself to become dependent
on visiting faculty can lead to highly unstable, fad-​of-​the-​month-​type
course offerings from one year to the next. A university takes a consid-
erable risk if it lets its curriculum depend on the availability of external
contracting parties. Ultimately, this approach might even jeopardize the
very survival of the university. In this sense, the reputation and the effi-
ciency perspectives are intertwined.

The General Applicability and Implications of the Residual

The ice hockey production and university teaching examples apply more
broadly to decision situations where an organization weighs the options of
performing work internally versus contracting it out to external providers.
The use of contract workers instead of employees is increasingly common in
many organizational settings: The military and the construction industries
make extensive use of contract workers; media companies buy content from
freelance journalists; real estate and retail companies use independent sales
agents instead of internal sales personnel; and so on. In all these contexts, the
way in which residual rights of control are distributed among the contracting
 The Efficiency Lens  51

parties have efficiency implications and, therefore, constitute an important

organization design consideration.
Similarly, the advantages and the disadvantages generalize to other
settings. If an industrial firm makes a component in-​house, it maintains re-
sidual rights of control regarding how it manages production. Analogously,
if a firm has its own legal department and patent lawyers, it enjoys the ben-
efit of having wide discretion over assigning particular lawyers to particular
cases. If an external law firm is used, the client can make proposals for the
timeline and request that specific lawyers be assigned to the case; however,
the external firm retains residual rights of control on both the assignment of
the client team and the timeline. The question of residual control has wide-​
ranging implications for the governance of contractual relationships.
Focusing on the residual also suggests an intriguing angle to ownership,
as it challenges the conventional definition of ownership based on who holds
the title to the asset. The conventional definition may be useful in many
contexts, but we propose that in the specific context of governance, the re-
sidual offers a more useful vantage point to ownership. Specifically, instead of
thinking of ownership as having the right to do with the asset as one pleases
(as was the case in Roman law), it is more useful to think of ownership in
terms of (1) who is entitled to exercise control over the decisions that are not
specified in contracts (residual rights of control), and (2) who is entitled to
the economic surplus the organization generates (rights to residual claims).
Considering residual rights of control and rights to residual claims sep-
arately is important, because as organization economist and Nobel Laureate
Oliver Hart (1989, 1766) noted, “these rights will often go together, but they
do not have to,” which is why the designer must give attention to both. Only
an explicit analysis of both residual claims and residual rights of control in the
context of the organization will uncover the ramifications in their entirety.
Understanding ownership as residual rights of control is particularly relevant
in knowledge work contexts, such as professional service organizations. A con-
sulting or a law firm does not own its employees in the conventional sense, but
questions of residual rights of control are of foundational importance.

Risk

Numerous individuals and collectives create economic value for an organi-

zation and benefit from the organization’s existence in one way or another;
52  Fundamentals of Efficient Organization

we use the term constituency to refer to these individuals and collectives.

However, the designer must acknowledge that some constituencies have
more at stake in the organization than others. Accordingly, we reserve the
term stakeholder to refer to those individuals and collectives that, by virtue
of their participation in the organization, have voluntarily put something at
risk, and consequently, have become vulnerable in one way or another. It is
imperative that the designer understand the variable degrees of vulnerability
among the organization’s constituencies.
To us, stakeholder is not a distinct category but, rather, a comparative no-
tion: Some constituencies have more of a stakeholder status with the organ-
ization than others. Categorizing constituencies into precisely demarcated
stakeholders and nonstakeholders is an oversimplification with potentially
adverse consequences.
What does it mean to put something at stake in an organization? In what
ways do constituencies become vulnerable? We propose a simple thought
exercise as a litmus test. Consider all the organization’s constituencies, one
at a time, and ask the following question: “If the organization unexpect-
edly ceased to exist, how would the specific constituency be affected?” All
constituencies would undoubtedly be affected at least to an extent, but the
rigorous designer analyzes degrees (see ­chapter 4 for details). Consequently,
the designer must incorporate the variable degrees of risk into governance
decisions. A failure to do so will jeopardize the credibility of the organiza-
tion in the eyes of those who are asked to put the most at stake.
Note that the issue is not whether a constituency is vulnerable in the
general sense but whether the constituency becomes vulnerable by virtue
of joining the organization. For example, money provides buffers against
uncertainty, and those with lower incomes tend to be in the general
sense more vulnerable than those with higher incomes. This difference
does not, however, mean that those who receive comparatively lower
salaries in the organization should be considered vulnerable in the gov-
ernance sense. The only situation in which low income implies vulnera-
bility in the governance sense is if the employees’ low income arises from
their membership in the organization. An example is the employee of a
startup. Startups are strapped for cash and cannot pay their employees,
even CEOs, salaries that are anywhere close to being competitive. The
designer must incorporate this discrepancy into governance decisions. In
startups, the standard response is to compensate inadequate salaries with
an equity stake.
 The Efficiency Lens  53

Shareholders as Residual Claimants

The canonical risk taker is the shareholder of the limited liability company. In
providing equity financing to the firm, the shareholder is guaranteed nothing,
only a claim to the potential residual. Unlike employees and suppliers who
are contractually guaranteed to receive fixed payments from the firm, the
status of a constituency that receives only residual payments is comparatively
vulnerable. The distinction of fixed versus residual payments is central.
An investor may pay €30 per share for a thousand shares, lose every last
penny, and not be entitled to compensation of any kind from anyone; the
shareholder is in principle risking everything without a safety net. This risk
is partly alleviated in situations in which the shareholder is free to buy and
to sell shares in an open market, thus transferring residual claims to another
investor. Organization economists speak of “freely alienable” residual claims
(e.g., Fama and Jensen 1983a, 312), which individual investors can use to
manage risk. If in addition the shares are publicly traded, the price at which
shares are bought and sold is unambiguous.
Free alienability enables risk management at the individual investor level.
However, this is not what is relevant to the designer who must acknowledge
that the organization’s residual claimants as an aggregate are always vulner-
able. Accordingly, the designer’s task is not to please any individual investor
but, rather, to ensure that the organization maintains its credibility in the
eyes of the providers of equity capital in the aggregate. The designer need
not worry about losing credibility in the eyes of any individual investor, but
a large number of investors losing credibility constitutes an imminent threat
to the value of shareholders’ equity and, in the long term, the organization’s
existence.
In startup firms, the stakeholder status of shareholders is even more pro-
nounced for two reasons. One is that there is no market price; all valuations of
equity are based on projections of highly uncertain future cash flows. In the
absence of unambiguous prices, trading shares becomes subject to difficult
negotiations due to asymmetric information. The other reason is that many
shareholders’ agreements in startup firms place restrictions on the aliena-
bility of residual claims. For example, founder-​shareholders may be required
to offer their shares to other founders before offering them to outsiders. Or,
if the company is party to the shareholders’ agreement, the company may be
entitled to purchase any shares offered for sale at their book value before they
can be offered to outsiders. In many startups, the book value of equity is zero,
54  Fundamentals of Efficient Organization

or close to zero. Agreeing to constraints on the alienability of residual claims

is an important design decision for a startup firm.
Nuances aside, providers of equity financing certainly have become
vulnerable by putting something at stake, which is why they should be
considered stakeholders of the limited liability company. At the same
time, they are not the only relevant stakeholder, perhaps not even the
most important one. Chapter 4 discusses the pertinent governance is-
sues in detail, but we will briefly consider here the idea that an em-
ployee can also become vulnerable, and therefore, should sometimes
be awarded stakeholder status instead of being considered merely a
constituency.

Employees and Risk

Let us apply the stakeholder litmus test to the organization’s employees: If the
organization ceased to exist, how difficult would it be for an employee, or a
specific employee group, to find alternative employment? The answers range
from not difficult at all to next to impossible. Another question is whether the
employee, or employee group, would be able to find alternative employment
at the same level of income. Again, the answers will vary. The variance in
answers to both questions must be analyzed and the implications incorpo-
rated into governance decisions.
The two authors of this book are business school professors who teach
comparatively general topics of strategy, governance, and operations man-
agement to undergraduate, graduate, postgraduate, and executive audiences,
using the English language. If for some reason our schools ceased to exist,
neither author would have considerable trouble finding new employment in
another business school, and neither would have to accept a job that pays
less than the current job. Securing new employment might involve incon-
venience, such as having to move to another city, but we must not con-
fuse inconvenience with economic risk. It would be hyperbole to propose
to the designer of the business school that professors are vulnerable be-
cause they would be inconvenienced by the termination of their employ-
ment relationships. Professors would be at risk only if their employment
relationships with the specific universities made them vulnerable. Insofar as
this vulnerability is concerned, what is most relevant is that neither author
 The Efficiency Lens  55

has committed to the kinds of skill sets that would tie us to a specific em-
ployer. Just the opposite, our expertise is readily redeployable in many other
organizations.
In stark contrast with business schools and their professors, there are nu-
merous contexts in which knowledge and expertise are organization specific,
or as is more commonly the case, become organization specific over time.
Consider the example of a software development company where engineers
are asked to commit to learning and further developing an organization-​
specific technology, such as a proprietary programming language or unique
software products. The engineers know that being skilled in the specific
technologies may have considerably less economic value in another organ-
ization. Committing to such specificity is a form of employment risk, and
something that warrants attention in organization design in general and em-
ployment relationships in particular.6
In the case of high levels of specificity, it may well be that the employment
contract alone will not be sufficient to secure the commitment of some of
the key employees. Or, the employment contract will be inefficient in the
sense that the salaries must be raised to unfeasibly high levels to offset the
perceived employment risk that arises from specificity. In such situations,
the organization is well advised to think of its members more broadly than
merely in terms of employment contracts where the employees exchange
their time and effort for a salary. Unlike investment portfolios, employment
portfolios are much more difficult for individuals to diversify, which means
employment risk is more difficult to manage. This is why employment risk
warrants the designer’s attention, particularly in the case of employees the
organization would have trouble replacing. In some contexts, employment
risk may be more central to the organization’s survival than investment
risk. This may occur in organizations that do not need to raise equity to fi-
nance their operations—​professional service firms such as law firms and ac-
counting firms are good examples.

6 This is an authentic example that is based on a conversation one of the authors had with the chief
technology officer (CTO) of a software company. The CTO candidly disclosed that the firm faced
considerable difficulties attracting talented programmers to the firm due to the specificity involved.
Prospective employees understandably viewed the commitment to specificity as risk, for which they
demanded a significant price (i.e., salary) premium. Considering specificity is central in the ex ante
stage when the organization seeks to strike a deal with an important constituency. In the case of the
software company, a failure to secure the cooperation of a sufficient number of skilled programmers
leads to underinvestment in a strategic technological skill. The author’s recommendation to the CTO
was to think of the relationship between the firm and the programmers in broader stakeholder terms,
not merely as an employment relationship.
56  Fundamentals of Efficient Organization

Specificity applies to nonhuman assets as well. An example is the use of

special-​purpose production equipment in a manufacturing firm. Often, in-
vestment in special-​purpose equipment is justified because it generates more
value than general-​purpose equipment. The downside is that its economic
value in the second-​best use may be considerably lower, even nonexistent.
An organization that requires firm-​specific assets to operate must often rely
on equity financing, because lenders will not accept firm-​specific technology
as collateral.
We conduct a more detailed analysis of risk in its various forms in c­ hapter 4
where we present a proposal for stakeholder analysis. In this analysis, speci-
ficity takes center stage.

Negative Externalities and Involuntary Risk

In this book, we focus primarily on risk in contexts where the contracting

parties enter into the relationship voluntarily. However, there is an important
aspect of involuntary risk, found in situations where risk is imposed without
consent. In some contexts, the problem is further exacerbated by the fact that
no consent is possible in the first place because the entity on which risk is
imposed does not have agency; the environment is the most salient example.
Imposition of risk links to negative externalities, or within an organiza-
tion, to negative spillover effects. In a negative externality, something that an
organization does has unintended and undesirable effects on entities that do
not have a contractual relationship with the organization. A case in point,
the Great Pacific Garbage Patch consists of 1.6 million square kilometers—​
about the size of Mexico—​of plastic waste, such as bottles, bags, and the like.
This obviously constitutes a risk to the environment, to wildlife, and to the
residents of the Pacific islands. Furthermore, it is a risk that has been imposed
without consent.
How should the designer approach the situation in which the
organization’s actions impose a risk on the environment? Should the envi-
ronment be considered a stakeholder? If the designer answers in the affirm-
ative, how should the designer incorporate the interests of a stakeholder that
is not a contracting party in the conventional sense? The question whether
noncontracting entities having neither discernible identity nor representa-
tion should be considered stakeholders is complex, and we have no definitive
answers. But we can propose a number of questions and issues to address.
 The Efficiency Lens  57

On the one hand, the natural environment is an important part of every

organization’s environment, and taking responsibility for protecting those un-
able to protect themselves seems like a prudent general principle. At the same
time, it is difficult to think of this responsibility in stakeholder terms. If the envi-
ronment is a stakeholder, what are the organization’s responsibilities toward the
environment, and the environment’s responsibilities toward the organization?
Trying to formulate the question in contracting terms runs into considerable
difficulties.
It seems that if the environment is to be considered in stakeholder terms, a
different approach is required. The general question of negative externalities
boils down to the question whether externalities should be viewed through the
lens of contracting (private ordering), legislation and policy (public ordering),
or moral obligation. The answer is that it is likely a combination of all three, and
all three have implications for governance. However, the nonobvious question
we present to the designer is, “In what ways should negative externalities be in-
corporated into governance as matters of private ordering?”
There is no consensus on the issue. Those who suggest private ordering is
less relevant tend to point to the centrality of public ordering: “The existence
of organizations that advocate for the environment, such as the Environmental
Protection Agency and the Department of the Interior as well as the Sierra Club
and the National Resources Defense Council and others, renders stakeholder
status for the natural environment redundant or unnecessary” (Phillips and
Reichart 2000, 188–​89). Those who take the opposite position tend to suggest
that “human proxies for the non-​human environment” (Phillips and Reichart
2000, 189) are necessary but not sufficient (see also Starik 1995).
One need not but take a quick look at the pollution indexes across the
globe to reach the conclusion that the public measures taken to protect the
environment are woefully insufficient. However, we must remain intellectu-
ally rigorous and acknowledge that this does not imply that private ordering
provides a comparatively efficient alternative. Whether further reinforcing
public ordering (through legislation, policy, and polity) is efficient compared
to private ordering requires further analysis. Those who think that public
institutions are inefficient should also consider the possibility that the same
inefficiency applies to private-​ordering initiatives. In his scathing Harvard
Business Review essay,7 management scholar Kenneth Pucker observed the
following:

7 https://​hbr.org/​2022/​01/​the-​myth-​of-​sust​aina​ble-​fash​ion
58  Fundamentals of Efficient Organization

Few industries tout their sustainability credentials more forcefully than the
fashion industry. But the sad truth is that despite high-​profile attempts at
innovation, [it has] failed to reduce its planetary impact in the past 25 years.
Most items are still produced using non-​biodegradable petroleum-​based
synthetics and end up in a landfill [ . . . ] [G]‌overnments need to step in
to force companies to pay for their negative impact on the planet. (Pucker
2022, summary paragraph)

The challenge of discussing environmental concerns in terms of private

ordering in for-​profit corporations is that these concerns ultimately tend to
become subservient to shareholder interests. Legal scholars Lucien Bebchuk
and Roberto Tallarita (2020, 109) are spot on in observing that in for-​profit
corporations “consideration of [environmental concerns] is a means to the
end of shareholder welfare.” Given both top management and director com-
pensation in large corporations is heavily equity based, the dispassionate
analyst must consider the possibility that private ordering as a response to
negative externalities may amount to an “illusory promise” (Bebchuk and
Tallarita 2020, 91).
Finally, even though invoking the moral aspect introduces an entirely new
dimension to the conversation, let us suggest that it can usefully be linked
to efficiency. A case in point, the taxi ride example in ­chapter 1 implicitly
embraces the general moral responsibility of waste avoidance in individual
decisions. In this sense, there is no immediate reason why efficiency cannot
be compatible with considerations of what is ethically justified. This topic is
discussed in more detail in c­ hapter 5.

How Management, Oversight, and Risk Relate

to One Another

Having defined the three key elements, we now turn attention to their
interrelationships, which are just as central to governance as are the concepts
themselves. Of particular importance is the extent to which management,
oversight, and risk are separated from one another. Let us consider two
examples.
Consider first the proportion of board members who should be inde-
pendent outsiders. In Efficiency Lens terms, the question is about the extent
to which oversight should be separated from management and risk.
 The Efficiency Lens  59

At least some separation seems necessary, because a board member who is

also involved in management may run into conflicts of interest in the board-
room. Suppose the board of a manufacturing firm contemplates the decision
to close one of its factories. The CEO may have a vested interest in closing
the plant, particularly if this strategic move will have a positive effect on fi-
nancial performance and, consequently, the CEO’s non-​equity-​based bonus
compensation.
A board member who also represents risk may similarly be unable to ap-
proach the question impartially. For example, union representatives on the
board will likely see the plant closure primarily as having negative outcomes
to their constituencies. Even though the union representatives’ concerns are
valid, as board members they do not represent the union but the entire or-
ganization. Indeed, under the law, board members have a duty of loyalty and
care not to their constituencies but to the corporation (Clark 1985; Blair and
Stout 1999). Consequently, the very notion of union representative is prob-
lematic; so are shareholder representative and stakeholder representative, for
that matter. Under the law, every board member is a representative of the or-
ganization, not its individual stakeholders.
Consider second the scenario of a risky strategic acquisition decision. To
what extent should those making the decision also bear the consequences if
the acquisition fails and destroys value instead of creating it? This question is
about the degree to which management overlaps with risk. On the one hand,
we can see how having no overlap between the two could be viewed as prob-
lematic. Specifically, is it fair to have someone who had nothing to do with
the decision bear the financial consequences of failure? On the other hand, it
is important to see also the problems that overlap creates. If members of the
top management team faced significant economic consequences in case the
investment failed, the predictable consequence is that they would become
highly risk averse in their decisions. Such investment paralysis might eventu-
ally jeopardize the survival of the entire organization.
More generally, we propose that not only can we express central gov-
ernance decisions in terms of how management, oversight, and risk are
related, but also the definition of efficient governance boils down to the
question, “What constitutes a credible configuration of management,
oversight, and risk?” By configuration, we refer specifically to how the
three elements are interrelated and the extent to which they overlap. Let
us examine two illustrative examples: the sole proprietorship and the open
corporation.
60  Fundamentals of Efficient Organization

The Sole Proprietorship

The simplest organization is one where management, oversight, and risk

are completely overlapping (fig. 2.2). Those who make the most important
decisions bear all the relevant risk and, therefore, are readily incentivized
to exercise oversight and control over their own decisions. The closest to
this ideal is the one-​person organization incorporated as a sole proprietor-
ship. The sole proprietor has all the incentives to get the job done in an effi-
cient way; there are no conflicts of interest, and because there are no agency
relationships, there are no potential agency problems either. In a sole pro-
prietorship, management, oversight, and risk are for all practical governance
purposes embedded in just one person.
Even in the case of the sole proprietorship, management, oversight, and
risk are, strictly speaking, not fully overlapping. Even a one-​person organ-
ization incorporated as a sole proprietorship must relinquish a number of
central oversight functions to other entities, such as accounting firms and
tax authorities. It is simply not good governance to let even the smartest and
most honest sole proprietor accountants perform financial audits of their
own firms. Therefore, management and oversight are always only partially
overlapping.
Similarly, due to externalities and spillovers, the sole proprietor is not the
only entity exposed to risk. For example, an incompetent or careless sole pro-
prietor accounting consultant can expose clients to unnecessary risks, even
litigation. Therefore, management and risk are not completely overlapping
either. This potential hazard gives rise to various institutional safeguards. In
the field of accountancy, a forward-​looking client is well advised to seek the
services of certified accountants, even if they are more expensive. Indeed, in
some countries only certified persons actively applying their certified skills
are allowed to use the label accountant, which provides an important safe-
guard that does not exist in many other settings. We conjecture that one of
the reasons the title consultant sometimes raises suspicions is because its use

Figure 2.2  The configuration of a sole proprietorship viewed through the

Efficiency Lens
 The Efficiency Lens  61

is not regulated in any way. Tales of shady chartered professional accountants

are much less frequent.

The Open Corporation

At the other end of the complexity spectrum is the organization where

management, oversight, and risk are separated from one another (fig. 2.3).
A large, publicly traded limited liability company is an example. In some
conversations, these organizations are dubbed public corporations or public
limited companies. We prefer using the word open instead of public, be-
cause from a governance perspective, the fact that the corporation’s shares
are publicly traded does not make the organization public in the relevant
sense. Because most publicly traded corporations are private organizations,
we find the term open corporation descriptively accurate (Fama and Jensen
1983a, 303).
Due to the fragmented ownership common in most open corporations,
management owns a very small fraction of shares; therefore, management
is materially separated from risk. As of December 28, 2020, Apple CEO Tim
Cook—​the person in charge of the most important strategic decisions—​
owned 837,374 shares of Apple stock, which represented 0.02 percent of
all outstanding shares. If Mr. Cook made a bad decision and the value of
Apple’s equity plummeted, 99.98 percent of the direct financial consequences
alone would fall on individuals and groups most of whom had no role in
the decision. If we considered other financial consequences, such as Apple
employees losing their jobs, Mr. Cook’s share of the loss would be even
smaller. Separation of management and risk should have profound organiza-
tion design and governance implications.

Figure 2.3  The configuration of an open corporation viewed through the

Efficiency Lens
62  Fundamentals of Efficient Organization

One implication of risk separating from management is the necessity

of separating management from oversight as well. The most conspicuous
manifestation is the fact that aside from the CEO, it is uncommon for any
other member of management to serve on the open corporation’s board of
directors. Instead, directors are outsiders employed in other corporations.

Focus on Material Separation

Given that management, oversight, and risk are neither completely

overlapping nor completely separated, the designer must direct attention to
the question whether the observed degree of separation is of material impor-
tance. To determine whether separation is material, we offer a simple litmus
test: Is the separation sufficient to merit a governance response? If the answer
is yes, then the separation is material.
The separation of management from risk in the open corporation is ma-
terial precisely because it necessitates a governance response, specifically,
material separation of oversight from management. An organization where
management is separated from risk but not from oversight is not credible
in the eyes of risk (see fig. 2.4). A look at just about any corporate scandal
illustrates the point.
One way to make sense of the events that took place at the Enron
Corporation at the turn of the millennium is to think of it as insufficient sepa-
ration of management and oversight when management and risk were mate-
rially separated. A plausible reason for Enron’s failure as an organization was
the fact that the integrity of the most important individuals in charge of over-
sight (chairperson of the board Kenneth Lay) and management (CEO Jeffrey
Skilling) was fundamentally compromised. The primary consequences of
Enron’s decline and bankruptcy fell on those bearing risk. The two obvious
groups affected were the shareholders, who lost a total of $74 billion, and the

Figure 2.4  A non-​viable configuration of management, oversight, and risk

 The Efficiency Lens  63

Enron employees, who lost their jobs. Those hit the hardest were employees
who were also shareholders; the pension plans of twenty thousand Enron
employees who lost their jobs were massively affected, some annihilated.
In summary, in the publicly traded corporation, the separation of manage-
ment and risk is always of material importance and requires the designer’s
attention; the obvious response is to separate oversight and management
powers from one another. In contrast, in a sole proprietorship, separation of
management from risk is not materially important. Although the sole pro-
prietor may not, strictly speaking, be the sole risk bearer, the amount of risk
sole proprietors carry tends to be sufficient to incentivize them to deeply care
about the fate of the organization. Being the primary risk bearer gives the en-
trepreneur the requisite high-​powered incentive to make business decisions
that are in the best interest of the organization, and indirectly, in the best in-
terest of all those who bear risk.

The Dynamic of Management, Oversight, and Risk

The third characteristic of the Efficiency Lens has to do with how manage-
ment, oversight, and risk, and their interrelationships, evolve over time.
A technology startup heading toward an IPO is a good example. On the IPO
path, there are three distinct but related dynamics (fig. 2.5):

Figure 2.5  The dynamic of a startup viewed through the Efficiency Lens
64  Fundamentals of Efficient Organization

1. As a startup organized as a private limited liability company heads to-

ward an IPO, it must add nonfounders and noninvestors to its board,
which means that management and oversight begin to separate. At
the IPO, the requirement for separation is explicit: The governance
guidelines of stock exchanges consistently require that at the time of
the IPO, the majority of board members be independent.
2. An IPO implies that ownership will become more dispersed as it is
opened to outside investors. This increased dispersion of owner-
ship means that those who make the key decisions will bear a smaller
portion of the overall risk, which reduces the overlap between man-
agement and risk. This alone requires that the organization must also
reduce the overlap between management and oversight. An organiza-
tion where management and oversight are highly overlapping with one
another but nonoverlapping with risk is not viable.
3. As risk and oversight separate from management, they tend not to move
in opposite directions, because new investors are reluctant to take on
risk unless they are awarded an oversight role. Separation of oversight
and risk accelerates when the startup nears an IPO, because ultimately
the firm must comply with stock exchange governance guidelines that
require clear separation of oversight from both management and risk.

Consistent with the First Premise from ­chapter 1 (“Organizations as

Deliberately Designed”), we approach the dynamics of management, over-
sight, and risk as a series of deliberate design choices that seek to maintain
the credibility of the organization in the eyes of its central stakeholders. As
the startup example shows, separation of powers is the outcome of conscious,
deliberate decisions that modify the composition of the board of directors.

Summary: Defining Governance

It may seem strange to offer a definition of a key term at the end of a chapter
instead of the beginning. But with a concept such as governance, we find it
impossible to define the term without first discussing and elaborating the
underpinning foundation on which it stands. After having introduced the
Efficiency Lens, we are prepared to offer our definition of governance.
Cambridge Dictionary defines governance as “the way that organizations
or countries are managed at the highest level, and the systems for doing
 The Efficiency Lens  65

this.”8 It is hard to find much utility in this definition. Furthermore, the idea
of defining governance as the way an organization is managed is unneces-
sarily narrow and potentially misleading.
The challenge with defining governance stems from the fact that in its
abstractness and generality, governance is analogous with concepts such
as sport, classical music, or mental health. These terms are not so much
concepts to be defined as they are umbrella terms whose domains must be
circumscribed. For instance, most discussions of what classical music is and
what it is not will likely involve a discussion not of definitions but, instead, of
composers and compositions that belong to its domain. This process is not
so much about defining what classical music is as it is about circumscribing
its domain. Indeed, Cambridge Dictionary “defines” classical music not by
offering a formal definition but by specifying its general domain as “a form
of music developed from a European tradition mainly in the 18th and 19th
centuries.”9
We suggest that a useful way of approaching governance as a concept is not
to try to offer a definition but to provide a list of questions that circumscribe
its domain. We circumscribe governance with three interrelated questions:

1. Who in the organization is trusted to make the most important

decisions about how activities are organized, how resources are
allocated, and how performance is evaluated? In Efficiency Lens terms,
who is in charge of management?
2. What general guidelines and principles govern decision making to en-
sure that decisions are made in the best interest of the organization? In
Efficiency Lens terms, how is oversight organized?
3. What safeguards are in place to ensure the cooperation of the
organization’s most important constituencies, particularly those who
have voluntarily put something at stake? In Efficiency Lens terms, how
does the organization address stakeholder risk?

Note that since these questions circumscribe governance primarily in

terms of private ordering, it would be inappropriate to offer this defini-
tion as universally applicable. We readily acknowledge that those who view
governance from an external point of view will emphasize compliance,

8 https://​dic​tion​ary.cambri​dge.org/​dic​tion​ary/​engl​ish/​gov​erna​nce
9 https://​dic​tion​ary.cambri​dge.org/​dic​tion​ary/​engl​ish/​classi​cal-​music
66  Fundamentals of Efficient Organization

law, regulation, and policy. A good example of an externally oriented def-

inition can be found in the Organisation for Economic Co-​operation and
Development’s (OECD) principles of corporate governance:

Effective corporate governance requires a sound legal, regulatory and in-

stitutional framework that market participants can rely on when they es-
tablish their private contractual relations. This corporate governance
framework typically comprises elements of legislation, regulation, self-​
regulatory arrangements, voluntary commitments and business practices
that are the result of a country’s specific circumstances, history and tra-
dition. (Organisation for Economic Co-​ operation and Development
2015, 13)

An externally oriented approach to circumscribing governance is fully

understandable given OECD’s emphasis on public policy and international
standards. OECD’s definition differs from ours not because we disagree with
one another but because we take an internal and OECD an external per-
spective. However, we view the two approaches as ultimately complemen-
tary. For instance, our private-​ordering view elaborates the self-​regulatory
arrangements, voluntary commitments, and business practices mentioned
in the OECD definition. Furthermore, in the discussion of startup govern-
ance in c­ hapter 6, we show how important it is for the designer to under-
stand how external rules and regulations impose limits on private ordering.
In short, private ordering occurs in a broader institutional context that both
enables and constrains private choices.
 PART II
GOVE R NA NC E WI T H I N A ND
AC RO SS ORGA N I Z AT IONS

In his groundbreaking 1937 article “The Nature of the Firm,” economics

Nobel Laureate Ronald Coase made a profound observation regarding the
organization of industrial production. When an industrial firm decides to
produce a component in-​house instead of buying it from an external sup-
plier, it is effectively choosing between transacting internally (within the
firm) and transacting in the market (across firms). Indeed, a market trans-
action and internal production are “alternative methods for co-​ordinating
production” (Coase 1937, 388).
In many contexts, the designer faces a similar choice of contracting either
within the organization or across organizations. The aim of c­ hapter 3 is to
show how these contracting decisions and choices can be subjected to a com-
parative efficiency analysis. To this end, c­ hapter 3 focuses specifically on the
costs of contracting.
In ­chapter 4, we take a contractual approach to stakeholder manage-
ment. An organization has numerous constituencies, some of which are
more critical to the organization’s viability than others. This chapter turns
attention to those constituencies that put something significant at stake by
striking a relationship with the organization. We call those with the most at
stake the organization’s stakeholders. The purpose of the stakeholder anal-
ysis discussed in this chapter is to evaluate which constituencies have most
at stake and how their cooperation can be secured by implementing the req-
uisite governance safeguards. Informed prioritization is essential to stake-
holder analysis.
In ­chapter 5, we analyze the for-​ profit/​nonprofit and public/​ private
distinctions from a governance perspective. A nuanced analysis reveals
that from the point of view of efficient governance, these distinctions, al-
though important, are not as central as one might think. This is because most
68  Governance within and across Organizations

nonprofits generate an economic surplus (just like for-​profits do) and most
public organizations are in fact public-​private partnerships. Consequently, in
addition to understanding and evaluating the general organizational form,
the designer must analyze the organization’s governance microstructure. An
examination of nonprofits and public organizations also helps explore the
central boundary conditions of the efficiency approach.
 3
Contracting within and
across Organizations

Think of the most substantial purchase you have ever made. For most of us,
it may have been the purchase of a house or an apartment. The chances are
that both before and after the purchase, you incurred various costs directly
related to the transaction. Before the transaction took place, you searched
for suitable alternatives and negotiated with prospective sellers. You may
also have incurred the cost of implementing various safeguards to avoid ex-
posure to hazards. Symmetrically, the seller incurred various costs, such as
searching for a buyer and implementing safeguards on the seller’s side. All
the costs incurred before the transaction are ex ante transaction costs.
After you made the purchase, you may have discovered that the contract
with the seller turned out to have a number of gaps and omissions that did
not clearly assign responsibilities in the event of unexpected disturbances.
Perhaps the house you bought turned out to have hidden flaws that were not
known to you at the time of the sale. As you raise the issue with the seller, the
seller appeals to caveat emptor (“let the buyer beware”) and maintains that
the flaws could have been discovered by reasonable inspection. All the time
and the effort you and the seller spend on settling the dispute after the trans-
action has occurred results in various ex post transaction costs.
This chapter is about the costs of transacting. In the context of governance,
attention turns to a comparative analysis of alternative ways of organizing
a transaction. Because many transactions can be organized as either intra-​
or interorganizational, the designer can seek efficiency by comparing the
costs of the two alternatives and choosing the comparatively efficient alter-
native. As table 3.1 illustrates, the designer often faces the situation of having
to choose either the intra-​or the interorganizational mode of transacting.
Both options involve ex ante and ex post transaction costs that the designer
should incorporate into the decision of how to structure the transaction or
the relationship.

Efficient Organization. Mikko Ketokivi and Joseph T. Mahoney, Oxford University Press. © Oxford University Press 2023.
DOI: 10.1093/​oso/​9780197610282.003.0003
70  Governance within and across Organizations

Table 3.1.  Examples of Intraorganizational and Interorganizational

Contracting as Alternatives

Contracting activity Intraorganizational mode Interorganizational mode

Organizing business Using internal faculty Contracting with visiting

school teaching and adjunct faculty
Producing a component Using an internal division Using an external vendor to
needed in automobile to produce the component supply the component
final assembly in-​house
Protecting the Having an internal legal Contracting with an external
intellectual property of a department with the law firm specialized in
high-​technology firm requisite expertise intellectual property
Long-​haul transport of Using a private fleet of Contracting with external
goods on trucks company trucks and owner-​operators
company drivers
A real estate company Hiring internal sales Contracting with
organizing transactions representatives as salaried independent sales agents
between buyers and employees who work exclusively on
sellers commission

British economist and Nobel Laureate Ronald Coase (1937) was the first
to point out that transacting within and across organizations are in many
instances alternatives to one another. What may now seem like a trivial point
turns out to have profound implications for the governance of contractual
relationships.
In Efficiency Lens terms, contracting decisions are usually thought to
belong to the domain of management. Specifically, whether to perform an
activity in-​house or to outsource it to an external organization is a matter
of planning and coordinating the value-​adding activities. However, we
seek to establish in this chapter that oversight and risk are always relevant
in these decisions as well. Consequently, the examples depicted in table 3.1
are ultimately not merely management decisions but indeed governance
decisions that involve management, oversight, and risk. Furthermore, the
Efficiency Lens prescribes the designer to choose the comparatively efficient
governance option, which in the case of intra-​versus interorganizational
contracting becomes particularly salient; the alternative governance modes
can be compared directly to one another in terms of their total costs.
Whether a contracting activity occurs within or across organizations
is based on the legal conception of the organization. If the transacting
 Contracting within and across Organizations  71

parties are two separate legal entities (e.g., separate firms), the transaction
is interorganizational; otherwise, the transaction is intraorganizational.
Adopting the legal conception tends to emphasize some aspects of the
contracting situation while abstracting out others. Because it is in our view
instructive to understand what is being abstracted out, we begin this chapter
by contrasting the efficiency logic with three other dominant logics: power,
competence, and identity.

Four Lenses to Organizational Boundaries

In their comprehensive review and analysis of organizational boundaries,

organization scholars Filipe Santos and Kathleen Eisenhardt (2005) noted
that the legal conception of organizational boundaries readily lends itself to
an efficiency analysis. Once the organization is defined in terms of its legal
boundaries, we can analyze whether the costs associated with crossing the
boundary are so high that they should be incorporated into the decision
of whether to organize the exchange as an intra-​or an interorganizational
transaction.
In addition to the efficiency approach, Santos and Eisenhardt (2005)
presented three other ways of conceptualizing organizational boundaries
(identity, power, and competence), which could be applied to the contracting
situations in table 3.1 to bring something essential about the situation into
focus. In fact, an entire book could be written to discuss any of these three
perspectives, which prompts us to give them voice here.1
To make all four perspectives salient, we apply each in turn to examine
the question whether a business school should organize its teaching activities
by employing internal faculty or by procuring teaching from the outside by
contracting with visiting and adjunct faculty, or by some combination of the
two. We start with the identity, power, and competence logics and then con-
trast them with the efficiency logic.

1 Just like the efficiency view, the three other views are based on and informed by decades of
academic research; Santos and Eisenhardt (2005) provide an excellent summary. To those in-
terested in reading more on the three other views, we recommend the following texts as the cen-
tral intellectual contributions (contributions within each view are listed in a chronological
order): (1) Power: Thompson (1967), Pfeffer and Salancik (1978), Pfeffer (1987), Clegg, Courpasson,
and Phillips (2006); (2) Competence: Penrose (1959), Chandler (1967, 1972), Porter (1985),
Chandler (1990), Barney (1991), Kaplan and Norton (2008), Gamble, Peteraf, and Thompson (2021);
(3) Identity: Albert and Whetten (1985), Weick (1995), Hatch and Schultz (2004), Gioia et al. (2013).
72  Governance within and across Organizations

Identity Lens: What Is a Business School?

Consider a prospective master’s student contemplating the choice of which

business school to choose for his or her MBA degree. The decision is conse-
quential, because tuition fees in top MBA programs are easily the same mag-
nitude as the person’s annual net salary postgraduation. Imagine then that
the student visits a business school’s website and, perusing the faculty pages,
discovers that the primary affiliation of every faculty member is in another
organization; all classes are taught by visiting faculty. Will this inspire con-
fidence or concern? Probably the latter. A legitimate business school has its
own faculty.
We submit that both students and faculty subscribe to this view. Those of
us teaching in various business school programs prefer not to have colleagues
change from one semester to the next. Instead, we like to get to know our
colleagues, their fields of expertise, their teaching styles, and so on. We fur-
ther benefit from cooperation and collaboration when we design our syllabi
to ensure that those who teach the same course are sufficiently consistent in
their content and approaches. This need for consistency across colleagues is
a question of organizational identity, and the associated lens through which
boundary decisions are viewed could be labeled the identity lens. The identity
lens emphasizes both the collective identity of the organizational members
as well as the image the organization projects to those outside the organiza-
tion. The key question is, “Who are we?”

Power Lens: On Whom Is the Business School

Dependent and How?

The business school, as an organization, is dependent on skilled teachers

to commit their time and efforts to teach classes. To what extent will these
individuals have power over the decisions of what to teach, to whom, and
when? Will the business school want to depend on the same individuals, or
different individuals, over time? Should the business school prefer long-​term
or short-​term contracts? More specifically, does the business school want to
depend on individuals whose commitment is secured by way of compara-
tively longer-​term employment contracts, or on individuals with whom
the business school enters into comparatively shorter-​term buyer-​supplier
contracts?
 Contracting within and across Organizations  73

The form of the contract has power implications. As we pointed out in

c­ hapter 2 in conjunction with the discussion on residual rights of control,
even though business schools do not instruct their teachers on the content
and the style of their teaching, they can exercise more residual control over
internal faculty than over visiting faculty.
When organizational boundaries are approached from a power perspec-
tive, the central question concerns the degree to which the organization
becomes dependent on others and, consequently, relinquishes some of its au-
tonomy. Consider again a business school that relies heavily on contracting
its teaching out to adjuncts and visiting faculty. If the school is dependent on
short-​term contracts with external professors, how are curriculum decisions
made? Will the person who taught the organization design elective last year
come back to teach this year as well? In case the professor is not available, are
there other external professors who could be contracted to teach the same
class? Will they teach using the same syllabus as the professor who taught the
class the previous year, or will they want to teach using their own syllabus?
If they want to use their own syllabus, does the scope of the elective change?
It is clear that decisions regarding what to do internally and what to out-
source link to considerations of autonomy, and losing autonomy is a cause
for concern. These decisions are the central issues considered within the
power lens.

Competence Lens: How Can the Business School

Achieve Its Strategic Goals?

Most business school deans would likely agree that the central objective of
business school activities is to offer high-​quality education. Accordingly, the
school should balance between the use of internal and visiting faculty in a
way that secures the highest quality of education, as measured by how much
students learn and how highly they value their degree in postgraduation
surveys. Understanding these outcomes is the focus in the competence lens.
There are many ways in which business schools seek to excel in pro-
viding their students with a meaningful and useful learning experience. In
contemplating the question of how to organize teaching in particular, the
competence lens would direct our attention to the quality of the students’
classroom experience and, consequently, teacher competence both individu-
ally and collectively. Many business schools boast of their faculty, and indeed,
74  Governance within and across Organizations

faculty scholarship is one of the central metrics used in business school

rankings. No matter what business school deans may think of the validity of
business school rankings and the metrics employed, everyone acknowledges
that they are influential.
Insofar as using internal versus external faculty is concerned, the key
question from the competence perspective is, “Which option leads to better
student learning outcomes?” As we think back to all our teaching in busi-
ness schools over the past thirty years, we conclude that students do not
base the evaluations of their learning experience so much on the individual
classes they took as much as they reflect on their entire learning journey.
Therefore, whether students genuinely learn the skills that are taught in the
classroom depends on the totality of their classroom experience; the struc-
ture and sequencing of the entire degree program matters just as much as,
if not more than, the content and the delivery of any individual course they
take. Consequently, we suggest that having an all-​star roster of individual
teachers (whether internal or visiting) is no guarantee of a high-​learning out-
come. Instead, organization matters as well, which provides a segue to the ef-
ficiency lens.

Efficiency Lens: Is the Business School Well Organized

in Its Teaching?

Consider the task of organizing the school’s teaching for the upcoming
academic year. Who is going to teach which courses? How should the syl-
labus of the mandatory core MBA seminar on strategic management be
revised? How much discretion is given to individual teachers to include
their preferred topics in the common syllabus? Which electives will the
Operations Management department offer? How does the organization
ensure that the individual professors’ teaching loads meet the minimum
required for the academic year? What should the sequencing of classes in
the highly technical Business Analytics concentration be?
These are all questions of coordination, which takes time and effort. The
efficiency perspective turns attention to how this coordination is organ-
ized. To the extent coordination is hampered by delays, confusion, sched-
uling conflicts, limited availabilities, lack of commitment, teacher turnover,
self-​serving behaviors, and the like, coordination costs increase and the
 Contracting within and across Organizations  75

organization creates waste. The efficiency logic prescribes that coordination

costs and avoidance of waste be incorporated into design and governance
decisions.
It is straightforward to see how the costs of coordinating teaching can link
to the mix of internal and external faculty. Specifically, if the exact same team
of internal faculty members teaches the classes from one year to the next,
the resultant continuity introduces stability, repeatability, and familiarity
that all work to lower coordination costs. The individuals involved know one
another and one another’s competences, they have experience coordinating
(and being coordinated) at the specific school, they have likely had many in-
formal conversations about course syllabi, topics, teaching materials, and so
on. Planning is easier when those involved in planning know one another
and have participated in the process in the past. Familiarity lowers adjust-
ment costs and breeds efficiency.
At the same time, hiring internal faculty involves long-​term contracts,
which may be comparatively more costly than short-​term contracting with
visiting faculty. In the efficiency logic, all relevant costs should be incorpo-
rated into the comparative assessment of alternatives.

Efficiency Lens vs. the Other Lenses

We consider the four lenses primarily complementary rather than

competing. To be sure, efficient coordination likely ultimately contributes to
the business school achieving its objective of providing high-​quality educa-
tion to students. Further, a business school faculty with a strong common or-
ganizational identity and ethos may alleviate coordination problems arising
from subgoal pursuit and self-​serving behaviors. Finally, much like a highly
vertically integrated industrial firm that produces internally most of the
components required in final assembly, a business school that is not funda-
mentally dependent on specific individuals external to the organization for
their contributions is more likely to be able to coordinate its activities more
efficiently.
Compatibilities and complementarities notwithstanding, focusing on effi-
ciency is important, in fact, so important that we thought the topic merits the
writing of an entire book. On the question of organizing teaching in business
schools, it is our impression and experience as faculty members that business
76  Governance within and across Organizations

schools sometimes tend to apply the competence logic with zeal by paying
disproportionate attention to outcomes that students directly experience and
evaluate.
We do not want to downplay the importance of student experience but
suggest a slight redirection to organizational outcomes, if only to comple-
ment the attention to student outcomes. This redirection invites an analysis
of governance decisions and formulation of the main problem as one of ef-
ficient organization. This redirection further shifts attention from strategic
outcomes to organizational ones. Strategic outcomes (e.g., business school
rankings) and organizational outcomes (e.g., coordination efficiency) are re-
lated and likely complementary, but they are not the same thing. It seems
logical to get the organizational outcomes right before turning attention to
the strategic outcomes.
The efficiency logic does not suggest that the designer should minimize
coordination costs, only that they should be incorporated into the efficiency
analysis as part of the total cost of transacting. Coordination costs, and trans-
action costs more generally, must be considered in conjunction with all other
relevant costs. Optimization of individual cost categories leads the designer
to the familiar suboptimization trap.
In the case of the internal versus external faculty decision, a central cost
category is teacher compensation. For the sake of argument, let us assume
that an analysis of compensation costs favors contracting with external fac-
ulty. After all, in the case of an employment relationship, the employer incurs
many costs that it would forgo if the teacher was hired as an external service
provider. In addition, employment relationships tend to involve longer-​term
commitments. However, the efficiency logic suggests that this potential com-
parative cost advantage must be considered in conjunction with the potential
increase in coordination costs.
To coordinate the teaching activities of external faculty members, the
business school would likely have to implement an administrative struc-
ture to coordinate and contract with visiting faculty. To this end, some
universities have established an entire department assigned to coor-
dinating visiting faculty, led by the associate dean of visiting faculty.
Importantly, the salaries of these administrative personnel should not be
considered general overhead costs. In the spirit of activity-​based manage-
ment, the costs of administering visiting faculty should be incorporated
as a relevant cost category in the decision of whether to use internal or
external faculty.
 Contracting within and across Organizations  77

An important revelation that systematic research on the costs of organizing

has produced over the past sixty years is that transaction costs may some-
times be so consequential that they tilt the efficiency analysis from favoring
one design alternative to another.2 It should go without saying that the total
cost of organizing teaching (not an individual cost category such as teachers’
salaries) should drive the decision. However, the challenge is that because
the costs of organizing are an elusive cost category (indeed, often assigned
to overhead), designers may focus on the more salient direct cost categories,
such as teachers’ salaries. Focusing on just one relevant cost category leads to
suboptimization.
The problem of ignoring total costs is relevant to all organizations. In
many industrial supply chains, firms tend to engage in make-​or-​buy deci-
sion analyses regarding components by considering only the comparatively
salient component prices and direct production costs, thus ignoring the
costs of coordination. A comparative analysis of external-​supplier prices and
internal-​supplier production costs may point toward buying a component
from an external supplier instead of producing it internally. At the same time,
the cost of managing the buyer-​supplier relationship may sometimes offset
this comparative advantage.

The Costs of Economic Exchange

Suppose you walk into a grocery store with the intent of engaging in a simple
(spot market) contract by which you exchange $3 of your wealth for a carton
of milk. You enter the store, get the milk from the refrigerator, pay $3 at the
register, and walk out.
The transaction in this example is so simple that we seldom stop to think
about it. Simplicity stems from two factors that most of us take for granted.
One, prices are salient. We know how much a carton of milk is supposed
to be priced at retail, and those who do not can find out at a negligible cost.
In economic terms, the price system works to the buyer’s advantage in the
transaction. Two, in most countries we can trust that the quality of dairy
products is intact. In the United States, for example, both federal and state

2 Economic research on coordination costs has been conducted mainly under the rubric of trans-
action cost economics (Williamson 1975, 1985, 1996). Some of the central academic publications that
have taken stock of the empirical research results include Shelanski and Klein (1995), Rindfleisch and
Heide (1997), Silverman (2002), Macher and Richman (2008), and Cuypers et al. (2021).
78  Governance within and across Organizations

governments have established and maintain safety regulations for dairy

products; violations are heavily sanctioned (e.g., Sumner and Balagtas 2002).
The system of institutions supports the exchange as well.
Because of the price system and the system of institutions, there is little
uncertainty associated with the grocery store transaction. For all practical
purposes, the cost of the transaction itself is negligible. Of course, you pay
$3 for the product, but that is the cost of the product, not of the transaction.
Finally, even if upon arriving at home you realize the milk you purchased
is past its expiration date, the transaction is easily reversible and the ex post
problem of inferior quality readily remediable. There is nothing transaction-
ally complex about buying a carton of milk.
The dairy product transaction looks simple to us because it is supported
by an institutional structure. Those who live in the developing countries, or
travel in them, know that simplicity is in fact only ostensible. Specifically,
absent the proper system of institutions, buying even dairy products may be-
come so complex and risky that many will refrain from purchasing and con-
suming them. In economic terms, the market fails: Even though both supply
and demand exist, the transaction will not take place. Market failures are al-
ways cause for concern, which is why a general understanding of the price
system and the institutional environment warrants the designer’s attention.
Consider in contrast a situation where the exchanging parties cannot fully
rely on the institutional pillar and must address contracting as a matter of pri-
vate ordering, thus relying on the contractual pillar instead (see ­chapter 1).
As an illustration, let us briefly put ourselves in the position of a purchasing
manager of an automobile manufacturer seeking a supplier for ten thousand
automatic transmission assemblies, priced at roughly a thousand dollars per
unit. The assemblies are make-​and model-​specific and must be designed
and engineered to model specifications. This means neither their precise
quality nor their exact price is necessarily known ex ante. Furthermore, the
purchasing manager may or may not have prior experience with the pool of
candidate suppliers. In this situation, both the buyer and the supplier expose
themselves to a potentially significant contracting hazard.

Ex Ante and Ex Post Costs

All contracting costs have a comparatively salient ex ante and a compara-

tively opaque ex post side. Ex ante costs consist of all the costs the contracting
 Contracting within and across Organizations  79

parties incur as they engage in negotiation, drafting, and agreeing upon the
contract that will be used to govern the transaction. Transacting parties often
incur ex ante costs even before they find one another. Indeed, various search
costs can be significant, even in seemingly simple situations, such as trying to
find a good lawyer, piano instructor, or therapist.
If the two transacting parties are legally separate entities, the relationship
will likely be governed by a formal contract, which involves both legal and
managerial effort and expertise. All the attention allocated to the transaction
upfront counts as an ex ante transaction cost. The ex ante cost can be thought
simply as the cost of “setting things up.”
Ex post costs involve all the costs that occur after the inception of the con-
tractual relationship. These costs link to the daily execution of the contract,
monitoring, enforcing, renegotiation, and conflict resolution. Conflict reso-
lution may range from the comparatively inexpensive renegotiation and joint
problem-​solving to the more expensive forms that involve third parties: me-
diation, arbitration, and litigation. In cases where contractual parties to the
dispute act in bad faith, ex post transaction costs may skyrocket.
Finally, terminating a contract may also have a cost, particularly if after
contract termination, one or both parties need to find a replacement ex-
change partner. Just think of the hassle associated with replacing your lawyer,
piano instructor, or therapist. Potential switching costs should be considered
ex post transaction costs as well.
When both ex ante and ex post transaction costs are incorporated into a
comparative efficiency analysis of alternative governance structures, the de-
signer may well realize that the costs are so significant that they have mate-
rial implications for the decision. For example, a make-​or-​buy analysis based
solely on the comparison of supplier prices and internal production costs
may favor outsourcing a component, but a fuller analysis that incorporates
both production and transaction costs may well make insourcing compar-
atively efficient. Consequently, it is not surprising that organizations find
outsourcing decisions not to lead to the kinds of cost savings that were
envisioned. Higher transaction costs associated with outsourcing offers a
plausible explanation.

Ex Ante and Ex Post Costs as Design Choices: The Case of Insurance

We counsel the designer not to take ex ante or ex post costs as givens but, in-
stead, treat them as matters of choice. As a general rule, it is more efficient
to prevent a problem from occurring than having to address it after it has
80  Governance within and across Organizations

occurred. Purchasing insurance offers a salient example of ex ante and ex post

problems, and an illustration of how the designer can turn an ex post problem
into an ex ante problem. In the following, we first discuss the general notion
of ex ante versus ex post costs, and subsequently, ex ante versus ex post aspects
of transaction costs in particular.
Obtaining a comprehensive insurance coverage policy for one’s automo-
bile ensures that no matter what undesirable event happens, be it collision,
theft, or a tree falling on the car, the economic consequences will be man-
ageable. An insurance policy effectively turns the ex post cost of a potentially
unmanageable economic loss into a more palatable ex ante cost of having
to pay monthly insurance policy payments. Predictably enough, since most
of us are both risk-​and loss-​averse, our preference is to address significant
economic loss as an ex ante problem. We may have only a vague idea of the
probability of getting into a car accident, but we take out an insurance policy
as a safeguard, just in case.
In other insurance situations, we accept the ex post problem as manage-
able and effectively self-​insure instead of shifting the risk of total loss to the
insurance company. When we purchase a smartphone, how many of us pay
the additional $50 for an extended warranty that covers hardware failures
and offers an express replacement service after the standard one-​year war-
ranty has expired? In this case, many of us find the risk reasonable to bear
ourselves. A total loss of a smartphone is something most of us are willing
to self-​insure, and the manufacturer’s standard warranty is sufficient to safe-
guard the transaction.
The logic of insurance effectively illustrates the ex ante and ex post aspects
of an economic problem, and how an ex post problem can, if desirable, be
converted into an ex ante problem. In addition to effectively establishing ex
ante and ex post costs as partial substitutes, a brief examination of the trans-
action costs involved in the insurance case is useful.
In the insurance case, the ex ante transaction costs would consist of the
costs of drafting and approving the insurance policy and ascertaining the
value of the insured assets. As anyone who has taken out an insurance policy
knows, these costs are, for all practical purposes, negligible. Indeed, it is in
the best interest of both the insurer and the insured that the ex ante process of
insuring property is not overly cumbersome.
In contrast, those who have filed insurance claims for substantial losses
know painfully well how significant ex post transaction costs may be. For ex-
ample, consider a car accident that involves personal injuries and where the
 Contracting within and across Organizations  81

settlement one person receives is paid through another person’s insurance

policy. Disagreements and bad faith make the situation even worse and may
result in litigation in which massive ex post transaction costs are incurred.
In fact, not only litigation but also other forms of dispute resolution such
as arbitration can be cumbersome, risky, and costly: Legal scholar Thomas
Stipanowich (2010) dubbed arbitration “the new litigation” due to the fact
that over time, arbitration has lost some of the benefits it had as an alternative
dispute resolution (ADR) mechanism.
The reason we self-​insure smaller losses is twofold. One reason is that the
economic consequence is manageable. Why insure something you can, in
some sense, afford to lose? But the other is that filing an insurance claim takes
time, and the outcome is uncertain. How much are we willing to spend in
terms of ex post transaction costs, which typically consist of having to spend
time filing claims? At any rate, this effort should be considered jointly with
the size of the prospective settlement. In some cases, self-​insurance may
present a more efficient alternative.

When Ex Post Costs Skyrocket: The Case of the Faulty

Ignition Switch

Ex post transaction costs consist primarily of the effects of unexpected events

not foreseen at the inception of the contract. Sometimes these unexpected
events are so consequential that the ex post transaction costs may not only
skyrocket but also seriously strain the contractual relationship. Nowhere is
this more evident than in the infamous case of the faulty ignition switches in
General Motors (GM) automobiles.
Delphi Automotive, a manufacturer of ignition switches, used to be a
part of GM. In 1999, it was spun off in an IPO. As an independent company,
Delphi continued its operation as an external supplier to GM. In early 2014,
GM had to recall 800,000 of its small cars due to faulty ignition switches.
The problem turned out to have catastrophic consequences: The faulty ig-
nition switches were ultimately credibly linked to more than one hundred
fatalities and hundreds of injuries. GM ended up paying over $2 billion in
fines, penalties, and settlements.3

3 https://​www.nyti​mes.com/​2017/​04/​24/​busin​ess/​supr​eme-​court-​gene​ral-​mot​ors-​ignit​ion-​flaw-​
suits.html
82  Governance within and across Organizations

Fully acknowledging that all economic ramifications pale in compar-

ison with the human tragedy involved, an examination of the economic
consequences can be instructive. Whereas poor-​quality inputs are always
cause for concern due to their direct cost consequences, we want to turn at-
tention here to the complications that arose from the supplier and the buyer
of the ignition switches being two separate firms.
Had Delphi still been a part of GM at the time of the recall, the situation
would have been comparatively simpler. For one, there would have been no
need to litigate whether the final assembler or the part supplier was at fault,
because both would have been part of the same legal entity. Even though it
would have been of little consolation to the victims and their families, having
GM and Delphi be part of the same firm would at the very least have paved
the way toward swifter dispute resolution. For example, the victims would
not have had to face the choice of whom to litigate, GM or Delphi. Some sued
GM, others Delphi, yet others named the two as codefendants.
Furthermore, whatever disputes GM had with Delphi could have more ef-
ficiently been addressed within GM as matters of internal management and
oversight instead of cross-​organizational negotiation or litigation. In this
sense, organizations have important quasijudicial functions that expedite in-
ternal conflict resolution (Williamson 1975, 30). Finally, there would have
been just one CEO, GM’s Mary Barra, appearing before Congress in 2014 to
speak on behalf of both GM and its internal supply division Delphi. In sum-
mary, the GM-​Delphi example shows that many potentially massive transac-
tion costs are ex ante hidden, because they do not manifest until something
unexpected happens. Uncertainty has profound implications for the ex post
costs of transacting.
Why did GM and Delphi not safeguard against uncertainty by vertical in-
tegration? Why did GM decide to divest Delphi? The main reason cited for
the divestment was that other auto manufacturers refused to contract with
Delphi as long as it was part of GM. It might have been transactionally more
efficient for GM not to divest Delphi, but this would have had drastic adverse
consequences to Delphi’s revenue. Governance decisions are never made in
isolation.
Did the problem with ignition switches have something to do with the fact
that GM and Delphi were separate companies? Evidence suggests that GM
as an organization exhibited a history of failing to address quality problems.
At the same time, there is no evidence that the problems with the ignition
switch were caused by the introduction of a legal boundary between GM and
 Contracting within and across Organizations  83

Delphi, and that if Delphi had been an internal GM division instead of an

external supplier, the quality problems would have received more attention.
In fact, external vendors are often scrutinized more rigorously than internal
suppliers. It seems implausible that the quality problems were caused by the
fact that the transaction was interorganizational; negligence seems like the
more plausible explanation.

The Importance of Oversight

A key lesson in the GM-​Delphi case pertains to oversight. As we noted
earlier, even though make-​or-​buy decisions are often thought of as matters
of management, oversight is relevant as well. The faulty ignition switch
case importantly illustrates how questions of oversight may be decisive.
Understanding the aftermath of the faulty ignition switch case requires an
understanding not only of the management but also the oversight of the GM-​
Delphi relationship.
In the GM-​Delphi case, the central question of interest was ultimately not
who designed and manufactured the part (issues of management) but specif-
ically how oversight was exercised. Even though Delphi not only manufac-
tured the switch but also owned the associated intellectual property rights,
GM was in charge of design specifications and approved both the final part
specifications and its installation into GM automobiles. During a congres-
sional hearing in 2014, Delphi CEO Rodney O’Neal maintained that GM
should be held responsible, because it had a central oversight role in ap-
proving the design.4
As another example where a technical failure can plausibly be linked to
oversight, consider the explosion of the Challenger space shuttle in 1986.
To attribute the disaster to the failure of the solid rocket booster sealing—​
this was the immediate cause—​overlooks the organizational failures that
preceded the technical failure. The Rogers Commission that investigated the
disaster described it not as a technical failure but as “an accident rooted in
history” (Rogers et al. 1986, 120).
There is indeed a parallel to the ignition switch case in that there were
concerns that the infamous “O-​rings” used in the rocket booster sealing
(supplied to NASA by an external supplier) were not necessarily flawed in

4 Hearing before the Subcommittee on Consumer Protection, Product Safety, and Insurance of the
Committee on Commerce, Science, and Transportations, United States Senate Hearing 113-​715, July
17, 2014.
84  Governance within and across Organizations

their design but unsuitable to be used in the specific operating conditions.

Once again, attention turned to oversight: Who approved the design of the
O-​rings and their installation into the space shuttle? Who approved the
launch?
The general governance question in the Challenger case pertains to the
rules and procedures that govern the launch decision process at NASA.
Another investigation into the Challenger disaster by the U.S. House of
Representatives Committee on Science and Technology echoed a more gen-
eral concern: “There is no clear understanding or agreement among the var-
ious levels of NASA management as to what constitutes a launch constraint
or the process for imposing and waiving constraints [and that] it is not al-
ways clear who has authority or responsibility” (Fuqua 1986, 144). It is also
instructive to observe that the report pointed to problems in the incentive
structures of buyer-​supplier relationships: “The [supplier contract] is a cost-​
plus, incentive/​award fee contract [which] provides far greater incentives to
the contractor for minimizing costs and meeting schedules than for features
related to safety and performance” (Fuqua 1986, 144). The target in these
critiques was not the specific decisions associated with the launching of the
Challenger but, rather, the organizational rules and principles that governed
decisions and contractual relations at NASA more generally—​in short,
matters of oversight. Understanding the key distinction between the man-
agement and the oversight of a contractual relationship is central.

The Determinants of Transaction Costs

Contracting across an organizational boundary tends to be more costly than

contracting inside the boundary. Complications arise from the fact that an
organizational boundary tends to be also a legal boundary, which means the
contracting parties in an interorganizational transaction represent different
legal entities. The representatives of separate legal entities are further con-
tractually obligated to act in the best interest of their respective organiza-
tions, which may cause friction in the relationship. This friction is attenuated
if the contracting parties are part of the same legal entity, such as the same
parent corporation.
The reason transaction costs should be incorporated into the analysis
is that the comparative efficiency of intraorganizational contracting may
be offset if the external supplier or service provider enjoys a productivity
 Contracting within and across Organizations  85

advantage due to economies of specialization and economies of scale. A case

in point, Bridgestone and Continental are more efficient in car tire produc-
tion than any automaker. In 2020, Bridgestone’s tire-​related revenue was $27
billion, which means it produced several hundred million tires that year. If
GM produced a set of tires for each of its seven million cars it sold annually,
the total number of tires would be only 28 million, an order of magnitude
smaller than Bridgestone’s production volume. Due to economies of scale
and economies of specialization, interorganizational contracting may be pre-
ferred even in cases where transaction costs are higher.
A fundamental question is whether the cost of transacting can be so
high that the comparative advantage of intraorganizational contracting
exceeds the external supplier’s productivity advantage, thus suggesting
that internalizing an activity would be more efficient even if productivity
was lower. To address the question, the designer must analyze the drivers
of transaction costs. The extensive research literature on the governance of
transactions has revealed three primary drivers: frequency, uncertainty, and
specificity. Note that all three are characteristics of the exchange relationship,
not the contracting parties.

Frequency of Transacting
Transaction costs tend to increase as a function of the number of transactions.
Importantly, this is not merely a matter of volume but, specifically, of the
number of transactions. For example, a supplier delivering a total of one
thousand units to the buyer once a month is different than the same sup-
plier delivering 250 units per week. When frequency increases, so does the
need for planning and scheduling, as well as the hazard of delays and other
disruptions.
Even though frequency is in many ways the most salient of the three
drivers, its effect on governance decisions is elusive. Higher frequency means
higher cost, but at the same time, it can also justify investment in more com-
plex and specialized governance structures. For example, suppliers often
appoint key account managers to handle the most important customer ac-
counts. This is analogous with the idea that large production volumes may
justify investments in automation. Williamson (1985, 60) elaborates: “[T]‌he
cost of specialized governance structures will be easier to recover for large
transactions of a recurring kind.”
The link between transaction frequency and the appropriate governance
decision remains tenuous. In their extensive review of the research literature
86  Governance within and across Organizations

on transaction costs, organization scholar Jeffrey Macher and legal scholar

Barak Richman (2008) noted that not only has transaction frequency re-
ceived less attention than uncertainty and specificity, but also the effects of
frequency on the governance mode are contested. These mixed empirical
results are plausible because, on the one hand, an increase in the frequency
of transactions provides a large “shadow of the future” in which contractual
parties have an incentive to meet their contractual obligations so that they
do not lose promising future business transactions, which can favor efficient
interorganizational contracting. On the other hand, increasing frequency,
dedicated investments, and more costly governance can be justified, which
can favor intraorganizational governance.

Uncertainty
Increasing uncertainty tends to make intraorganizational contracting more
efficient due to comparatively lower ex post transaction costs. However, in-
stead of thinking of uncertainty in the general sense, the designer must un-
pack the concept and look at its diverse sources and manifestations. Here, we
highlight uncertainty of three different kinds: technological, behavioral, and
demand uncertainty.
Technological uncertainty has to do with changes and developments in
technology that are often difficult to anticipate. In the case of two firms en-
gaging in R&D collaboration, for example, technological uncertainty is likely
the most significant reason why the collaborating firms must adapt over
time. Technological uncertainty is driven by the simple fact that the pace and
the direction of innovative activities are unpredictable.
Behavioral uncertainty suggests that it is generally impossible for one ex-
change party to predict how the other party will behave in an unforeseen
circumstance that the contract does not cover. Behavioral uncertainty can
occur not only because of self-​serving and opportunistic behaviors but also
because of honest disagreements as the contractual parties fail to converge in
their expectations. Because of behavioral uncertainty, there is a need for con-
tractual safeguards. Safeguarding against the unpredictable directs attention
to situations in which the contracting parties find themselves in unforeseen
and unprecedented circumstances that are outside the scope of their con-
tract. How does one contracting party know how the other will behave in
a situation that has not occurred before? How does the party know how it
will itself behave in an unprecedented situation? Such unpredictability is the
 Contracting within and across Organizations  87

essence of behavioral uncertainty and can lead to significant ex post transac-

tion costs.
Finally, buyer-​supplier relationships are affected by demand uncertainty.
A case in point, demand for automobiles is significantly affected by factors
outside the control of the exchange parties; input prices, interest rates, and
general consumer confidence are but a few examples. The number of seats
that GM will need from its seat supplier Adient in any given month or year
is not only variable but also unpredictable. Demand unpredictability leads
to various maladaptation problems such as order cancellations (Williamson
1983, 526), which requires contractual safeguards.
The research evidence of how uncertainty affects governance modes is
mixed as well, which is likely due to the fact that different researchers look
at different dimensions of uncertainty. Furthermore, different dimensions
associate with the decisions in different, context-​dependent ways. For
example, some forms of technological uncertainty may point to the
intraorganizational mode as comparatively efficient, but other forms, such
as the possibility of technological obsolescence, may have just the opposite
effect.5
The other reason for the mixed evidence is that uncertainty tends not
to operate on the governance mode independently of other factors. For
instance, uncertainty presents a contractual hazard primarily in contexts
where specificity is present as well. We therefore counsel the designer not
to draw any conclusions based on uncertainty alone. More generally, any
one-​factor-​at-​a-​time approach to governance decisions is likely to mis-
lead the designer. Even though the designer can benefit from thought
exercises of the consequences of “tweaking one factor at a time,” govern-
ance decisions call for an examination of the organization in its entirety.

Specificity
Whenever contracting parties commit to something that makes them de-
pendent on one another, specificity builds up. For example, an industrial

5 Strategy scholars Srinivasan Balakrishnan and Birger Wernerfelt (1986) suggested that the effects
of technological uncertainty must be analyzed by disaggregating technological uncertainty into its
constituent subdimensions. They proposed that even though increasing technological uncertainty
tends to be associated with a higher likelihood of transacting within the organization (i.e., vertical
integration), technological obsolescence has the opposite effect: As the likelihood of technolog-
ical obsolescence increases, the expected benefits of the investment decrease, as do the benefits of
vertical integration. Therefore, when the likelihood of obsolescence increases, interorganizational
transacting may become comparatively efficient.
88  Governance within and across Organizations

supplier may build its subassembly plant next to the customer’s final assembly
plant. In making the relation-​specific investment, the supplier commits to
specificity: The economic value the subassembly plant generates would suffer
greatly should the exchange relationship terminate. Specificity is about the
difference of the value of an asset, skill, or activity in its best versus second-​
best use.
Specificity takes many different forms (Williamson 1985). The location-​
specific investment is an example of site specificity. The supplier may also
commit to physical asset specificity by investing in tools and technologies
that can only be used to serve the specific customer. Employees in turn may
commit to human capital specificity by developing organization-​specific
skills.
Specificity links directly to the cost of switching from one contracting
party to another. Low switching costs alleviate the contractual hazard
arising from uncertainty, because undesirable behavior by the other
contracting party can be remedied by switching suppliers. In the taxi ride
example in c­ hapter 1, the passenger need not worry about the behavioral
uncertainty of a specific taxi driver. If the passenger has an unpleasant
experience with a particular driver, all one has to do is to use different
drivers in the future. The likelihood of the same driver picking up the same
customer again is very low anyway. The same applies to drivers having an
unpleasant experience with a particular passenger. In sum, competitive
markets produce efficiency because with many alternative suppliers and
customers, “large numbers” provide a sufficient safeguard and curb op-
portunistic behaviors.
Of the three drivers of governance decisions, asset specificity is
corroborated by strong and unambiguous evidence. In their review
of the research literature, Macher and Richman (2008, 42) noted that
incorporating asset specificity has significantly improved our under-
standing of why contracting parties favor intraorganizational contracting
to serve the purpose of efficient governance. Here, understanding that
contracting parties would be wise to provide mutual commitments to be-
come bilaterally dependent due to relation-​specific investments is central
for maintaining efficient governance. We therefore counsel the designer
to give careful attention to specificity. At the same time, specificity should
be considered in conjunction with other relevant factors, most notably
uncertainty.
 Contracting within and across Organizations  89

The General Governance Decision

Whether a parts supplier should be an independent company or an internal di-

vision of the final assembler is the canonical example of the make-​or-​buy deci-
sion. The efficient governance decision is one that is lower in its total cost, where
all ex ante and ex post costs relevant to the transaction are incorporated.
The make-​or-​buy decision is but one manifestation of the general question of
how to govern a contractual relationship. There are numerous variations on the
theme of efficient contracting, and making connections is useful. Williamson
(1994, 86 [emphasis added]) elaborates:

[Seeking efficiency] is mainly responsible for the choice of one form of capi-
talist organization over another. It thereupon applies this hypothesis to a wide
range of phenomena—​vertical integration, vertical restrictions, labor organ-
ization, corporate governance, finance, regulation (and deregulation), con-
glomerate organization, technology transfer, and, more generally, to any issue
that can be posed directly or indirectly as a contracting problem. As it turns out,
large numbers of problems that on first examination do not appear to be of a
contracting kind turn out to have an underlying contracting structure.

In this section, we explore in detail what this general contracting structure

is and how it can be applied.
Table 3.2 gives examples of contracting situations that range from man-
aging buyer-​supplier relationships and R&D alliances to the organization
of the legislature (Ketokivi and Mahoney 2017). All these examples involve
central governance decisions that two (or more) contracting parties must
collectively make. We further propose that contracting efficiency is relevant
concern in all these decisions, only the manifestations and drivers differ.
Table 3.3 summarizes the efficiency criteria and efficiency drivers for each
example.
In the following sections, we discuss four of the examples in Table 3.3 in
detail: vertical integration, corporate diversification, R&D collaboration,
and the legislature. We cover the corporate governance example in detail in
­chapter 4 in conjunction with the discussion of stakeholder analysis. We in-
vite the reader to engage in a similar reflection and analysis of the remaining
two, franchising and corporate finance.6

6 Rubin (1978) discussed the franchising example and Williamson (1988) corporate finance.
90  Governance within and across Organizations

Table 3.2.  A Selection of Governance Contexts with the Corresponding

Governance Questions and Contracting Partiesa

Context Representative governance question Contracting parties

Vertical Which components should a Buyer and supplier

integration manufacturer outsource and which
should it make in-​house?
Franchising Should the franchising fee that the Franchisor and
franchisee pays to the franchisor be based franchisee
on franchise revenue or profit?
Corporate Which individual businesses should the Horizontally
diversification corporation own? equivalent firms
Corporate Who should have a seat on the The firm and its
governance corporation’s board of directors? stakeholders
R&D When should R&D collaboration Alliance partners
collaboration between two firms be organized as a joint
equity alliance instead of collaborative
contracting?
Corporate Which assets are financed through debt The firm, lenders, and
finance and which through equity? providers of equity
Legislature What kinds of cooperation and exchange Members of the
should occur among members of legislature
Congress?
aAdapted, with some modifications and additions, from Ketokivi and Mahoney (2017). Reprinted
with permission from Oxford University Press © 2017.

Vertical Integration

Which components should the final assembler produce in-​house, and which
should it purchase from external suppliers? Should a firm have its own legal
department to handle intellectual property issues, or should it contract with
an external law firm? Should a university hire internal faculty or contract
with adjunct and visiting faculty? More generally, how should the organiza-
tion approach the make-​or-​buy decision?
In contemplating the make-​or-​buy decision, two cost categories become
relevant. One is the cost of performing the activity itself: producing a com-
ponent, providing legal advice on an intellectual property dispute, teaching
a course at the university, and so on. Comparing the production costs of the
feasible governance options is a salient part of the analysis.
The more elusive part is the analysis of the cost of contracting. In indus-
trial production, if the parts needed in final assembly are standardized and
 Contracting within and across Organizations  91

Table 3.3.  Examples of Governance Decisions, Efficiency Criteria, and

Efficiency Drivers

Context The efficiency criterion Efficiency drivers

Vertical Is the degree of vertical integration set Productivity and

integration such that the total cost of governing the coordination
buyer-​supplier relationship (the sum of
production costs and transaction costs) is
minimized?
Franchising Do the franchisor and the franchisee Incentive alignment
have the proper incentives to create value within and across
through the operations each manages and organizations
controls? Is free riding discouraged?
Corporate Do the individual business units create Economies of
diversification collectively more value when they are scope within the
divisions of the same corporation, organization
as opposed to operating as separate
businesses?
Corporate Does the composition of the corporation’s Cooperation of
governance board of directors secure the long-​ the organization’s
term cooperation of the corporation’s stakeholders
stakeholders?
R&D To manage uncertain interfirm Flexibility to address
collaboration collaboration, is a joint equity alliance unanticipated
a more flexible governance mode than disturbances
collaborative contracting?
Corporate Does the mix of debt and equity financing Cost of capital
finance minimize the cost of capital?
Legislature Do members of the legislature, individually Alignment of
and collectively, serve the interests of their legislative work with
constituencies? constituency interests

alternate suppliers are available, then purchasing the parts would amount
to little more than “ordering them from a catalog.” The issue becomes more
complicated for customer-​specific parts that require customer-​specific en-
gineering. Parts are no longer simply picked from a catalog. Instead, they
are designed, redesigned, and exchanged in collaborative long-​ term
relationships. Car seats and entire car interiors are good examples in the con-
text of automobile assembly. These relationships include various relation-​
specific investments, which are usually comparatively easier to manage if
the buyer and the supplier are divisions of the same firm. Internal disputes
are alleviated by the fact that they are subject to corporate intervention.
In contrast, if the transacting parties are separate firms, such managerial
92  Governance within and across Organizations

intervention is not possible. Furthermore, in the case of internal transactions,

there is also a strong incentive to address problems through internal organ-
ization because litigation is not an option—​courts do not hear or settle in-
ternal disputes.
The obvious designer’s mistake is to engage in suboptimization by
basing the decision exclusively on a comparison of production costs.
Analyses that incorporate only the direct costs of production tend to point
to outsourcing as the more efficient option, because specialized suppliers
often enjoy economies of scale and scope not available to the buyer. But be-
cause transaction costs in interorganizational transacting are often higher
than in intraorganizational transacting, failing to incorporate the costs of
contracting may mislead the designer to consider outsourcing preferable
even when it is less efficient in terms of total cost. Indeed, many outsourcing
decisions have not led to the kinds of cost savings first envisioned. It is pos-
sible that the differences in contracting costs are significant enough to offset
any productivity advantages.7
Finally, we should note that the comparative analysis does not require
make and buy to be the only options; an efficiency analysis can incorporate
hybrid governance forms, such as leasing. In their study of shoemaking ma-
chinery, organization economists Scott Masten and Edward Snyder (1993)
concluded that the manufacturer of shoemaking machines should structure
its contracts with those who use the machines (the shoemakers) as long-​
term leases. The reasoning was that a lease would help avoid possibly costly
haggling over prices and free-​riding concerns on both the buyer and the
seller side. Leasing the machines instead of buying them would be more lu-
crative to the shoemaker, because “[r]‌eceiving payment up-​front, a manufac-
turer has no interest, beyond its reputation, in the ultimate performance of
its product” (Masten and Snyder 1993, 42). But if the equipment were leased
and maintenance performed by the party that designed and built the equip-
ment, these performance concerns would be alleviated.
Making the lease long term would be beneficial to the equipment man-
ufacturer who has invested significant resources in the design and further
development of the machines. Masten and Snyder (1993, 38) noted that at
the time, United Shoe was one of the five largest patent holders in the United

7 The comparison should also incorporate potential differences in the quality of internally
produced and externally procured parts. But since we are operating under the assumption that make
and buy are feasible alternatives, the premise is that in both cases, quality is adequate. If this is not the
case, then the obvious choice is the option where quality is adequate.
 Contracting within and across Organizations  93

States. A long-​term lease would serve as a safeguard to the manufacturer

because having to commit to a long-​term lease would incentivize the shoe-
maker to ensure the machines would also be used.
In summary, a long-​term lease would strike a balance between the interests
of the two contracting parties: “[L]‌easing makes the incentives to develop
and support quality machinery largely self-​enforcing and thus avoids the
practical limitations of contractual guarantees” (Masten and Snyder 1993,
42). The long-​term lease would thus offer a similar built-​in high-​powered
incentive on the seller side we observed in the taxi ride example in c­ hapter 1.

Corporate Diversification

Volkswagen Group consists of the Automotive Division and the Financial

Services Division. The Automotive Division is further divided into the
Passenger Cars Business Area, the Commercial Vehicles Business Area, and
the Power Engineering Business Area. Finally, the Passenger Cars Business
Area consists of the individual brands such as Volkswagen, Audi, Škoda,
Seat, Bentley, and Porsche (Volkswagen Group 2020).
Why does it make sense for Volkswagen Group to house all these different
car brands—​horizontally equivalent activities in economic terminology—​
under the same corporate umbrella? In 2020, Volkswagen Group sold a total
of one million Audis for a total of €50 billion of revenue. What, if anything,
would be different if Audi operated as an independent automaker?
It should not come as a surprise that individual car brands are not man-
aged independently of one another within Volkswagen Group. Because pas-
senger car brands have similar technological needs such as fuel efficiency,
significant amounts of R&D effort can be shared and leveraged across in-
dividual brands that “not only work with each other, but also for each other
on key technologies, forming cross-​brand networks of expertise to address
topics of importance for the future” (Volkswagen Group 2020, 145 [emphasis
added]).
Obviously, nothing would stop Audi and Volkswagen from collaborating
even if they did not belong to the same corporation. Therefore, the central
governance question is not whether the brands should collaborate but, rather,
whether the collaboration should be organized within the same firm or across
independent firms. The efficiency view suggests that due to the highly inten-
sive and idiosyncratic needs for collaboration that likely involves frequent
94  Governance within and across Organizations

joint problem-​solving, sharing of classified information, and proprietary in-

tellectual property, bringing both brands under the same corporate umbrella
is comparatively efficient. To be sure, having a legal boundary in between two
.

car brands would not prevent collaboration, but it might cause significant
inefficiencies due to the contracting hazards it would present. Williamson
(1979, 250) elaborates the efficiency logic: “The nonstandardized nature of
these transactions makes primary reliance on [interorganizational] gov-
ernance hazardous, while their recurrent nature permits the cost of the spe-
cialized [internal] governance structure to be recovered.” Interpreted in the
context of Volkswagen Group, the complex corporate governance structure
is justified because the advantages it bestows more than offsets the costs,
resulting in net gains.
Note also that the question here is not whether the individual car brands
enjoy a competitive advantage in the marketplace or whether Volkswagen
Group’s individual business units and car brands are profitable. The ques-
tion is squarely on the efficiency of coordination: Are horizontally equiva-
lent activities more efficiently structured as intra-​or interorganizational
relationships? Whether the efficiency Volkswagen Group gains from
internalizing horizontally equivalent activities has competitive implications
would require an analysis that is outside the scope of this book. Also, com-
bining horizontally equivalent activities is not always beneficial. In the auto
industry, Volvo Cars seems to be doing much better as an autonomous car
brand than it did when it was a division of Ford Motor Company.

R&D Collaboration

Consider a scenario where two technology companies with partially

overlapping interests seek the best way to structure joint R&D. Two
alternatives are collaborative contracting and a joint equity alliance. In the
former, the relationship would be governed through a series of contracts that
specify each party’s rights and responsibilities. In the latter governance alter-
native, a separate legal entity—​a joint limited liability company—​would be
established. Both companies would make contributions to its equity and ap-
point members to its board of directors. Which alternative is comparatively
efficient?
Let us start at the main problem the designers of the collaboration are trying
to solve. In the case of risky R&D collaboration, a common way of framing the
 Contracting within and across Organizations  95

governance problem is in terms of how the contracting parties will address is-
sues that may arise unexpectedly during the collaboration. If the relationship
is governed by collaborative contracts, addressing issues that the contracts do
not cover presents a recurring contractual challenge. If these emergent issues
are frequent, the contracting parties may find themselves committing lots of
resources and attention to negotiations and exception management.
Under conditions of high uncertainty, the joint equity alliance has sev-
eral benefits. For example, the contracting parties need not write complex
contracts, because they can assign problem-​solving responsibility to the
alliance’s board of directors which is entrusted to address problems when-
ever they arise. Furthermore, because board members have a fiduciary duty
of loyalty and care to the alliance, the contracting parties can expect fewer
agency problems. Finally, contributions to joint equity can be thought of as
safeguards by which the contracting parties signal their strong commitment
to the collaboration (Gulati and Singh 1998).
In summary, whereas governance efficiency in the vertical integration cen-
tered on production and coordination costs, in the case of R&D collabora-
tion efficiency pertains to the ability to address unanticipated developments
in interfirm collaboration. A joint equity alliance is the comparatively effi-
cient response under conditions of high uncertainty. If uncertainty is lower,
as may be the case in short-​term collaborations, collaborative contracting
may be preferred.

Organizing the Legislature

In their insightful economic analysis of the legislature, political scien-

tist Barry Weingast and management scholar William Marshall (1988)
concluded that exchange in the United States Congress was organized in a
seemingly efficient way. But what does it mean for a legislature to be efficient?
It is useful to start at the fact that lawmakers have considerable incentives
to exchange support with one another. Indeed, legislators often actively seek
“trading partners” to further the interests of their own constituencies. The
question is not whether the legislature makes smart decisions and passes
high-​quality laws but, rather, whether legislators, individually and collec-
tively, are able to introduce legislation that serves the interests of their respec-
tive constituencies. The question is not one of quality of the decisions but,
rather, the alignment of legislative work with the interests of the electorate.
96  Governance within and across Organizations

The exchange of votes runs into severe problems in daily legislative prac-
tice. The issues of interest of two “trading partners” do not come up for a vote
simultaneously, which makes packaging bills into a single “market exchange”
infeasible. When trading is nonsimultaneous, how can the legislator whose
turn is to deliver first trust that the other will deliver later? Opportunism
aside, what if an unforeseen circumstance, such as the failure to get re-​
elected, led to a situation where the other party was simply unable to fulfill its
side of the bargain? Enforceability of bargains becomes problematic, and the
simple market form of exchange becomes inefficient.
The ubiquitous committee system found in legislatures around the world
provides a comparatively efficient alternative, because it provides protec-
tion against uncertainties associated with market exchange. In many ways,
the legislature functions more like a firm than a market. Just like the R&D
department of an industrial firm focuses on product and service develop-
ment tasks and the sales and marketing departments on revenue creation,
the United States House Committee on Energy & Commerce controls the
agendas within its jurisdiction by deciding which bills to bring to the House
floor for a vote.
An important function of the committee system is to introduce stability
that enables comparatively more efficient legislative bargaining. Although
it is by no means perfect (no governance alternative is), it can be argued
to be more efficient than simple market exchange, because it addresses the
nonsimultaneity problem. If all trades had to occur simultaneously in the
very same bill, the ability of legislators to serve their constituents would be
significantly hampered. The stability that the committee system provides
contributes to efficient organization.

Conclusion: Toward Understanding

the Entire Organization

Suppose we analyze all the organization’s contractual relationships and ex-

amine whether they are structured as intra-​or interorganizational. In figure
3.1, there are a total of 16 tasks, seven of which are conducted within the
organization and nine by external actors. For the sake of argument, let us
assume T1–​T7 represent seven different R&D teams of a research lab that
collaborate both with one another as well as with nine other research teams
(T8–​T16) belonging to external organizations.
 Contracting within and across Organizations  97

Figure 3.1  Organizational boundary and the scope of the organization

Some of the relationships are more intensive than others in terms of the
collaboration requirements. Perhaps the relationship T4↔T6 requires in-
tense daily communication, joint problem-​solving, and sharing of facilities
for the collaboration to be successful; the double-​headed arrow indicates re-
ciprocal interdependence. In contrast, the relationship T3→T4→T5 may rep-
resent a series of supplier relationships where T4 requires inputs from T3 in
order to supply T5; the single-​headed arrows indicate sequential interdepend-
ence. Finally, T1 may be responsible for simply pooling the contributions
from three external suppliers T8, T10, and T11; the set of single-​headed
arrows leading to T1 (T8→T1, T10→T1, and T11→T1) indicate pooled in-
terdependence. The three types of interdependence were first introduced to
the analysis of organizational relationships by sociologist James Thompson
(1967).
98  Governance within and across Organizations

The efficiency logic predicts that one is likely to find the most intensive
interdependencies in intraorganizational relationships. Importantly, inten-
sity is not merely a function of interdependencies stemming from the work-
flow (the conventional view) but also by interdependencies that have their
roots in the organization (our view), that is, transaction frequency, uncer-
tainty, and relation-​specific investments. Indeed, not only reciprocal but also
pooled and sequential interdependencies in the workflow can be associated
with strong reciprocal organizational interdependencies. For instance, the
relationship T3→T4→T5 may be comparatively simple in terms of its se-
quential workflow interdependency, but there may be organizational factors
that necessitate the internalization of all three activities. All three activities
might involve proprietary knowledge possessed only by the organization’s
best internal experts.
At the same time, some contractual relationships may be characterized
by strong interdependencies and still remain interorganizational. Task T14,
for example, may be a task that is infeasible for the organization to inter-
nalize, because it relies on a technological competence the organization
could not internalize even if it were desirable. These relationships may give
rise to contracting hazards, and therefore, require special attention from the
designer.
The implications of figure 3.1 are profound. Having mapped all the relevant
tasks, their interdependencies, and their contracting modes, what emerges is
an understanding of the entire organization. Specifically, having conducted
an efficiency analysis of all the relevant contractual relationships, we will
have made sense of not only individual contractual relationships but also the
scope of the entire organization. Some organizations have broader scopes be-
cause many transactions are so complex that they are prohibitively expen-
sive to manage through interorganizational contracting. Others have narrow
scopes due to the relative ease at which contractual relationships across or-
ganizational boundaries can be governed. To understand contracting is to
understand the organization in its entirety.8

8 The organization economist would say that we have arrived at a theory of the firm. The purpose of
a theory of the firm is to explain why organizations exist in the first place and what determines their
scope (Holmström and Tirole 1989).
 4
Stakeholder Analysis

One of the authors of this book chaired the board of an industrial startup.
In his discussions with factory workers, it dawned on him that employees
had no idea how equity operated in a limited liability company. The workers
had never owned shares, and understandably, the notions of residual
claimancy and residual risk as well as the distinction between fixed and re-
sidual payments were alien to the young engineers and technicians. The
workers did not know how fundamentally vulnerable shareholders were due
to being entitled only to residual payments. The employees had understand-
ably analyzed their own employment risk, but acquiring an understanding
of shareholder risk led to productive conversations regarding the broader
notion of stakeholder risk and, in particular, how important it was for the
startup to maintain a buffer of positive shareholders’ equity; not to be distrib-
uted to the shareholders but to ensure the small, risky organization’s survival.
The conversations were not only about informing employees on the ec-
onomic realities of a high-​risk startup. One of the prospective employees
taught the board an important lesson about how becoming an employee
would effectively commit him to site specificity (see c­ hapter 3). The man,
in his early thirties, had two children ages four and six. Becoming an em-
ployee at the plant would require that he either commute a total of sixty miles
every day from his hometown to the plant or relocate his family to the small
town where the plant was located. The former would effectively have added
an hour and a half to his workday. However, the latter would have meant that
his family would commit to their six-​year-​old child starting school in a new
town. It is reasonable to view this decision as the employee committing to
specificity, even though it was only indirectly related to the man’s potential
employment contract. The man took the job, commuted for a few months,
and then resigned. It was fully understandable that he was reluctant not
only to accept employment at a high-​risk startup but also to commit to site
specificity.
Unfortunately, many stakeholder conversations are about constituencies
engaging in advocacy to protect their interests. In an attempt to counter

Efficient Organization. Mikko Ketokivi and Joseph T. Mahoney, Oxford University Press. © Oxford University Press 2023.
DOI: 10.1093/​oso/​9780197610282.003.0004
100  Governance within and across Organizations

purely self-​interested behavior on all sides, we present in this chapter a frame-

work aimed at helping the designer conduct an explicit stakeholder anal-
ysis by a comparative assessment of the vulnerabilities of the organization’s
constituencies. A central objective of a dispassionate analysis is that the
organization’s stakeholders become aware not only of their own risk but also
the risk that other stakeholders bear.
It is perhaps realistic to assume that self-​interest will likely continue to
dominate stakeholder conversations. However, our objective is to transform
hardball advocacy into informed, reasoned advocacy. To this end, we empha-
size throughout this chapter both the importance of analysis and what legal
scholars Margaret Blair and Lynn Stout (2001, 404) labeled other-​regarding
behaviors, not only by the designer but by all constituencies and stakeholders.
The problem with the purely self-​interested approach to stakeholder manage-
ment is that a self-​interested constituency will always find a way to argue for a
stakeholder status. Other-​regarding constituencies, in contrast, will evaluate
their own position by an explicit comparison to other constituencies. This
comparative analysis goes to the very essence of value-​creating cooperation.
In addition to downplaying advocacy, another objective is to overcome
the unfortunately common problem in stakeholder conversations: the adop-
tion of excessively expansive approaches that tend to be “so broad as to be
meaningless and so complex as to be useless” (Orts and Strudler 2002, 218).
In many conversations, the word stakeholder is simply used as rhetorical
shorthand to convey that someone, or something, is important. To counter
this lack of nuance, in addition to analyzing the stakeholder status of the
organization’s constituencies, we also examine specific governance responses
available to the designer in safeguarding the most important stakeholder
relationships. Awarding stakeholder status to a constituency is meaningless
unless it is associated with a governance response of some kind.
Most fundamentally, a nuanced analysis is needed for stakeholder govern-
ance to become practically relevant for the designer. To this end, we start this
chapter by making the distinction between an organization’s stakeholders
and its constituencies more generally. We evaluate the potential stakeholder
status of a constituency by asking whether a constituency has voluntarily
put something at stake by joining the organization. Legal scholars and busi-
ness ethicists Eric Orts and Alan Strudler (2002, 217–​18) crystallized what it
means to be a stakeholder of an organization: A stakeholder “holds a bet or a
wager on the outcome of an enterprise.” Our version of stakeholder analysis
adopts a similar approach.
 Stakeholder Analysis  101

To be sure, some constituencies of the organization can make the

claim of having voluntarily put something more at stake than others.
Stakeholder analysis is therefore ultimately a matter of analyzing degrees, not
categories: Some constituencies have more of a stakeholder status than others.
This comparative approach paves the way toward informed prioritization of
stakeholders, which is essential to governance decisions.
Designers face numerous dilemmas in prioritizing constituencies,
and a dispassionate look at stakeholder interests reveals fundamental
incompatibilities; a benefit to one stakeholder may come at the expense
of another. In fact, if stakeholder interests were compatible, there would
be no need for stakeholder analysis, or stakeholder governance for that
matter. Instead, the designer would simply select the governance alterna-
tive that benefits all stakeholders. Such idealized thinking amounts to “a
mischaracterization of reality” (Bebchuk and Tallarita 2020, 129).
Only partially overlapping and at times conflicting interests consti-
tute a stakeholder dilemma, which presents the designer with governance
alternatives each of which has at least one undesirable consequence. The
designer must resist attempts to escape difficult choices and trade-​offs
by appealing to the idea of balancing stakeholder interests. Jensen (2002,
251) bluntly described balancing as a “hurrah word” that “cannot ever substi-
tute for having to deal with the difficult issues associated with specifying the
trade-​offs among multiple goods and bads.”
This chapter echoes Jensen’s call and counsels the designer to approach
stakeholder issues analytically. The greatest threat in failing to prioritize is
that the organization risks its credibility in the eyes of those who have the
most at stake. In many organizations, those who have put the most at stake
tend to be both critical to the viability of the organization and particularly
difficult to replace. A stakeholder approach that either deliberately or inad-
vertently ends up trying to pacify those with less at stake poses an immediate
threat to viability.

Stakeholders vs. Other Constituencies

We consider any individual or collective that has a contractual relationship

with the organization a constituency. Constituencies have duties toward the
organization, and for the organization to be credible in the eyes of the con-
stituency, the organization must offer something in return. Barnard (1938)
102  Governance within and across Organizations

wrote about the importance of organizations offering all its constituencies the
proper inducements in exchange for the contributions the organization asks
in return. Barnard (1938, 93) further linked inducements and contributions
directly to organizational efficiency and viability: “[E]‌fficiency of an organ-
ization is its capacity to offer effective inducements in sufficient quantity to
maintain the equilibrium of the system. It is efficiency in this sense and not
the efficiency of material productiveness which maintains the vitality of or-
ganizations.” Barnard’s position is fully consistent with the aims of this book.
In the case of most constituencies, inducements and contributions can
be adequately addressed and secured in formal contracts: employment
contracts, buyer-​supplier contracts, licensing agreements, leases, and the like.
The need for a more nuanced analysis arises when constituency relationships
involve nontrivial degrees of risk, which gives rise to the need to implement
additional safeguards to secure the cooperation of the risk-​bearing constit-
uency. Some of these safeguards can be enfolded into the formal contract.
For example, employee job security may be introduced by the employer
committing to a long-​term employment contract.
However, sometimes all formal contractual arrangements fall short of pro-
viding the requisite inducements. Some relationships involve so much risk
that adequately safeguarding them must extend beyond formal contracts and
become embedded in the very design of the organization. The relationship
between the limited liability company and its shareholders offers the canon-
ical example.

The Shareholder as a Stakeholder of the Limited

Liability Company

Inducements in the case of employees take the form of fixed salary payments
once or twice a month. If the employer defaults on these payments, employees
have multiple alternatives for recourse. If negotiations fail and the employer
still refuses to pay, the employees can take the employer to court. If litiga-
tion is ineffective, they can take the corporation to bankruptcy; upon liqui-
dation of the corporation’s assets, employees would be among the first to be
compensated. In bankruptcy proceedings, not only the firm but also its top
executives and board members could be held liable.
Bankruptcy would, of course, ultimately have adverse consequences for
the employees as well, but the central governance point is that due to strong
 Stakeholder Analysis  103

institutional forces bolstering employment contracts, the employer is ex ante

strongly incentivized to maintain the contractually guaranteed inducements to
its employees. In the case of organizational demise, everybody loses, including
those making decisions on behalf of the organization in its management and
oversight functions. This should properly incentivize management to ensure
that the organization makes all its contractually required fixed payments.
Consider in contrast the shareholders. What contractual safeguards do
they have? More fundamentally, what kind of a contract do they have with
the firm in the first place? The firm is asking for financial contributions in
the form of equity financing, but what inducements does it offer in return? If
these inducements are not maintained, what recourse do shareholders have?
Whereas employees are contractually guaranteed to receive fixed
payments from the top line of the income statement, shareholders receive
only residual payments from the bottom line in case there is something left
over. It is illegal to pay dividends unless all other contractual obligations are
met; if there is no residual, there can be no dividend. Organizations cannot
give any guarantees to shareholders that they will receive residual payments.
This risk makes shareholders a contractually disadvantaged and a highly vul-
nerable constituency—​shareholders have essentially no protection.
In the absence of a formal contract, the organization must embed in its
governance structures the requisite inducements to shareholders to secure
their contributions. One common safeguard is to make the board of directors
a governance instrument of the shareholders. Specifically, the organization
signals credibility toward shareholders by appointing a board of directors
whose primary task is to secure the rights of the residual claimants. Therefore,
although the corporation can make no promises of residual payments, it can
assign its board the role of ensuring that residual interests are incorporated
into governance. The vulnerability of the shareholder has prompted a gov-
ernance response, and therefore, the shareholder has been converted into a
stakeholder possessing a safeguard.
Buying shares in a limited liability company is an obvious bet or wager on
the outcome of the organization, but there may be others. What are they?
What bets and wagers have others made on organizational outcomes? Do
employees make a bet when they sign the employment contract and ex-
change their time and effort for fixed salary payments? Do suppliers make a
wager as they agree to supply products and services to the buyer? The aim of
a stakeholder analysis is to analyze all bets and wagers and work out the gov-
ernance ramifications.
104  Governance within and across Organizations

Consistent with Orts and Strudler’s (2002) idea that the bets and wagers
are made specifically on organizational outcomes, we offer residual risk as
the starting point of a stakeholder analysis. Just like the assessment of risk in
general, the analysis must incorporate the magnitude of residual risk each
stakeholder bears; the associated comparative analysis provides the founda-
tion for stakeholder prioritization. If designers start at the premise that a con-
stituency either is or is not a stakeholder, then the predictable outcome is that
every constituency is classified as a stakeholder, and all the designers’ work
is still ahead of them; if everyone is important, then no one is important.
Prioritization of COVID-​19 vaccine delivery provides a useful illustration.

COVID-​19 Vaccine Delivery: Efficient and

Transparent Prioritization

It is difficult to disagree with the general principle that the COVID-​19 vac-
cine should be offered free of charge to everyone who wants it. However,
this principle is analogous to the idea that everyone is a stakeholder, and as
such, is useless to the designer of the COVID-​19 vaccine delivery organi-
zation. If everyone is entitled to the vaccine, who should get it first? Should
vaccinations be given on a first-​come, first-​served basis? This approach
would be not only unethical but also potentially dangerous.1 Not surpris-
ingly, we are not aware of a single country, state, municipality, or other ju-
risdiction where a detailed, reasoned, and transparent prioritization was not
implemented and clearly communicated to the public.
It seems prudent to prioritize healthcare professionals, essential workers,
senior citizens, and individuals with compromised immune systems, over
others. To this end, the COVID-​19 Vaccination Plan issued by Illinois
Department of Public Health (2021, 13) stipulated the following: “Local
public health jurisdictions should plan to collaborate with their regional
healthcare coalition, hospitals, long-​term care and/​or assisted living facilities,
and other potential vaccine providers that serve frontline essential workers

1 To those who believe that the first-​come, first-​served principle is efficient in situations where a lot
is at stake, we recommend entering “Black Friday crowd rushing into the mall” into a search engine
and watching a few videos of how people behave when the primary thing at stake is a discount on
a smartphone. What would happen if the first-​come, first-​served principle was applied in a matter
involving personal health?
 Stakeholder Analysis  105

in their jurisdiction to ensure full coverage of vaccine first to the designated

priority groups and then to the general public.”
In the state of Illinois, which was in our view sufficiently representative of
most jurisdictions, the first priority group consisted of healthcare workers
and long-​term care facility residents and staff; the second consisted of per-
sons over sixty-​five years old, front-​line essential workers, persons between
the ages of sixteen and sixty-​four with high-​risk medical conditions; and so
on. In all these groups, the key terms were specified in great detail to make
prioritization exceedingly transparent. For example, healthcare worker was
defined using the Centers for Disease Control and Prevention (CDC) defini-
tion as “paid and unpaid workers in healthcare settings who have the poten-
tial for direct or indirect exposure to patients or infectious materials.” Note
that this definition did not require that the person necessarily work with
patients.
Similarly, high-​risk medical conditions were listed in detail: diabetes, pul-
monary disease, cancer, and heart condition, among others. Smoking was
also included as a high-​risk medical condition, which was a source of some
controversy. We propose that the prioritization of smokers offers an excellent
opportunity to exercise other-​regarding stakeholder thinking.
It may be appealing for a nonsmoker to criticize the prioritization of
smokers. Why should people receive preferential treatment simply because
of a personal health choice they had made? The designers of the vaccine pri-
oritization principles were well aware that prioritization of smokers would
stir controversy and, consequently, fully understood the importance of
communicating an explicit justification to the public. The justification for
prioritizing smokers was based on two related issues. One was the scientific
finding that cumulative smoking exposure (measured by pack-​years) had
an empirically corroborated adverse association with COVID-​19 outcomes
(e.g., Lowe et al. 2021). The other was that vaccine delivery would be
prioritized such that the comparatively more vulnerable receive protection
first regardless of the origin of their vulnerability.
For the sake of argument, suppose that the designer had begun to clas-
sify dozens, if not hundreds, of preexisting conditions based on whether they
had plausibly resulted from a personal choice. Is obesity a personal choice?
How about diabetes, the origins of which can in part be traced to personal
choices? High cholesterol? How much more challenging would the (already
highly complex and contested) prioritization protocol become? The guiding
principle in vaccine delivery was that designers considered only the hazard
106  Governance within and across Organizations

the preexisting condition presented, regardless of its source. In other words,

vaccine delivery would be addressed exclusively as a matter of public health,
not a judgment of personal choices. Incorporating personal choices would
not only have been ethically questionable, but it would also have led to pro-
hibitively complex governance structures.
Having followed the discussions on vaccine prioritization closely, we
observed that as a general rule, other-​regarding behaviors were more the rule
than the exception. The twenty-​one-​year-​old college athlete with no respira-
tory issues readily understood and accepted that a sixty-​five-​year-​old with
asthma had more at stake. Both should have the right to receive the vac-
cine, but it was clear that the designer would incorporate into governance
decisions the fact that the sixty-​five-​year-​old asthmatic must be prioritized
over the twenty-​one-​year-​old college athlete.
We submit COVID-​19 vaccine prioritization as an instructive example of
why both explicit stakeholder prioritization and other-​regarding behaviors
are essential to governance decisions. Everyone can argue to be ultimately at
risk, but some are more at risk than others. We further propose that the ques-
tion “Who needs the organization the most?” is sufficiently analogous with
the question “Who needs the vaccine the most?” Accordingly, in evaluating
the potential stakeholder status of an employee, an investor, a supplier, or a
customer, we ask, “How would the person or entity in question be affected if
the organization ceased to exist?” Employees would lose their jobs, investors
would lose their wealth, suppliers would lose customers, customers would
lose suppliers, and communities (municipalities, states, nations) would lose
tax revenue. That some would lose more than others should be clear. For ex-
ample, some employees would be able to find alternative employment very
quickly, with little or no reduction in pay. But others might have considerable
problems finding new employment, particularly if their skills are highly or-
ganization specific. To arrive at an informed understanding of stakeholders,
the analysis must involve a detailed examination of the implications of or-
ganizational failure. This is why the focus turns specifically to organizational
outcomes.

Defining Stakeholder

Management scholar Bidhan Parmar and colleagues (2010, 412) insightfully

noted that “[r]‌ather than seeing the definitional problem [of stakeholder] as a
 Stakeholder Analysis  107

singular and fixed, admitting of only one answer, we instead can see different
definitions serving different purposes.” This is exactly why trying to formu-
late a generally applicable definition is not useful. For the purposes of our
exposition, we formulate a definition that helps the designer think of ways by
which the organization can ensure that the requisite bets and wagers the or-
ganization needs from its constituencies are in fact made. For the industrial
startup whose survival hinges on successfully raising equity to finance highly
specific technological assets, the most important bets and wagers likely in-
volve investments in equity; for organizations that rely on organization-​
specific innovation and R&D, bets and wagers may involve not only equity
financing but also employees committing to organization-​specific skills.
The failure to attract the requisite bets and wagers means the organization
underinvests in some of its central activities, which jeopardizes viability.
Once all constituencies have made all the bets and the wagers that link
to organizational outcomes, they will also have developed a residual interest
and therefore become, by definition, interested in the organization’s success.
Similarly, especially in cases where constituencies commit to specificity and
become difficult to replace, the organization symmetrically develops an in-
terest in their success. As a result, the conditions of a stakeholder relationship
(see ­chapter 1) are fulfilled: By virtue of having responsibilities toward one
another and being interested in one another’s success, the contracting parties
have become one another’s stakeholders. The magnitude or the intensity of
the stakeholder relationship can be established by examining the size of the
bets and the wagers made.
We maintain that our approach to stakeholder analysis avoids the problem
of both being too narrow and being overpermissive. To illustrate the approach,
we use in the following publicly available information to conduct a stakeholder
analysis of HP Inc., a global provider of personal computing devices and
accessories in both business-​to-​business and business-​to-​consumer markets.

A Stakeholder Analysis of HP, Inc.

In its 2020 annual report, HP (2020, 4) described its business as follows:

We are a leading global provider of personal computing and other access

devices, imaging and printing products, and related technologies, solutions
and services. We sell to individual consumers, small-​and medium-​sized
108  Governance within and across Organizations

businesses (“SMBs”) and large enterprises, including customers in the gov-

ernment, health, and education sectors.

Based on this description alone, it is straightforward to see how HP is in-

volved in a large number of diverse long-​term contracts and relationships
with employees, suppliers, customers (consumers, businesses, and other
institutions), communities, governments, investors, banks, and the like.
Indeed, HP’s 131-​page Sustainable Impact Report (HP 2019) discussed the
diversity of HP’s stakeholders in great detail. The complexity and the vast
heterogeneity of stakeholders provides us with a great opportunity for
conducting a stakeholder analysis. We first briefly discuss HP’s own ap-
proach and then extend it by our own efficiency-​based analysis.

HP’s Perspective in Brief

In its Sustainability Impact Report, HP (2019, 5) listed the following entities

as stakeholders: HP employees, suppliers and supply chain workers, local
and global communities, customers, channel and retail partners, recycling
vendors, and regulators. Interestingly, the 131-​ page report mentions
shareholders only once and does not list shareholders as stakeholders.
However, we think it is not only reasonable to count shareholders as
stakeholders—​ their wagers are unambiguous—​ but also ultimately mis-
leading to make a distinction between the two. Our position is that the stake-
holder status of all constituencies, shareholders included, has a common
origin in residual risk.
The Sustainability Impact Report also reveals that HP engages in explicit
analysis of stakeholder issues: “We identify appropriate stakeholders based
on factors such as expertise, willingness to collaborate, reputation, location,
and sphere of influence” (HP 2019, 14). Moreover, the focus is on engage-
ment: “We gain valuable insight through our regular engagement with a
range of stakeholders—​including employees, investors, suppliers, customers,
peer companies, public policymakers, industry bodies, nongovernmental
organizations (NGOs), sector experts, and others” (HP 2019, 14). This ap-
proach squares perfectly with our idea of focusing on mutual value creation
and collaboration. We want to add to this idea the importance of assessing
residual risk.
 Stakeholder Analysis  109

At HP, stakeholder issues are elevated to the top management and board
levels. Most notably:

HP’s Board of Directors’ Nominating, Governance and Social Responsibility

(NGSR) Committee oversees the company’s policies and programs relating
to global citizenship and the impact of HP’s operations; provides guidance
and recommendations to the Board on legal, regulatory, and compliance
matters relating to political, environmental, global citizenship, and public
policy trends; and reviews the annual Sustainable Impact Report. (HP
2019, 15)

Publicly available documents do not describe in detail how HP prioritizes

its stakeholders. In the following section, we use publicly available infor-
mation in conjunction with the Efficiency Lens to make inferences about
which particular relationships might be more likely to exhibit stakeholder
characteristics.

A Stakeholder Analysis of HP’s Constituencies

Prioritization of constituencies should be performed not only across but

also within constituency groups. In fact, we propose that prioritization
within categories is both more meaningful and has better tractability than
prioritization across categories. The question, “Which specific employees,
or employee groups, should be considered stakeholders?,” is more tractable
than the question, “Are shareholders more important stakeholders than
employees?” We consider the latter question ambiguous not because it is
difficult to compare an employee to a shareholder but because to compare
the two is to compare two heterogeneous groups to one another. Put differ-
ently, fine-​grained within-​group analyses must precede any between-​group
comparisons.

Suppliers
We estimate the total number of HP’s suppliers to be around a thousand, at
least in terms of the order of magnitude. It is further immediately clear from
basic accounting data that suppliers in the aggregate are crucial to HP’s busi-
ness. We estimate HP’s total annual spending in its supply chain—​purchase
of components, products, and services—​to be at the very least 40 percent
110  Governance within and across Organizations

of HP’s annual revenue.2 Not surprisingly, HP openly acknowledges that it

not only depends on third-​party suppliers, but also that its financial results
would suffer if it failed to manage its suppliers effectively, and that the “busi-
ness could be negatively impacted if key suppliers [were] forced to cease or
limit their operations” (HP 2020, 16).
The magnitude of HP’s purchasing costs clearly establishes that HP is fun-
damentally dependent on its supply chain; if purchasing costs were 4 instead
of 40 percent of revenue, we would be less concerned. However, it does not
immediately follow that suppliers, as constituents, are also stakeholders. The
idea that HP would have a thousand external organizations as its stakeholders
would lead to prohibitively complex governance structures. Consequently,
significant prioritization must take place.
What would happen if HP lost Company X as its supplier? What would
happen if Company X lost its HP account to another vendor? It is always
important to analyze stakeholder relationships through the lens of bilateral
dependency because relationships characterized by strongly unilateral de-
pendence do not qualify as stakeholder relationships. Specifically, it makes
little sense to label HP’s relationship with Company X a stakeholder relation-
ship if Company X is a supplier of standard parts that HP can purchase from
a dozen other vendors. To be sure, Company X might like to claim a stake-
holder relationship with HP, but since the arm’s-​length relationship with
Company X constitutes no risk to HP, awarding stakeholder status would
direct unwarranted attention to a simple transactional relationship, and pos-
sibly away from the genuinely complex relationships that merit the designer’s
attention.
What kind of a cost would HP incur if it had to replace Company X with
a new supplier? Supplier switching cost offers a salient metric for evaluating
the potential stakeholder characteristics of a buyer-​supplier relationship.
Furthermore, bilateral dependency would prompt us to ask the symmetrical
question, “What kind of a cost would Company X incur if it had to find a new
buyer to replace HP?” In some buyer-​supplier relationships, switching costs
are negligible; in others, they are astronomical.
HP’s annual reports provide insight into the prioritization of suppliers and
the identification of those suppliers that are likely considered stakeholders.

2 In 2020, HP’s revenue was $56 billion and cost of revenue $46 billion (82 percent of sales). Cost of
revenue consists primarily of purchases and wages of those employees working directly on the gener-
ation of sales. We hypothesize purchases to be the single most important source of costs, which is why
we estimate it to be at least 40 percent of revenue. Obviously, this is merely our own estimate.
 Stakeholder Analysis  111

A good litmus test for whether a supplier is a stakeholder is whether its name
appears on the 10-​K form.3 If a supplier is mentioned by name, its identity
matters, and a stakeholder status is more likely.
Suppliers whose names appear repeatedly on HP’s 10-​K form are AMD
(a supplier of x86 processors; mentioned four times in the 2020 10-​K
form), Canon (a supplier of laser print engines and laser printer cartridges,
mentioned eleven times), Microsoft (a supplier of software products;
mentioned five times), and Intel (a supplier of x86 processors; mentioned
four times). Not surprisingly, HP (2020, 6) openly admits to being “de-
pendent upon” these suppliers. In the case of AMD, Canon, Microsoft, and
Intel, switching costs are so high that losing these suppliers would likely
threaten HP’s survival.4
AMD’s, Canon’s, Intel’s, and Microsoft’s annual reports reveal a symmetric
dependence on HP. For example, HP is Intel’s third-​largest customer, ac-
counting for 10 percent of Intel’s $78 billion revenue in 2020. HP cannot af-
ford to lose Intel as a supplier, but we are almost certain that the feeling is
mutual.
At the other end of the supplier spectrum are scores of vendors that HP
(2020, 5) described as follows: “For most of [the products and components
we purchase], we have existing alternate sources of supply or alternate
sources of supply are readily available.” The switching costs with re-
gard to these suppliers are intentionally kept low by managing them as
market transactions instead of long-​term contracts. In general, switching
costs offer a straightforward tool for analyzing the stakeholder status of
suppliers.

3 Form 10-​K is an elaborate document each publicly listed company must file with the Securities
and Exchange Commission every year. The 10-​K form is a source of useful information on the
company’s business environment, strategy, management, and financial data. It is also useful in
conducting a rudimentary stakeholder analysis.
4 In the early 2000s, one of the authors worked with a manufacturer that performed the final as-
sembly of tractors. The manufacturer bought most of the components and subsystems from external
suppliers. One of these suppliers designed and manufactured tractor cabins, one of the most expen-
sive subsystems in the final assembly. Furthermore, there was only one supplier of cabins, which
made the final assembler highly dependent on the specific supplier. Given that each cabin was
customized, cabins had to be delivered on a just-​in-​time basis from the supplier’s factory to final
assembly. By the early 2000s, the buyer-​supplier relationship had been ongoing for fifty years, and
predictably enough, developed all kinds of relation-​specific features. When the general manager of
the final assembly operation was asked how long it would take for his factory to replace the cabin sup-
plier, the manager replied, without missing a beat: five years. He then added: “We had no option but
to put all our eggs in one basket, so my task is to watch that basket!” To be sure, five years would be
sufficient to take the company to bankruptcy many times over.
112  Governance within and across Organizations

Customers
An identical switching-​cost analysis can be applied to customers. The stake-
holder status of customers becomes salient in the HP case when we consider
the large institutional and corporate clients. For example, CDW Corporation
is a multibrand provider of information technology solutions to various insti-
tutional customers (e.g., government, healthcare, and education). A Fortune
500 company with annual sales of $20 billion in 2020, CDW is also one of
HP’s large corporate clients, and we suspect the relationship exhibits many
characteristics of a stakeholder relationship. But since HP is CDW’s supplier,
we have essentially covered this issue in the previous discussion on suppliers.
HP is mentioned several times—​and referred to as CDW’s partner—​in
CDW’s 10-​K form.5
Considering customers as stakeholders does raise an interesting additional
question not covered in the previous section on suppliers: When should the
specific customer category of consumers be considered stakeholders? The in-
tuitive answer may be always, but we propose the answer is actually rarely.
Suppose you purchase an HP laptop computer. Does this make you a
member of the HP organization? No. What obligations do you have toward
HP? None. What obligations does HP have toward you? Aside from having to
ensure that the laptop works the way it is expected until the warranty period
ends, not much else. It is hyperbole to describe a relationship with no dis-
cernible reciprocal characteristics or bilateral dependency as a stakeholder
relationship. Of course, we are fully aware that corporations often consider
not only customers in general but consumers in particular their stakeholders.
We dare suggest this is simply to convey the sentiment that consumers are an
important constituency because without consumers there will be no viable
business. To those who disagree with our position that consumers are rarely
stakeholders, we present the following challenge: Show us how consumer
interests are incorporated into governance structures, and where the impor-
tant safeguards that secure the rights of consumers are implemented.

5 On its 10-​K form for the 2019 fiscal year, CDW stated the following: “[S]‌ales of products man-
ufactured by Apple, Cisco, Dell EMC, HP Inc., Lenovo and Microsoft, whether purchased directly
from these vendor partners or from a wholesale distributor, represented approximately 60% of our
2019 consolidated Net sales [ . . . ] The loss of, or change in business relationship with, any of these
or any other key vendor partners, or the diminished availability of their products, including due to
backlogs for their products, could reduce the supply and increase the cost of products we sell and
negatively impact our competitive position” (CDW Corporation 2019, 11). We have no doubt that
it is in the best interest of both CDW and HP to safeguard the critical buyer-​supplier relationship
through bilateral credible commitments.
 Stakeholder Analysis  113

To be sure, HP must take care of its consumer customers; after all,

consumers are a significant source of revenue. In the 2020 fiscal year, HP’s
revenues from consumer hardware were $2.5 billion. But just as in the case
of suppliers, high volume must not be confounded with stakeholder status.
We see no discernible governance implications in the HP-​consumer rela-
tionship. HP’s relationships with large institutional clients are obviously an
entirely different matter, because business-​to-​business relationships often
involve significant bets and wagers on both sides. In business-​to-​consumer
relationships, it is difficult to see bets and wagers that would require the
designer’s attention.
We suggest that as a general rule, consumers are unlikely to be stakeholders
of the companies whose products and services they use. The simple reason
is that switching costs tend to be negligible both to the consumer and the
organization. We might in fact go even so far as to suggest that not being
a stakeholder is what consumers should prefer. Imagine the scenario of not
only being dependent on the use of a personal computer (as many of us are)
but also being dependent specifically on one provided by HP. Why would you
choose to commit to such specificity? What are the benefits of such “lock-​in”
to you? Will they offset the increased switching cost such that net gains are
obtained? With most consumer products, voluntarily putting something at
stake with a particular brand seems like an unnecessary wager.
HP laptops are not cheap, but it is an overstatement to suggest that by
buying one you have put so much at stake that you should be able to claim
stakeholder status. If HP ceased to exist, you would have no trouble finding
another brand of computers. Besides, HP ceasing to exist would not mean
that your HP laptop would be immediately useless. IBM laptops certainly did
not stop functioning on May 1, 2005, when IBM sold its PC hardware busi-
ness to Lenovo.
Symmetrically, it cannot possibly be surprising to learn that the owners
and executives at HP would lose no sleep over losing you as a customer.
Even though HP is dependent on both businesses and consumers buying
its products, it depends on the consumer market only in the aggregate; the
cost of switching from one particular consumer to another is negligible.
Only a mass exodus of consumers would be a cause for concern. To be sure,
this sometimes happens, but we propose that the failure to incorporate
consumers as stakeholders is not the reason.
We do not know whether HP’s leadership considers consumers
stakeholders in our meaning of the term. However, insofar as efficient
114  Governance within and across Organizations

organization is concerned, we do not see how it would be justified for HP

to elevate consumers to a stakeholder status. Symmetrically, no economic-​
efficiency rationale suggests that consumers should seek to put anything
at stake at HP. One does not need to become HP’s stakeholder to use HP’s
products, and HP does not have to consider the consumer a stakeholder to
sell its products and to take care of all its customers’ needs. The relationship
between the two is purely transactional.
The fundamental problem in thinking of the relationship between
corporations and consumers in stakeholder terms is that the relationship is
typically asymmetric. When the consumer becomes “locked in” to the spe-
cific supplier of products or services, the supplier becomes in no appreciable
way dependent on the individual consumer. This is hazardous to the con-
sumer, because the supplier can effectively hold the consumer hostage:

[O]‌nce the relationship has begun, the supplier will be isolated to some
degree from competition and will be in a position to “hold up” the con-
sumer. . . . Generally, after the consumer has entered into the relationship
with the producer, [he or she] will find [himself or herself] vulnerable to
price increases or the threat of termination; the producer will be in a posi-
tion to price discriminate in an attempt to capture [more value]. (Goldberg
1976, 439)

Employees
For the purposes of an efficiency analysis, it is instructive to consider
employees and suppliers as analogous: Employees can usefully be regarded as
“supplying” their time and efforts in return for compensation. Accordingly,
an employment contract is, in some sense, a special case of the more gen-
eral buyer-​supplier contract. Furthermore, just like in the case of analyzing
suppliers and customers as stakeholders, the focus should be on switching
costs on both sides.
Who are the employees HP cannot afford to lose because they would be
difficult to replace? These are the employees whose relationship with the
organization should be considered in stakeholder terms instead of being
viewed merely through the lens of an employment contract.
Are those who create greater value to HP more likely to be stakeholders
than those who create less value? This is not necessarily the case. The ques-
tion of stakeholder status is not about how much value an employee creates
 Stakeholder Analysis  115

but whether the employee possesses skills that are organization specific. It is
not value but specificity that makes the employee more difficult to replace.
In discussing employee skills, it is crucial to distinguish between specificity
and specialization. As an example, consider the orthopedic surgeon working
at the Johns Hopkins University in Baltimore. The surgeon is both highly spe-
cialized and highly valuable to the hospital. At the same time, the surgeon’s
technical skills are not organization specific.6 The surgeon can easily seek
employment at other hospitals, and symmetrically, the hospital may look to
hire a replacement surgeon from another hospital. Committing to speciali-
zation makes individuals profession specific, committing to specificity makes
them organization specific. Because the profession (orthopedic surgery) is a
broader entity than the organization (Johns Hopkins University), those who
commit to organizational specificity tend to be more difficult to replace, be-
cause the pool of replacement candidates is smaller.
Individuals who have committed to specialization are found throughout
the HP organization; it is probably reasonable to suggest that all HP
employees are specialized in one way or another. But where might we find
individuals who have committed to specificity? In an attempt to pinpoint the
parts of HP’s organization where specificity might be found, it is useful to
start with the question of what is distinctive about the organization.
Strategy scholars C. K. Prahalad and Gary Hamel (1990) suggested that an
organization distinguishes itself from the rest through its core competences.
For example, Prahalad and Hamel (1990, 83) described Honda’s and Canon’s
core competences as follows: “It is Honda’s core competence in engines and
power trains that gives it a distinctive advantage in car, motorcycle, lawn
mower, and generator businesses. Canon’s core [competences] in optics,
imaging, and microprocessor controls have enabled it to enter, even dom-
inate, markets as seemingly diverse as copiers, laser printers, cameras, and
image scanners.” In Prahalad and Hamel’s logic, core competences become
embedded in core products, which are then leveraged to create revenue in
the corporation’s business units.
But if the organization’s core competences make the organization unique,
then it is only logical to look for commitments to organizational specificity
within these competences. Commitments to specificity may be technological

6 Surgeons work as members in surgical teams. Over time, a fundamental transformation may
occur as team members learn to work together. As a result, some specificity may develop, and moving
to another organization would involve having to learn to work with a new surgical team. However,
the surgeon’s general technical skills, while highly specialized, are not team or organization specific.
116  Governance within and across Organizations

(physical asset specificity), but to the extent they involve employees devel-
oping and nurturing organization-​specific skills (human capital specificity),
these employees should be considered not merely constituencies but indeed
stakeholders. The organization cannot expect employees to commit to speci-
ficity without the requisite safeguards.
By applying Prahalad and Hamel’s logic, we find it plausible that in the case
of HP, human capital specificity is found in HP’s Printing Segment, and in
particular, within the units that develop graphical solutions to deliver “large-​
format, commercial and industrial solutions, and supplies to print service
providers and packaging converters through a wide portfolio of printers
and presses (HP DesignJet, HP Latex, HP Indigo and HP PageWide Web
Presses)” (HP 2020, 70). It is likely that at least some of HP’s technical experts
who work on commercial and industrial printing solutions possess skills that
are in varying degrees HP specific, because printing and imaging have al-
ways been at the core of HP’s strategy. Not surprisingly, this segment is also
where many of HP’s central patents are found. Those employees whose work
is related to HP’s intellectual property are also likely to exhibit human capital
specificity. When employees commit to the development of organization-​
specific technologies, they make an unambiguous wager on organizational
outcomes.

Financiers
Which providers of capital have a stakeholder relationship with HP, and why?
Here, it is instructive to examine the kinds of contracts financiers have with
the firm. The obvious distinction is between the providers of debt and equity
capital. Even though the debt-​equity categorization is a simplification and
there are many hybrid forms of financing that exhibit features of both (e.g.,
Pratt 2000), the distinction is still analytically useful. Specifically, whether the
financing instrument is “pure” debt or “pure” equity or some combination
of the two, understanding the essence of the contractual relationship with
the financier is central. Insofar as governance is concerned, understanding
the difference between fixed payments and residual payments becomes cen-
tral. The starting point for a stakeholder analysis is provided by “pure” debt
representing the “unqualified obligation to pay” and “pure” equity the “un-
limited claim to the residual benefits of ownership” (Pratt 2000, 1067).
With providers of debt capital, HP enters into a formal contract that
stipulates the material conditions of the debt: amount, interest, payback
schedule, collateral, and the like. Lenders will seek contractual safeguards,
 Stakeholder Analysis  117

collateral being the obvious example. Furthermore, there is also a well-​

developed aftermarket for debt instruments, which means that the lender
can transfer its contract with HP to another lender. Consequently, because
the relationship with lenders can be safeguarded through a formal con-
tract, it is difficult to see why lenders should be considered stakeholders in
any but the most extreme of circumstances. Lenders should be considered
stakeholders only if they genuinely make bets and wagers on organizational
outcomes—​this is seldom the case.
Just like in the case of suppliers, customers, and employees, we can ana-
lyze this further by asking the question, “How would the lender be affected
if HP ceased to exist?” Here, it is relevant to note that lenders receive prefer-
ential treatment over shareholders in bankruptcy proceedings. Specifically,
if HP were to declare bankruptcy, its debt obligations would not disappear,
they would only be transferred to the bankruptcy estate. Bankruptcy, there-
fore, does not mean that the lender faces an immediate loss. There are well-​
developed safeguards in place.7
Stakeholder analysis of financiers leads to a very different result when
providers of equity capital are considered. In short, shareholders are prom-
ised nothing and can, at least in principle, lose everything. If HP were to
declare bankruptcy, HP’s shares would in all likelihood be worth nothing,
which means no obligations to shareholders would be transferred to the
bankruptcy estate. Even in the case something was transferred, shareholders
would remain residual claimants, the last in line to receive benefits from the
estate. Bankruptcy proceedings seldom result in there being a residual be-
cause corporations tend not to declare bankruptcy when their net worth is
significantly positive.
The wagers that shareholders make on organizational outcomes are so
salient that it is straightforward to see why shareholders should, as a gen-
eral rule, be considered stakeholders. Furthermore, in comparison with
suppliers, customers, employees, and lenders, shareholders are the compar-
atively disadvantaged contracting party in that they are recipients of only

7 One of the authors of this book was an investor in a consumer-​products firm that declared
bankruptcy. In the bankruptcy, shareholders lost everything when the shares lost all their value;
suppliers lost as some of their invoices were left unpaid; employees lost as they did not receive all their
paychecks and the firm failed to make even the mandatory pension payments on the employer side;
customers lost as they did not receive products for which they had already paid. In stark contrast,
not a single bank incurred an economic loss. With the requisite safeguards in place, banks had fully
avoided exposure to residual risk. Of course, we do not wish to suggest banks never lose money in
bankruptcies; the point is that they not only have access to safeguards others do not (e.g., collateral),
but they are also more informed and skilled in implementing these safeguards.
118  Governance within and across Organizations

residual payments. At the same time, the problem resides in the failure to see
the other, less conspicuous wagers that other constituencies make. Designers
should be compelled to examine stakeholder issues in their entirety, which
means analyzing all wagers made, not just the immediately observable and
salient ones. Once all wagers have been identified, the designer should think
of how to safeguard them through the appropriate governance responses.

Governance Responses to Stakeholder Interests

Identifying the organization’s stakeholders is merely the first step. The second,
more challenging step, is to work out the governance implications: How
should stakeholder interests and stakeholder prioritization be incorporated
into governance? What kinds of safeguards are needed to secure the most
critical relationships that exhibit residual risk?
The enduring problem with stakeholder conversations is that they tend not
to have actionable implications for governance decisions. For example, sup-
pose the designer concludes that a group of employees should be considered
stakeholders. What are the governance implications? Should these employees
be awarded a formal role in oversight, such as representation on the board of
directors? Furthermore, does broader board participation offer a remediable
solution to an efficiency problem, that is, will net gains be realized for the or-
ganization? The designer must understand that governance is not a system
that can be “tweaked” one factor at a time; instead, individual decisions tend
to have various indirect, systemic effects.
In this section, we discuss the governance implications of stakeholder
analysis. In the discussion, we focus on the third question that circumscribes
the essence of governance (see ­chapter 2): What are the reciprocal credible
commitments that secure the continuing cooperation of the organization’s
constituencies in general and its stakeholders in particular?
The premise in structuring exchange relationships is that the contracting
parties seek a solution that aligns with the central characteristics of the spe-
cific relationship. In seeking alignment, the designer must be able to dis-
criminate, that is, to understand the efficiency implications of the feasible
alternatives in the given context. In economic terms, the designer must seek
discriminating alignment in governance decisions (Williamson 1991, 277).
We propose incorporating commitments to specificity and vulnerability
as the central characteristics of the organization’s relationship with a given
 Stakeholder Analysis  119

constituency. Those who have by virtue of joining the organization become

vulnerable have made a credible commitment toward the organization; the
organization must reciprocate by properly safeguarding the relationship.
The resultant mutual credible commitments secure the cooperation of the
constituency.
For contracting to be efficient, the relationship must have the requi-
site safeguards to secure the cooperation of the contracting parties. In
the following, we discuss the different ways of safeguarding employment
relationships, buyer-​supplier relationships, and relationships with financiers.

Safeguarding Employment Relationships

Because employment relationships are heterogeneous, they also vary greatly

in the degree of safeguarding required. This variation can be linked directly
to levels of specificity and contract duration.
Think of your first summer job, whether it was mowing lawns, selling ice
cream, or delivering newspapers. The essence of your employment relation-
ship was little more than you offering your free time in exchange for mone-
tary compensation in an intentionally transient relationship. Such a simple
contracting setup allows the contracting parties to draft an essentially com-
plete employment contract; it is hard to see justification for burdening the
relationship with additional safeguards. The only relevant governance choice
is whether a low-​or a high-​powered incentive is preferred. If you mowed
lawns, the chances are the employer (or your customers) paid you based on
the number of lawns you mowed (a high-​powered incentive). In ice cream
sales, an hourly rate (a low-​powered incentive) is probably more appropriate.
If compensation was based on the number of customers served, the worker
might mysteriously call in sick on cold and rainy days, only to miraculously
recover when the weather got warm and sunny again. Linking employee
compensation to arbitrary factors such as the weather tends to be inefficient.
At the other end of the spectrum are employment relationships where
employees commit to specificity by developing and nurturing skills that
create strong bilateral dependency between the employee and the organi-
zation. Organization-​specific skills have low redeployability, and therefore,
committing to them must be considered a form of risk—​they are bets on or-
ganizational outcomes. One way for the designer to think about discrimi-
nating alignment is to acknowledge that employees committing to human
120  Governance within and across Organizations

capital specificity have a legitimate residual interest. Accordingly, one option

is to award employees a formal residual claimant status.
In limited liability companies, employee stock ownership plans are one
alternative of awarding employees a residual-​claimant status; opening the
board of directors to the group of employees who represent organization-​
specific skills is another. The latter would effectively make some employees
not merely residual claimants but also trustees of the organization. Employees
would participate in the ratification of management decisions and would ex-
ercise oversight over their implementation. In the governance literature, em-
ployee participation in oversight is sometimes discussed under the rubric of
codetermination. Countries such as Germany have mandated codetermina-
tion for companies that have more than five hundred employees. The most
salient manifestation of codetermination is employee participation on the
board of directors.
Codetermination in Germany (and a number of other countries) is legally
mandated, but there is no reason why the designer could not incorporate co-
determination as a private-​ordering response as well if it offered an efficient
governance alternative. Williamson (1985, 303) pointed to the informational
advantages of employee representation on boards: “[Employee] membership
on boards of directors can be especially important during periods of actual
or alleged adversity, especially when firms are asking workers for give-​backs.
Labor’s board membership might mitigate worker’s skepticism by promoting
the exchange of credible information.” In this sense, employee participation
can function as an instrument for securing the credibility of management
decisions in the eyes of employees. This credibility is essential in situations
where employees commit to specificity.
Opening up the board of directors for broader participation is a “heavy-​
duty” measure that will likely have undesirable systemic effects. However,
there are also targeted options available to the designer. These more targeted
measures can be implemented at the level of both the individual employment
contract and a collection of similar contracts. In some European countries,
for example, contracting with employees is subject to externally mandated
industry-​wide standards embedded in various collective-​action agreements.
The private-​ordering counterpart to externally mandated obligations is
multiunit bargaining where a representative of employees (typically a union)
negotiates with a specific employer on behalf of an entire group of employees
whose contractual relationships with the organization are appreciably
similar.
 Stakeholder Analysis  121

For example, the general agreement between the United Automobile,

Aerospace and Agricultural Implement Workers (UAW) and General
Motors (GM) is a seven hundred-​page document that contains scores of
rules and principles that protect unionized GM employees. The agreement
stipulates, among other things, that the employer is not allowed to locate out-
side suppliers at a plant if it “conflicts with UAW assigned operations that
could cause a loss of jobs” (UAW-​GM 2015, 655). Similarly, the employer’s
discretion regarding sale of assets belonging to the bargaining units (e.g., a
specific plant) is severely restricted.8
Negotiations between employers and employees frequently involve deep-​
seated and adversarial power dynamics. However, in our attempt at un-
derstanding efficient organization, we direct the designer’s attention here
squarely to the cooperative aspects of employer-​union relationships and
private-​ordering approaches to codetermination. We choose to view these
relationships less as hardball advocacy and more as genuine attempts at
safeguarding contractual relationships under conditions of bilateral depend-
ency. We are not surprised that the need for multiunit bargaining arises in
oligopolistic industries in particular (Davidson 1988): When an industry is
dominated by a small number of employers, the chances are their key ac-
tivities involve various kinds of specificity, including human capital speci-
ficity. The automobile industry and commercial airlines are representative
examples.

Safeguarding Buyer-​Supplier Relationships

A similar specificity examination can be applied to relationships with

suppliers and customers. Even though the nature of specificity in buyer-​
supplier relationships is more likely to involve physical asset specificity and
dedicated assets instead of human capital specificity, the designer’s task re-
mains essentially unchanged: Contractual relationships that involve spec-
ificity merit the designer’s attention because they may require additional
safeguarding.

8 “[T]‌he Company will not close, idle, nor partially or wholly sell, spin-​off, split-​off, consolidate
or otherwise dispose of in any form, any plant, asset, or business unit of any type, beyond those
which have already been identified, constituting a bargaining unit under the Agreement” (UAW-​GM
2015, 356).
122  Governance within and across Organizations

Arm’s-​length buyer-​supplier relationships that involve neither specificity

nor residual risk require no additional safeguards; when switching costs
are low, the availability of alternative sources of supply provides the requi-
site protection. Symmetrically, suppliers having alternate sources of demand
for their products and services safeguards the supplier side. All additional
safeguards would be redundant and, therefore, a form of waste. For this
reason, neither the taxi driver nor the passenger needs to safeguard the con-
tractual relationship; there is little at stake and a negative experience can be
addressed by blacklisting the specific exchange partner.
Specificity changes everything. The case of shoemaking (see
­chapter 3) provides an excellent example of contracting parties seeking
discriminating alignment under conditions of specificity (Masten and
Snyder 1993). The long-​term lease arrangement between the shoemaker
and the equipment manufacturer safeguarded the high levels of physical
asset specificity. Specifically, the shoemaker agreeing to a long-​term lease
and committing to using the shoemaking machines ensured that the equip-
ment manufacturer’s massive investments in R&D would not be wasted.
The shoemaker would in turn benefit from the equipment manufacturer’s
continuing commitment to further developing the machinery. Another
benefit was that the equipment manufacturer would be incentivized to per-
form adequate maintenance on the equipment, because the manufacturer
only leased the equipment to the shoemaker. If the shoemaker had bought
the machines, the equipment manufacturer might still have performed
maintenance but would have lower incentive intensity since it would have
given up title to the machines.
Some relationships are so hazardous that the exchange is most efficiently
organized as an intraorganizational relationship. As an example, consider
the oil refining value chain (Klein, Crawford, and Alchian 1978). Suppose
an oil company owns and operates the oil fields and the refineries, both of
which are subject to considerable site specificity and physical asset speci-
ficity. Suppose further that an oil pipeline is the only way of transporting the
oil from the fields to the refineries. The oil company operating the oil fields
and the refineries might be well advised to integrate vertically into pipelines
as well; contracting with an external firm for pipeline service would expose
the oil company’s highly specific assets to an economic “holdup problem”
(Goldberg 1976) by the pipeline service company. Owning and operating
not only the oil fields and the refineries but also the pipelines would effec-
tively eliminate the holdup problem. Vertical integration is the ultimate
 Stakeholder Analysis  123

“heavy-​duty” governance mechanism for safeguarding complex transactions

that involve high degrees of specificity.
Unlike in the case of securing the cooperation of employees who exhibit
high human capital specificity, board membership is seldom considered in
the case of buyer-​supplier relationships. As the United Shoe example shows,
even significant commitments to specificity can be addressed through care-
fully tailored and specialized buyer-​supplier contracts. It is hard to think of
a situation in which a buyer-​supplier relationship should be safeguarded by
awarding the buyer an oversight role in the seller’s organization, or vice versa.
Williamson (1985, 308) elaborates: “Considering the variety of widely ap-
plicable governance devices to which firms and their suppliers have access,
there is no general basis to accord suppliers additional protection through
membership on the board of directors [ . . . ] [T]‌he governance structure that
firm and supplier devise at the time of contract [ . . . ] will afford adequate
protection.” Williamson’s point applies equally if we replace the word supplier
with the word customer.

Safeguarding Relationships with Financiers

In contrast with contractual relationships with employees, buyers, and

suppliers, it is awkward to think of contractual relationships with financiers in
terms of specificity. Neither lenders nor providers of equity commit to spec-
ificity when they finance an organization. However, they do expose them-
selves to varying degrees of financial risk. Therefore, instead of analyzing the
contractual relationships with financiers through the lens of specificity, let us
turn to the general notion of vulnerability and look at whether the financier
receives fixed or residual payments. Those who receive fixed payments re-
quire fewer safeguards than those entitled only to residual payments; in fact,
the contractual guarantee of a fixed payment is a safeguard in and of itself.
The organization’s lenders have a comparatively complete contract with the
organization. The loan agreement guarantees the creditors fixed payments of
interest and capital. The lender may also require the borrower to offer collat-
eral, such as general-​purpose equipment or the organization members’ per-
sonal property, to guarantee the loan. In case the organization defaults on the
fixed payments, lenders have powerful tools at their disposal to recover their
investments: They can take the borrower to court, make claims for the collat-
eral, or, if everything else fails, take the borrower to bankruptcy. Even in the
124  Governance within and across Organizations

event of a bankruptcy, the organization’s debt obligations will not disappear,

they will instead be transferred to the bankruptcy estate. In summary, debt
financing can usually be sufficiently safeguarded through a formal contract
that guarantees fixed payments.
Providers of equity face a completely different set of circumstances. Unlike
lenders, shareholders receive only residual payments from the organization.
Because the existence of a residual is a performance outcome subject to un-
certainty, it is essentially noncontractible: The organization simply cannot
contractually commit to a certain return on investment to the shareholders.
Furthermore, if the organization is taken to bankruptcy, only residual
claimancy (no contractual obligations) will be transferred to the bankruptcy
estate.
The relationship between the organization and providers of equity must be
safeguarded. Absent sufficient safeguards, who would ever contribute to the
equity of the organization? A common way of safeguarding the rights of the
residual claimant is through participation on the board of directors. In con-
trast, there are no compelling reasons why providers of debt financing should
be awarded board seats.

The Paradox of Stakeholder Participation

Both the specific topic of shareholder participation and the more general
notion of stakeholder participation on the board of directors have received
considerable attention in governance practice and research. Some scholars
and practitioners maintain that the board should be an instrument of the
shareholders, whereas others suggest it should be opened for broader stake-
holder participation.
In a stark departure from both views, we propose that it is fundamentally
misguided to think of board membership in terms of participation. In fact,
we submit that the very notion of participation on boards is paradoxical.
Unpacking the paradox requires that we make the distinction between stake-
holder interests and stakeholder representation.
What does the suggestion that the organization’s most important
stakeholders should be represented on its board of directors ultimately mean?
Does it mean that the largest shareholder gets to appoint their agent to the
board of directors to secure the specific shareholder’s interests? Similarly,
should companies whose employees commit to high levels of human capital
 Stakeholder Analysis  125

specificity let the employees appoint their agent to the board to ensure em-
ployee interests are incorporated to board-​level decisions?
We are constantly amazed at how casually those engaging in stakeholder
conversations and debates gloss over the fact that the only constituency the
board member represents is the organization. This is not only a governance
principle but it is also the law: The only beneficiary of the board member’s fi-
duciary duty of loyalty and care is the organization, not the constituency that
appointed the member.
It is hazardous to confound stakeholder interests with stakeholder partic-
ipation. Appointing a representative of debt financing to the board provides
a cautionary example. Are we surprised to find that when bankers are ap-
pointed to the board of directors, the firm starts leaning more heavily toward
debt financing? Governance scholars David Larcker and Brian Tayan (2015,
chap. 5) accurately noted that exhibiting such bias constitutes a violation
of fiduciary duty; what is worse, such violations are very difficult to detect.
Larcker and Tayan further noted that research results unfortunately suggest
that when bankers serve on boards, they indeed tend to behave in ways that
privilege the interests of their employers over those of the organization. This
constitutes a breach of fiduciary duty.
The paradox of stakeholder participation on boards is effectively
crystallized by the question, “Assuming board members represent their re-
spective constituencies instead of the organization as a whole, how could
they arrive at decisions that are in the best interest of the organization?” How
does a car manufacturer’s board of directors make a decision regarding plant
closure if its board of directors consists of representatives of management,
shareholders, banks, and employees, each advocating the interests of their
respective constituencies? What, if any, is the common interest that all these
stakeholders share?
Might an independent board of directors whose task is to incorporate
stakeholder interests without directly representing any of them offer the com-
paratively efficient option? Posing this question effectively introduces the
notion of director independence.

Independent Thinking in Oversight

Suppose a stakeholder group has two options. One is that it gets to appoint
a representative who is directly incentivized to promote the interests of the
126  Governance within and across Organizations

specific stakeholder; the appointed board member would therefore effec-

tively become the stakeholder’s agent on the board. The other option is that
the board consists of independent individuals whose fiduciary duty of loyalty
and care is to consider what is best for the organization. In this second sce-
nario, the board member would become a trustee of the organization.
Should the stakeholder prefer a tailored agent or a general trustee? This
choice situation makes the distinction between the board consisting of stake-
holder representatives and the board serving stakeholder interests salient.
Let us consider the question by examining shareholders. In many large
corporations, board member compensation is significantly equity-​based.
In fact, this is not only common practice, but it is also a principle that the
New York Stock Exchange Governance Guide endorses: “Currently, it
is common to have equity represent a slight majority of regular annual
compensation—​such as a pay mix of equity compensation 55 percent and
cash compensation 45 percent [ . . . ] The emphasis on equity compensation is
also directionally consistent with the typical pay mix for senior executives.”9
At first glance, equity-​based director pay seems reasonable, as it aligns
the interests of oversight and management toward increasing the value of
shareholders’ equity. In reality, this alignment may actually have just the op-
posite effect. Specifically, if both managers and directors are incentivized
through equity, directors may be less inclined to exercise oversight over man-
agement. Counterintuitively enough, a more efficient way of incorporating
shareholder interests into oversight is to make the board independent of
shareholder interests; in short, populate the board with trustees of the or-
ganization, not agents of the shareholders. Finance scholars Harley Ryan and
Roy Wiggins (2004, 500) elaborate: “[I]‌ndependence enhances shareholder
welfare since board independence results in compensation contracts that
provide directors with stronger incentives to monitor.” In other words, in-
dependent directors are more efficient in monitoring top management pre-
cisely because their interests do not align with the interests of those whose
actions they are expected to monitor. This observation reveals something
essential about the value of independent judgment: All stakeholders should
take seriously the possibility that the key to efficient oversight may be found
in independence, not in representative participation.
Union representation on boards offers another example of the paradox
of stakeholder participation. Suppose the workers’ union appoints a union

9 New York Stock Exchange Corporate Governance Guide chap. 21, p. 156.
 Stakeholder Analysis  127

representative to the car manufacturer’s board of directors. Suppose then

that the board must consider the choice of closing down a large assembly
plant located in a small town. In formulating his or her position on the issue,
will the union representative think of what is best for the organization or
what is best for the workers? The law requires the former, but the union
representative’s own constituency would favor the latter. The resultant di-
vided loyalty is so profoundly problematic that few organizations open
the board of directors to employee participation. Instead, the relationships
with high-​specificity employees are addressed as a combination of targeted
individual-​level contracting and multiunit bargaining.
To be truly independent in judgment, a board member must be free of all
other loyalties. Consequently, the bar is set at a very high level: Independent
directors should be independent of not only the organization on whose
board they serve but also those who appoint them. Only absolute indepen-
dence secures independent judgment.
If absolute independence sounds idealistic, let us examine two examples.
Consider first the organization of the judiciary. Even though members of
the Supreme Court of the United States are appointed by the President, the
Supreme Court does not represent the President’s interests in any capacity.
Article III, Section 1 of the Constitution of the United States declares that
the Supreme Court wields the judicial power of the United States, and
therefore serves the interests of the people. As a telltale sign of what we
think comes fairly close to absolute independence in the governance sense,
we witness time and again how Supreme Court Justices vote in a way that
is completely at odds with the preferences of the President who appointed
them.10
As another example, consider Neste, a Finnish oil company organized
as a limited liability company with an annual revenue of €12 billion. Three
aspects of Neste’s board of directors merit attention here (Neste 2020): (1) all
eight board members are independent of both the company and its largest
shareholders; (2) only two of the eight board members have a nontrivial eq-
uity stake in the company (four board members own no shares at all); and
(3) board compensation is limited to modest annual retainers that range

10 In January 2022, the Supreme Court rejected former President Donald Trump’s request to
block the release of White House records to the select committee of the House of Representatives
investigating the events of January 6, 2021. None of the three conservative Justices appointed by
President Trump sided with Trump’s request.
128  Governance within and across Organizations

from €35,700 for regular board members to €67,900 for the chairperson;
there is no equity-​based pay.11
In the case of the Supreme Court, the idea that integrity must prevail over
loyalty to any individual constituency is clear. But the fact that a for-​profit
oil company would appoint a completely independent board suggests that
even in the for-​profit setting, the idea of an exclusively fiduciary (as opposed
to stakeholder or shareholder) duty is not an unreasonable proposition. We
might therefore ask, “When does integrity not merit the designer’s atten-
tion?” Applying this principle to board composition, it does not seem at all
utopian to us that individual board members should never represent indi-
vidual stakeholder interests but, rather, should have a fiduciary duty to the
entire organization.

Summary: Returning to the Main Problem

Stakeholder conversations are simultaneously frustrating and indispensable.

They are frustrating because they often regress to uncompromising advocacy
of individual stakeholder interests. We share both Orts and Strudler’s (2002,
218) criticism that stakeholder conversations tend to be “so broad as to be
meaningless and so complex as to be useless” and Bebchuk and Tallarita’s
(2020, 91) concern that stakeholder governance may offer merely an “illu-
sory promise.” At the same time, stakeholder conversations are indispensable
because designers must address the only partially overlapping and at times
inconsistent interests of those who create value for the organization. We find
overly optimistic pronouncements of compatible stakeholder interests intel-
lectually dishonest, misleading, and potential sources of significant organ-
izational waste. We must not let stakeholder governance regress to wishful
thinking, appeasement, and mere rhetoric.
The problem, as we see it, is that many stakeholder conversations tend “to
put the cart before the horse” and seek answers without being explicit about
the questions. Why do organizations need stakeholder governance? If they
do need it, should it be considered a matter of private ordering? If yes, why? If
it is a matter of private ordering, how should it be implemented?

11 In April 2022, Brand Finance (one of the leading independent brand valuation firms) ranked
Neste as the second most valuable brand in Finland with a brand value of €2.2 billion, a 20 percent
increase from 2021.
 Stakeholder Analysis  129

In their impressive and detailed review of the research literature on stake-

holder management, Parmar and colleagues (2010) presented three different
problem formulations for stakeholder management: (1) the problem of value
creation and trade; (2) the problem of the ethics of capitalism; and (3) the
problem of the managerial mindset. Even though all three are worth our at-
tention and even though the three are interconnected, we find it infeasible to
address all three in one tractable approach. Simultaneously trying to incor-
porate the managerial mindset, value creation, and the ethics of capitalism
into the same conversation is bound to lead to confusion.
In this chapter, we have implicitly adopted the objective of value crea-
tion by asking, “How does the organization secure the voluntary coopera-
tion of those constituencies who, by virtue of becoming contributors to the
organization, are asked to become economically vulnerable?” Furthermore,
we propose that it is useful to link vulnerability to voluntary, informed
commitments to specificity.
We hope this chapter enables the designer to move toward rigorous, ex-
plicit, and transparent stakeholder analysis. Among other things, our ap-
proach can serve as a guide to prioritization: Which employment contracts
require more of the designer’s attention than others? Which supplier and
customer relationships are more vulnerable than others? How are the
relationships with the organization’s financiers different from one another,
and why does it matter? Our approach offers actionable tools for the anal-
ysis of specific constituency groups. Within each group, the designer is likely
to identify contractual relationships that should be treated as stakeholder
relationships in governance decisions. Awarding stakeholder status to a con-
stituency that has nothing at stake is economically wasteful and distracts at-
tention from the relationships that merit the designer’s attention.
That some suppliers are more important than others is salient. How
about comparisons across stakeholder groups? Are shareholders a more
important constituency than suppliers? Frankly, we find such intergroup
comparisons illogical, which is why we have not discussed them in this
chapter. As we have shown, each constituency group exhibits within-​
group heterogeneity that is often sufficiently significant to have governance
implications. Some employees are stakeholders, others are not; the same
applies to suppliers, customers, and financiers. The only group where all
members can justifiably be argued to be stakeholders are the providers of
equity financing, because they are all, individually and collectively, vulner-
able due to their status as residual claimants. It does not, however, follow
130  Governance within and across Organizations

that because all shareholders are stakeholders, their interests supersede

the interests of those stakeholder groups where less than 100 percent of
constituents are stakeholders.
Comparing shareholder interests to the interests of other stakeholder
groups in the aggregate is ultimately a case of the composition fallacy. A com-
position fallacy occurs when one mistakenly attributes to a whole a charac-
teristic that applies only to some of its individual members, and then draws
conclusions from or acts based on this misattribution.
Designers who are unable to incorporate nuance into governance
decisions may fall prey to the composition fallacy. A one-​size-​fits-​all ap-
proach to stakeholder governance will result in excessive safeguards in some
relationships and insufficient safeguards in others. Designers must under-
stand that high-​specificity employees require a different contracting ap-
proach than low-​specificity employees, that fixed payments and residual
payments cannot be handled by the same governance structures, and that al-
though some suppliers should be considered stakeholders, the vast majority
probably should not.
The approach proposed in this chapter hopefully offers a remedy to
the composition fallacy. Instead of adopting a one-​size-​fits-​all approach,
designers should analyze each contractual relationship by identifying its
critical characteristics (giving special attention to contractual hazards and
vulnerabilities) and then seek discriminating alignment by devising tailored
safeguards to secure the cooperation of the contracting party throughout the
duration of the contract.
 5
Nonprofit and Public Organizations

Nothing in the preceding chapters, or the subsequent ones, suggests that ef-
ficiency is an exclusive concern of organizations that seek profits. To make
this argument clear, we devote an entire chapter to discussing efficiency in
nonprofit and public organizations. We start this chapter by making two im-
portant distinctions: (1) for-​profit versus nonprofit, and (2) public versus
private. These distinctions are not as straightforward as one might think,
and a closer look at how governance structures are designed is required. We
then discuss the nonprofit theater as an example of governance in a non-
profit organization. Finally, the context of public organizations (or public-​
private partnerships) offers an opportunity to examine some of the crucial
assumptions and boundary conditions of efficiency analysis. For example,
the idea of net gains requires a number of assumptions that are not met in
some public organizations. This insight links to the idea that the outcomes of
governance decisions may not be commensurate, and consequently, an anal-
ysis of net gains is impossible. In such contexts, the idea of governance as ef-
ficiency may have to yield to governance as integrity (Williamson 1999, 340).
Later in the chapter, we discuss the efficiency/​integrity distinction in light of
examples from contexts where efficiency analysis is infeasible or, at least, of
secondary importance.

The Key Distinctions

Figure 5.1 shows a two-​ by-​two matrix with four example organiza-
tions: (1) Finnair, the majority-​state-​owned Finnish commercial airline;
(2) the University of Illinois, a public US university; (3) Caterpillar Inc., a pri-
vate corporation incorporated in the state of Delaware; and (4) Real Madrid
Club de Fútbol, a sports club in Madrid, Spain. We use these four examples as
illustrations as we make distinctions between for-​profit versus nonprofit and
public versus private organizations.

Efficient Organization. Mikko Ketokivi and Joseph T. Mahoney, Oxford University Press. © Oxford University Press 2023.
DOI: 10.1093/​oso/​9780197610282.003.0005
132  Governance within and across Organizations

Figure 5.1  Examples of for-​profit, nonprofit, public, and private organizations

For-​Profit vs. Nonprofit

Both for-​profits and nonprofits often seek a surplus, only their motivations
differ. The for-​profit seeks a surplus to provide sufficient investment returns
to the residual claimants to secure their continuing cooperation; the non-
profit seeks “to provide a reasonable cushion or reserve against a rainy day or
provide for future growth plans of the organization” (Gross, McCarthy, and
Shelmon 2005, 15). In fact, even some of the motives for seeking profits are
similar, as both for-​profits and nonprofits may be interested in a surplus to
provide for future growth.
As illustrations of nonprofit organizations, we intentionally selected two
that show both a substantial economic surplus in their income statements
and a significant net worth on their balance sheets: the University of Illinois
(net worth of $4 billion), and Real Madrid Club de Fútbol (net worth of
€533 million). In fact, nonprofits that produce a surplus and have net worth
are easier to find than those that do not, which is consistent with legal and
economic scholar Henry Hansmann’s (1980, 838) observation that “[m]‌any
nonprofits in fact consistently show an annual accounting surplus.” What is
central for governance purposes is how the surplus is governed.
Let us first examine the annual revenues and expenses of the University
of Illinois, a public nonprofit organization. Revenues consist of student
tuitions and fees, grants and contracts, state appropriations, investment in-
come, and private gifts, among other sources. Expenses consist of payments
to employees, employee benefits, payments to suppliers, scholarships, and
fellowships, among others. In the fiscal year 2020, the University of Illinois’
 Nonprofit and Public Organizations  133

revenues exceeded expenses by $237 million, increasing the university’s net

position to about $4 billion. It is useful to think of the net position in balance-​
sheet terms as analogous to shareholders’ equity. The University of Illinois’s
net position is held in the form of capital assets: treasury bonds and bills,
asset-​backed securities, equity funds, and money market funds. In sum, the
University of Illinois is a public, nonprofit organization that has revenue,
produces an economic surplus, and has net worth.
A private nonprofit organization is similar in many respects. Real Madrid
Club de Fútbol is one of the largest sports organizations in the world. In 2019,
it produced an after-​tax surplus of €38 million from a revenue of €757 mil-
lion. This annual surplus is added to the organization’s net worth (or equity).
In 2019, Real Madrid’s net worth was €533 million. Because Real Madrid is
a cooperative owned by its 91,000 members, or socios, this equity belongs to
the organization—​the socios are not residual claimants.
Despite seeking and producing an economic surplus and having net
worth, both the University of Illinois and Real Madrid are nonprofit or-
ganizations: Their net worth belongs to the organization and cannot be
appropriated by private actors. Indeed, one of the central tasks of oversight in
a nonprofit organization is to prevent such appropriation.
Rules that regulate the governance of net worth have both public-​and private-​
ordering properties. In the spirit of this book, our focus is on the latter, but since
understanding the governance of net worth in its entirety is important, both
merit attention. In the case of the University of Illinois, the $4 billion net posi-
tion consists of three main categories of net assets: restricted nonexpendable,
restricted expendable, and unrestricted (and expendable) net assets (see Gross
et al. 2005). Expendable assets can be used for their intended purposes whereas
nonexpendable assets must be held in perpetuity. Furthermore, restricted as-
sets are subject to externally mandated restrictions. These constraints imply that
only unrestricted assets effectively belong to the category of private ordering
in the sense that they may be allocated to specific purposes at the discretion of
management (by observing the rules set by oversight).
At the University of Illinois, both university management and oversight
are responsible for “the university’s reputation and strong financial posi-
tion” (University of Illinois 2020, 12). No private appropriations can be made
from the university’s assets. Importantly, this restriction does not arise from
the university being a public but a nonprofit organization, “barred from
distributing its net earnings, if any, to individuals who exercise control over
it, such as members, officers, directors, or trustees” (Hansmann 1980, 838).
134  Governance within and across Organizations

The situation is appreciably similar for Real Madrid. Because the socios
are not residual claimants, their only prerogative is that they are entitled to
“enjoy the club’s activities” (Real Madrid 2019, 10). Real Madrid is governed
by the General Assembly, the President, and the Board of Directors, who
collectively decide whether Real Madrid’s annual surplus is entered into
retained earnings, spent on the renovation of the Santiago Bernabéu sta-
dium, used to finance Real Madrid youth teams, or some other uses. Real
Madrid’s €533 million net worth cannot be appropriated privately.
In sum, the essential governance aspect of the nonprofit organization
stems from the fact that the organization has no residual claimants other
than the organization itself. Consequently, the nonprofit’s surplus must be
properly governed and managed by the organization to prevent both inap-
propriate allocations and expropriations. Moreover, in the case where the
nonprofit organization is awarded preferential tax treatment, efficient over-
sight of the nonprofit is no longer merely a matter of private ordering; it also
becomes a matter of regulatory compliance.

For-​profit and Nonprofit as Design Choices

The nonprofit form of organizing is more common in some contexts than
others; education, healthcare, and publishing are examples of contexts where
nonprofits are prevalent. However, we frame the question whether to or-
ganize as a for-​profit or nonprofit ultimately as the designer’s choice. Will the
designer want to impose a distribution constraint on the surplus or not? This
design choice is effectively illustrated by nonprofit entrepreneurs.
Why would a self-​interested entrepreneur choose the nonprofit form
of organizing? To be sure, the nonprofit form weakens the entrepreneur’s
profit-​seeking incentive. However, economists Edward Glaeser and Andrei
Shleifer (2001) suggested that deliberately lowering the entrepreneur’s incen-
tive intensity is the whole point. Specifically, by choosing the nonprofit form
of organizing, the entrepreneur sends a strong signal to the organization’s
constituencies that the entrepreneur is not interested in profiting—​indeed,
cannot profit—​ from their contributions. Signaling a self-​ imposed low-​
powered incentive confers several benefits: “When customers, employees, or
donors feel protected by the nonprofit status of the firm, the entrepreneur has
a competitive advantage in the marketplace” (Glaeser and Shleifer 2001, 100).
Incorporating as a nonprofit can be directly linked to stakeholder gov-
ernance (­chapter 4). By lowering incentive intensity, the entrepreneur, as
a designer, works toward securing the cooperation of the organization’s
 Nonprofit and Public Organizations  135

stakeholders. When entrepreneurs self-​impose a low incentive intensity,

stakeholders may see their own commitments to specificity as less hazardous,
because the economic holdup hazard is significantly reduced: “Employees
may invest more in specific human capital at not-​for-​profit firms because
these firms have less financial incentive to cut wages or perquisites ex post”
(Glaeser and Shleifer 2001, 101).
Similarly, information asymmetry concerns are alleviated. Those who have
comparatively limited access to information (e.g., employees or suppliers)
need not be concerned about those with better access (e.g., managers and
owners) exploiting the asymmetry.
In sum, commitment to weak incentives as a deliberate design choice
may foster efficient governance (Williamson 1999). Low incentive intensity
becomes a potential problem only if it emerges inadvertently or as an un-
desirable side effect of a governance decision in a situation in which high-​
powered incentives would be comparatively efficient. If this is the case, then
the beneficial, intended consequences of the governance decision must be
considered simultaneously with the inefficiency consequences of the unde-
sirable loss of incentive intensity (see c­ hapter 8 for a detailed discussion).
How does the self-​interested entrepreneur benefit from low incentive in-
tensity? If the nonprofit status of the organization indeed signals credibility,
the organization increases its likelihood of attracting talented and com-
mitted employees, a broad customer base, and donors as financiers. The
organization’s customers may even be willing to pay more for the products
and services of a nonprofit than a for-​profit because they know that the sur-
plus will not be appropriated by a profit-​seeking entrepreneur. Furthermore,
if the entrepreneur is also an employee of the organization, the organization
can pay the entrepreneur a salary with reasonable perquisites.
Given the considerable failure rate of entrepreneurial firms (up to 90 per-
cent by some estimates), organizing the entrepreneurial firm as a nonprofit
may provide the self-​interested entrepreneur with comparatively low-​risk
access to a steady income. In addition, the entrepreneur not only achieves
this steady income in a reasonably comfortable and meaningful environment
but may also gain the genuine respect and admiration of the organization’s
stakeholders.
The nonprofit entrepreneurial venture is, of course, subject to the
familiar drawbacks of low incentive intensity. Indeed, the price is
paid precisely in terms of the consequences of low incentive inten-
sity: “[N]‌onprofit firms might be expected to be slower in meeting
136  Governance within and across Organizations

increased demand and to be less efficient in their use of inputs than for-​
profit firms” (Hansmann 1980, 844). However, organization scholars
Akhil Bhardwaj and Anastasia Sergeeva (2022) challenge this line of rea-
soning by suggesting that a nonprofit cooperative does not necessarily
forgo high-​powered incentives, because residual claimancy is not the
only means of achieving high-​powered incentives. As always, it is the task
of the designer to engage in an analysis of which feasible governance al-
ternative is comparatively efficient.

How about Tax Minimization as the Main Problem?

Is the decision to incorporate as a nonprofit not driven primarily by the pref-
erential tax treatment of nonprofits? This question merits a closer look.
In the United States, the nonprofit is exempt from federal income taxes,
which provides “a particularly strong subsidy to the non-​profit form”
(Hansmann 1987, 79). As with all governance decisions, the situation must
be considered in its entirety. To be sure, understanding tax implications is
important, however, proposing tax considerations as the default explana-
tion for the choice of the nonprofit organizational form is an oversimpli-
fication (see also Glaeser and Shleifer 2001). In fact, we suggest that from
the governance point of view, choosing not to have a residual claimant
because of tax benefits is misguided. What if there are constituencies with
a legitimate residual interest, which should be incorporated into govern-
ance? Will the designer simply “trade in” their residual claimancy for the
tax-​exempt status? Are those asked to waive their residual claimancy still
expected to bear residual risk? Will they accept their de facto transforma-
tion from a residual claimant to a donor?
It is of course possible that the designer chooses the nonprofit form
with the intent of abusing it by providing private residual payments
through expropriation. Designers who think such abuse of the non-
profit form will pay off need to consider the potentially significant ex
post transaction costs arising from the fact that when an organization is
tax exempt, “the Internal Revenue Service may well take an interest in
whether there is any distribution of profits” (Hansmann 1980, 874). How
does the magnitude of the ex post transaction costs of having to conceal
expropriation (and the consequences of potentially failing to do so) com-
pare with the magnitude of the tax-​exemption benefits? We propose that
behaving opportunistically in choosing the organizational form is ulti-
mately myopic.
 Nonprofit and Public Organizations  137

Same Competitive Context, Different Choices

The world of competitive team sports offers a telling example of the notion
that organizational form is not fully determined by the organization’s envi-
ronment but is indeed a matter of design choice. In contrast with the two
Spanish football clubs Real Madrid and FC Barcelona, both organized as
nonprofit cooperatives, many other European football clubs are incorpo-
rated as limited liability companies. The English football club Manchester
United F.C. was founded as a closed (or close) limited liability company
that went public (i.e., became open) in 2012. Manchester United Plc’s stock
is currently traded on the New York Stock Exchange, and in 2020, paid its
shareholders $30 million in dividends. The French football club Paris Saint-​
Germain F.C. is organized as a closed limited liability company, fully owned
by Qatar Sports Investments.1
Organizational form notwithstanding, the similarities between Real
Madrid (a nonprofit) and Manchester United (a for-​profit) are striking.
Both (1) have income statements and balance sheets that are, for all prac-
tical purposes, identical in their structure; (2) have the same three sources
of revenue (commercial, broadcasting, match day); (3) rely heavily on debt
financing; (4) have a board of directors and a management team; (5) pay high
salaries to their executives; (6) seek to maintain a healthy positive cash flow
so as to generate an economic surplus; (7) attract the top football talent by
offering them generous salaries, and (8) have the identical objective of win-
ning championships, which includes beating one another in the European
Champions League.
What, then, should we make of the fact that one is organized as a for-​profit
corporation and the other a nonprofit cooperative? As far as the two organ-
izations are concerned, the material difference is that in one of these organ-
izations, the owners are residual claimants who are entitled to a return on
their investment; in the other, the owners are merely entitled “to enjoy” the
organization’s activities. This difference merits attention, but considering it
simultaneously with all the similarities, it really does not seem like these two

1 Discussing the key differences between Manchester United and Paris Saint-​Germain effec-
tively shows the value of using the words open and closed instead of private and public. Even though
Manchester United’s shares are publicly traded, it remains essentially a private firm because its
owners are predominantly private investment companies and private individuals. Using the word
open prevents confusion. Similarly, the essential characteristic of Paris Saint-​Germain is that its
shares are not publicly traded even though the organization itself is best described as public: PSG’s
owner Qatar Sports Investments is part of Qatar Investment Authority, a sovereign wealth fund.
Using the word closed prevents confusion.
138  Governance within and across Organizations

organizations are all that different. This brings us to the important conclu-
sion that we should be cautious in drawing inferences about an organization
simply based on the legal form it has adopted. The choice of the organiza-
tional form is an important design decision, but it is equally important to
acknowledge the importance of governance microstructure. This consider-
ation is important in the for-​profit/​nonprofit distinction but becomes even
more crucial as we make the distinction between private and public organ-
izations. In the public/​private distinction, labels can be misleading, because
whereas the designer can often choose between the for-​profit and the non-
profit forms, the design choice is almost never between the public and the
private forms. Instead, the governance decision is more fine-​grained, be-
cause the key question is, “What are the roles of public and private actors in a
partnership of the two?”

Private vs. Public

The public/​private distinction is more challenging than the for-​profit/​non-

profit distinction because whereas the latter describes actual organiza-
tions, the former defines two ends of a continuum. Specifically, even though
there are both exclusively public and exclusively private organizations, a
vast number of organizations are best described as hybrids of the two, that
is, public-​private partnerships of sorts. For this reason, public and private
are separated by a dashed line and for-​profit and nonprofit with a solid line
in figure 5.1. That the public/​private distinction constitutes a continuum
becomes salient in the context of limited liability companies.
Many limited liability companies, particularly smaller ones, are exclu-
sively private in that all shares are owned by private entities holding the cen-
tral decision rights and rights to the residual. Even large, open corporations
are predominantly private even though public actors may hold an equity
stake in them. For example, Norges Bank Investment Management, the in-
vestment fund of the Government of Norway, owns about one percent of
shares in Apple.
There are also fully state-​owned limited liability companies such as the
federally owned electric utility corporation Tennessee Valley Authority
(TVA) in the United States. However, fully state-​owned corporations are
uncommon. Most “state-​owned” corporations are like Finnair, the flag car-
rier and the largest airline in Finland. Only 56 percent of Finnair’s shares
 Nonprofit and Public Organizations  139

are owned by a public actor, the Republic of Finland. Securing the republic’s
interests as a shareholder is assigned to the Ownership Steering Department
of the Prime Minister’s Office. The remaining 44 percent of shares are owned
by private institutions and individuals.

What Does It Mean for Finnair to Be a Public Organization?

The descriptively accurate term to describe Finnair would be a public-​private
partnership, but in case a definitive categorization is desirable for some pur-
pose, most of us would probably categorize Finnair as a public organization
because the state has a majority ownership stake. However, there are three
reasons why classifying Finnair as a public organization may be misleading.
First, Finnair’s employees are not civil servants; instead, they have private
employment contracts with the organization. In Efficiency Lens terms, man-
agement consists exclusively of private actors with a private-​law employment
contracts with the organization.
Second, seven of eight board members are either private board
professionals or executives of private firms; only one board member
represents the Prime Minister’s Office and, therefore, the public interest. In
Efficiency Lens terms, those who exercise oversight are primarily private citi-
zens with private-​law employment contracts in private organizations.
Third, even though 56 percent of equity is state-​owned, we propose that
the majority of residual risk (when considered more broadly than just in-
vestment risk) is borne by private actors due to commitments to specificity.
Pilots, maintenance technicians and engineers, and cabin crew all commit
to specificity by completing training that is not only highly specialized but
also highly specific. Even though a pilot certified to fly the Boeing 737-​200
aircraft could in principle fly the aircraft for any commercial airline, the like-
lihood of a Finnair pilot finding employment in another commercial airline
is low.2 The same restriction applies to a maintenance technician specialized
in component repair of the Airbus A350-​900 aircraft. Consequently, the re-
sidual risk is borne not only by shareholders (44 percent of whom are pri-
vate organizations and individuals) but also by employees (practically all of
whom are private individuals with private-​law employment contracts).

2 About ten years ago, one of the authors had a pilot from the German airline Lufthansa as a stu-
dent in his MBA seminar. The author asked the student how many Finnish nationals were employed
by Lufthansa as pilots. The student promised to look into it and came back the next day with the an-
swer: zero. Pilot mobility is severely constrained.
140  Governance within and across Organizations

In sum, viewed through the Efficiency Lens, we see how Finnair, and
many other organizations like it, are only ostensibly public. In fact, it
would be perfectly justifiable to label Finnair “more private than public”
even though in most categorizations (including fig. 5.1) Finnair is
considered a public organization. A telltale sign that Finnair is indeed
considered public is that one of the persistent topics in public discussions
is Finnair’s potential privatization, operationally defined as the state
giving up its majority equity stake. In trying to follow these privatiza-
tion conversations, it has been difficult to pinpoint an explicit, let alone
agreed-​upon, main problem that privatization would address. In what
ways would the organization be more efficient if, say, 51 percent (in-
stead of 44 percent) of Finnair’s shares were privately held? What, if an-
ything, would change in the management and oversight where the vast
majority of actors are already private? Pilots would still fly the planes,
cabin crew would serve the customers, technical experts would per-
form maintenance and repair duties (all under private-​law employment
contracts), and the board of directors would still serve as a fiduciary of
the organization.
To be sure, Finnair has been at the center of many controversies. Like
many other airlines, public and private alike, Finnair has had frequent and
severe clashes with unions. Without going into detail on these clashes, we
see commitments to specificity and debates over residual rights of control
as the main drivers.3 It is difficult to see how “privatization” would alle-
viate these problems; that the state owns a majority of the voting rights
in the corporation seems like an ancillary issue. Consequently, we invite
those promoting privatization of public organizations—​both in the case of
Finnair and more generally—​not only to clearly define what they mean by
the term privatization but also to explicate the main problem that privati-
zation would address.
Let us return briefly to the Finnish oil company Neste (see ­chapter 4).
The Republic of Finland owns 44 percent of the shares in Neste; the second
largest shareholder is a private insurance company with a 1.3 percent equity
stake. What purpose would be served in classifying Neste as a private and

3 A recurring conflict between Finnair’s management and the pilot’s union concerns the rights to
residual control over asset utilization: Can management contract with non-​union pilots to fly the
planes? This effectively raises the question who de facto owns Finnair’s fleet of aircraft. In the conven-
tional sense, the fleet is of course the property of the legal entity Finnair Plc. However, the question at
hand is who owns them in the rights-​to-​residual-​control sense, where the answer is not as clear.
 Nonprofit and Public Organizations  141

Finnair as a public corporation? What aspects of governance would these

classifications elucidate? To us, a much better option is to examine in detail
the way these two companies incorporate into their governance decisions the
fact that their residual claimants involve both private actors and the public
interest. We find it plausible that shareholder demographics are not as rele-
vant as one might think. Again, we must look beyond the form into the gov-
ernance microstructure.

Caterpillar, the Archetypal Private For-​Profit

Caterpillar Inc. is a categorically private organization, because the vast ma-
jority of its shares are held by private actors, most notably private institutional
investors such as Vanguard, BlackRock, and Bill & Melinda Gates Foundation.
Much as in the case of large publicly traded corporations, Caterpillar’s own-
ership is highly fragmented across thousands of shareholders. Caterpillar is
in many ways representative of the large, modern corporation. The compen-
sation structure of Caterpillar’s board of directors is also more representative
of large open corporations in the United States: Nonemployee directors re-
ceive an annual cash retainer of $150,000 and an equal amount in restricted
stock with a one-​year vesting period. Directors are also required to own
Caterpillar stock equal to five times their annual cash retainer (Caterpillar
2021). Caterpillar’s board compensation is strikingly different in its struc-
ture compared to that of Neste and Finnair; the latter two neither incentivize
board members with equity nor oblige them to become residual claimants
through stock ownership.

Summary: Both Form and Microstructure Matter

The main conclusion from the preceding discussion is that the choice of
organizational form, while foundational, establishes only broad guidelines
for governance. The key message to the designer is that choosing the proper
governance structure is a matter of both design choice and microanalyt-
ical detail. The latter becomes particularly relevant when attention turns
to governance dynamics. Specifically, adaptation over time is seldom a
matter of changing from one form to another but, rather, implementing
finer-​grained adaptations within the chosen form. Organization scholar
Ilya Cuypers and colleagues (2021, 129) offer an example in the context of
a joint venture:
142  Governance within and across Organizations

[I]‌n response to contextual changes, firms might continue with a joint ven-
ture [joint equity limited liability company] rather than move to a wholly
owned subsidiary [vertical integration], but they might increase the levels
of hierarchical control within the joint venture by altering ownership stakes
or reshaping the board of directors.

This example reveals the essence of governance adaptation: It is about

changes within the form, not changes of the form.
In the spirit of analyzing governance microstructure, we examine in the
following governance decisions in the context of a nonprofit performing
arts organization. Our aim is to establish efficiency as a relevant topic in
nonprofits as well. In the subsequent section, we take a closer look at organi-
zations that involve a public interest of some kind. In these contexts, we pro-
pose that comparative efficiency analysis may not be central.

The Nonprofit Organization in the Performing Arts

How does a theater company organize itself? Should it choose a govern-

ance structure that includes a residual claimant? If the theater organizes as a
nonprofit, how does it attract donors? Who is in charge of management and
makes the most important decisions about repertoire? Who has the key over-
sight role? Is there residual risk? Who bears it and how?
In performing arts organizations, the artistic director has conventionally
been the most important decision maker and, therefore, in charge of man-
agement. The artistic director is not only responsible for curating the season
by choosing programming and content but also the person who, along
with the artistic staff, has a key role in selecting artists with whom the the-
ater contracts. The artistic director must also interact with the prospective
financiers of the theater, that is, both prospective audiences (earned revenue)
and donors (unearned revenue). In sum, the artistic director’s role is central
in the theater’s relationship with its key constituencies: donors, audiences,
artists, and staff. Nonprofit theaters are often in many ways expressions or
extensions of one person’s vision, especially in theaters where the artistic di-
rector is also the founder.
As the theater grows, its organization professionalizes in several ways. One
sign of professionalization is that management is divided into “managing
the art” and “managing the business” (cf. Bhardwaj and Sergeeva 2022). The
 Nonprofit and Public Organizations  143

artistic director is responsible for the former and a managing (or executive)
director for the latter.
As the theater organization professionalizes, a number of constraints
emerge on the artistic director’s discretion. Instead of curating the season
in a way that expresses the artistic director’s vision, “the season planning
process is determined by the needs of the theater, the community, and the
artists” (Colburn 2007, v). The emergence of organizational constraints gives
rise to oversight, which is exercised by the theater’s board of directors.
Some nonprofit theaters succeed in covering their costs with earned and
unearned revenue; others fail. Failure has variable consequences to donors,
artists, audiences, and managers. Just like in other organizations, theater or-
ganizations have constituencies who, by virtue of their relationships with the
organization, bear risk.
In sum, the professionalized nonprofit theater contains all three elements
of the Efficiency Lens: management, oversight, and risk. We may thus apply
the Efficiency Lens and conduct a stakeholder analysis to examine the gov-
ernance ramifications.

A Stakeholder Analysis of the Nonprofit Theater

In the nonprofit, there are no residual claimants expecting residual payments,

and some of the revenue is unearned as it originates in donations. These two
idiosyncrasies make the nonprofit theater different from a for-​profit com-
mercial organization. At the same time, the general logic of stakeholder
analysis is still equally applicable. Specifically, the designer can still evaluate
each constituency for potential residual risk to determine whether the rela-
tionship between the constituency and the organization exhibits stakeholder
characteristics. After having identified the stakeholders, the designer can
seek the appropriate governance responses to safeguard the most important
relationships.

Donors as Stakeholders
The absence of residual claims must not be interpreted as the absence of re-
sidual risk (Fama and Jensen 1983a). All that is required for residual risk to
exist is a constituency that becomes vulnerable by virtue of its participation
in the organization. In the nonprofit theater, the obvious bearers of residual
risk are the donors, who are putting their wealth at stake.
144  Governance within and across Organizations

At the same time, becoming a donor is not merely a matter of putting

one’s wealth at stake. There are also important reputation effects that fur-
ther deepen the stakeholder relationship between the donors and the theater.
Specifically, donors tend to offer not only their wealth but also their identities
to the service of the theater. Furthermore, donors do not act independently
of one another; instead, they are members of communities of donors. A well-​
known member of the community donating to a theater sends a signal to
other members that the theater is a legitimate target for donations, which ef-
fectively ties not only the prominent donors’ wealth but also their reputation
to organizational outcomes. Therefore, even though donors are not residual
claimants, they have a conspicuous residual interest in the theater organiza-
tion and, consequently, a vested interest in efficient governance, oversight in
particular.
The absence of residual claimancy does not make the agency problem
irrelevant either. All that is required for an expropriation hazard (and an
agency problem) to exist is the separation of management from risk. In fact,
“agency problems between donors and decision agents in nonprofits are sim-
ilar to those in other organizations where important decision managers do
not bear a major share of the wealth effects of their decisions” (Fama and
Jensen 1983a, 319). Because donations are inalienable (the donor cannot re-
nege on a donation already made), the expropriation hazard is in some ways
more acute in the nonprofit than it is in an open for-​profit corporation where
residual risk is freely alienable.4
Given that donors put both their wealth and their reputation at stake, it
is understandable why board members of donor-​dependent organizations
often either represent donors or are in fact donors themselves (Fama and
Jensen 1983a, 319). Donors exercise oversight to ensure their donations are
used in the best interest of the organization and not expropriated as private
benefits by powerful members of management.
That donors face residual risk is not novel (e.g., Turbide and Laurin 2014).
In fact, the very idea of organizing as a nonprofit can be viewed as a deliberate
signal to donors that their donations are safe (Glaeser and Shleifer 2001).
Although the nonprofit form does not safeguard against expropriation of

4 Even though recovering donations already made is not feasible, donors have efficient ex post
remedies available in case donations are poorly managed by the organization. If a prominent donor
concludes that the theater is mismanaging donations, one option is to refrain from future donations.
The name of a prominent top-​tier donor disappearing from the donor list sends a strong signal to the
donor community that the theater is no longer a legitimate target for donations.
 Nonprofit and Public Organizations  145

donor funds, it prevents their excessive appropriation. The difference be-

tween appropriation and expropriation is that appropriation can occur, even
in excess, without breaking the rules. In contrast, expropriation tends to in-
volve taking possession of wealth in ways that are not only inappropriate but
possibly also illegal. The absence of a residual claimant creates a set of rules
that mitigate excessive appropriation, but these rules are ineffective if they
are not obeyed.5

Artists and Staff as Bearers of Residual Risk

The constituencies whose potential stakeholder status has received less at-
tention are the artists and the staff. Just like donors, artists are not residual
claimants. However, to the extent they commit to specificity or become vul-
nerable in other ways, they may bear residual risk.
Contracting relationships between a theater and its employees are subject
to the fundamental transformation just like any other contracting relation-
ship. In fact, this process was explicitly embedded in the governance of film
studios in the early and mid-​1900s. In what was known as the studio system
(Gomery 2005), actors would sign long-​term contracts with specific studios.
For example, the child superstar Shirley Temple contracted in the 1930s pri-
marily with 20th Century Fox and its predecessor Fox Film Corporation.
Under the studio system, the names of specific actors became intertwined
with the names of specific studios. The fundamental transformation would
occur over time as the actors and the studios learned to contract and collab-
orate with one another.
Analogously with the Hollywood studio system, when artists and non-
profit theaters develop reciprocal relationships over time, stakeholder
relationships emerge. Artists in the nonprofit theater context tend to develop
long-​term relationships with specific theater organizations by becoming res-
ident actors. Specificity develops both inadvertently and through deliberate
commitments and, as a result, switching costs becomes sufficiently high to
merit the designer’s attention. Importantly, dependency is bilateral, because

5 The distinction between appropriation and expropriation is subtle but important. The
organization’s constituencies appropriate revenue in various legitimate ways: employees appropriate
fixed salary payments, shareholders appropriate residual dividend payments, and so on. To the extent
appropriation is excessive, it may become a cause for concern; excessive executive compensation is
a representative example. In contrast, expropriation is always not only improper, but it may also be
illegal. Insiders expropriating private benefits from a nonprofit organization is a representative ex-
ample. Expropriation may result in the nonprofit organization losing its tax-​exemption privileges for
the fiscal year in which expropriation is discovered.
146  Governance within and across Organizations

both artists and staff may possess idiosyncratic skills that become intertwined
with the identity and the repertoire of the theater organization. The depar-
ture of one well-​known actor cannot be addressed simply by contracting an-
other well-​known actor.
The relational process is a general phenomenon in the context of the arts
and entertainment: The identities of individual directors, actors, producers,
and studios matter. How would Gracie Films and 20th Television, the pro-
duction companies of the animated television series The Simpsons, replace
Nancy Cartwright, the actor who gives her unique voice not only to one
of the central characters, Bart Simpson, but also to a half-​dozen of others?
Indeed, the world of arts and entertainment offers perhaps the most salient
examples of specificity and high switching costs.
The idea that actors may bear significant residual risk has been incorpo-
rated into theater governance, but only very recently. In January 2022, after
several years of internal conflicts and turmoil, American Shakespeare Center
(ASC), a Virginia-​based regional theater company, appointed Brandon
Carter both as its artistic director and an ex officio member of its board of
trustees. The interesting fact about Mr. Carter is that he is a resident actor
at ASC. Moreover, at ASC, curating the season is not the exclusive prerog-
ative of the artistic director; instead, it is based on a management structure
described as “a coequal group of individuals.”6 This coequal group currently
consists of the artistic director, director of creative planning, programming
coordinator, digital projects coordinator, performance studies manager,
and a community programs manager, all of whom are either artists or staff
members. In many ways, ASC has embraced the idea that those in charge of
management need not necessarily be managers (see ­chapter 2).

Audiences and Residual Risk

In ­chapter 4, we suggested that consumers are unlikely to bear sufficient
residual risk to justify a stakeholder status. In the case of many consumer
products and services, switching costs are low and the availability of alter-
native suppliers of products and services provides a sufficient safeguard.
Interestingly, the nonprofit theater might constitute an exception to the
rule: In the nonprofit theater context, audiences are not merely consumers
of art but may also have a relationship that exhibits elements of residual risk.

6 Wash. Post (Jan. 10, 2022). Retrieved at https://​www.was​hing​tonp​ost.com/​thea​ter-​dance/​2022/​

01/​10/​car​ter-​ameri​can-​shak​espe​are/​
 Nonprofit and Public Organizations  147

An analogous phenomenon is found in the world of professional team sports

where spectators are not merely consumers of sports entertainment; over
time, many become loyal, lifetime fans of specific teams.
The audience members with a clear residual interest are the donors who,
in addition to funding the organization, are holders of season tickets. The re-
sidual interest of the nonprofit theater donor is found not in residual claims
(there are not any) but in the effect the theater organization has on the sur-
rounding community. Unlike Hollywood film studios and the for-​profit
world of Broadway theater, nonprofit theaters tend to be small, local organ-
izations whose primary audiences reside in the local communities in which
the theaters are located. Donors and actors are often residents of the local
community as well. Local theaters are therefore not only commercial organ-
izations but also social communities whose role extends beyond providing
entertainment; they may become avenues of activism, social commentary,
even instruments of social justice. In the research literature, these are some-
times labeled hybrid organizations in that they have both an economic and a
social mission (e.g., Ebrahim, Battilana, and Mair 2014). To be sure, artistic
expression can be intimately intertwined with social and political interests.
When a theater organization links to broader social objectives, we can
see how not only donors and artists but also audiences may develop a re-
sidual interest. If the local nonprofit theater fails to curate the season in a way
that meets the broader social objectives in the community, switching to an-
other nonprofit theater in another community does not constitute a feasible
remedy. Therefore, both the idea of switching and the associated switching
cost seem inapplicable in this context.

What’s Next?

Salma Qarnain, cofounder and executive producer of the New-​York-​based

Black Man Films and former managing director and chief operating officer
(COO) of Synetic Theater in Washington, DC, described (personal commu-
nication, March 1, 2021) an intriguing, emerging way of thinking about risk
in theater organizations. Specifically, the conventional focus on donors and
founding artistic directors as the sole stakeholders is being countered with the
inclusion of artists. Artists have always been considered constituencies of the
theater organization, but elevating them to a stakeholder status implies that
their contractual relationships are now considered in broader governance
148  Governance within and across Organizations

terms that extend beyond employment contracts. Appointing Mr. Carter, a

resident actor, as ASC’s artistic director and member of the board of trustees
is a representative example. Just like donors make a wager on organizational
outcomes, so do artists and staff.
In the spirit of comparative analysis, the implications of this new gov-
ernance thinking must be addressed in their entirety. Does the new model
enhance organizational viability? Does a shared leadership model increase
or stifle innovation? Does a broader consideration of stakeholder interests
dilute the ability of the board to engage in efficient oversight of unearned
revenue and therefore jeopardize the organization in the eyes of the donors?
Does the new model lead to more diversity and inclusion? According to Ms.
Qarnain, it is too early to tell whether the new governance models result in
net gains for the organization—​the unintended consequences are still largely
unknown.
We see the situation as analogous to opening up the board of directors
in for-​ profit corporations to broader stakeholder interests. Williamson
(2008, 250) noted that “giving the board stakeholder responsibility dilutes
its credible contracting support for equity.” In the context of donor-​based
nonprofits, the concern translates to the question, “How will donors react
when the organization elevates artists from a constituency to a stakeholder
and incorporates their interest in board-​level decisions?” We submit this to
the designer as a central question to be addressed in thinking of board com-
position. In the following, we take a closer look at the nonprofit board in light
of its fiduciary duty.

Fiduciary Duty of the Nonprofit Board

If we adopt the premise that the board has a fiduciary relationship with the
organization, then the tasks of the nonprofit and the for-​profit boards are es-
sentially identical. In both contexts, there are those who bear residual risk.
Whether those bearing residual risk are also residual claimants may be rele-
vant but not central. More significant differences would emerge if we viewed
only the shareholder, not the organization more broadly, as the beneficiary
of the board’s fiduciary duty in the for-​profit setting. But as we established in
­chapter 4, such thinking is misguided.
Drawing on the insights of Fama and Jensen (1983a), we counsel the de-
signer to view the fiduciary duty of the board in both for-​profit and nonprofit
organizations not in terms of residual claimancy but in terms of residual risk.
Consequently, the board should incorporate into its decisions especially
 Nonprofit and Public Organizations  149

those constituencies that bear residual risk. In the nonprofit context, the
sources of unearned revenue (the donors) are the obvious constituency with
a stakeholder status. But to the extent that other constituencies (e.g., those
with long-​term employment relationships) bear residual risk, they should be
considered stakeholders as well.
Unfortunately, there are reasons to believe that the boards of nonprofit
performing arts organizations are inefficient in their oversight role (Galli
2011). It is also often the case that board members are beholden to the artistic
director because they have been introduced to the board either by the artistic
director or a fellow board member. When this relationship is combined with
the principle of appointing donors to the board, the de facto targets of the
board’s attention are the donors and the artistic director. As we discussed in
­chapter 4, stakeholder representation on boards may jeopardize the board’s
independence.
Again, it is crucial to distinguish between stakeholder representation on
the boards and incorporating stakeholder interests at the board level. Much
as in the case of for-​profit corporations where the CEO may be able to ex-
ercise excessive control over the board (Mace 1971), oversight in the non-
profit theater may be unduly influenced by the artistic director. Indeed, the
artistic director may not only lead the board of directors but also choose its
members. This influence leads to the undesirable outcome of insufficient
separation of management and oversight, which may have adverse efficiency
consequences. The proposition that incorporating employee interests into
theater governance enhances governance efficiency merits attention. This
consideration does not, however, necessarily imply artist representation
on the board. Giving artists more prominent roles in managing the theater
might offer a viable option. Again, appointing a resident actor as the ar-
tistic director is an illustrative example of a governance choice that might
lead to an efficient outcome. Specifically, promoting artists to top managerial
positions gives them more voice in the organization without running the risk
of diluting donor interests at the board level.

The Boundaries of Efficient Governance

Throughout the preceding chapters of this book, we have promoted an effi-

ciency approach to governance. However, efficiency is not always applicable
in governance decisions; sometimes focusing on efficiency may be either
150  Governance within and across Organizations

premature or downright misguided. Whereas the boundaries of governance-​

as-​efficiency thinking should be acknowledged in all contexts, we find that
they become particularly salient in organizations that involve public actors
and the public interest, which is why we find this chapter a suitable context
for critically examining the boundary conditions and the limits of applica-
bility of the comparative efficiency approach.
Let us start at a context in which a comparative efficiency analysis is
conspicuously applicable: the make-​or-​buy decision in an industrial firm
(­chapter 3). In the make-​or-​buy decision, the trade-​off is between produc-
tion costs and transaction costs: Outsourcing the production of a compo-
nent lowers production costs (because specialized suppliers are often more
productive) but increases transaction costs (because transacting across a
legal boundary complicates exchange). Applying the efficiency logic suggests
that if the savings due to increased production efficiency offset the increased
transaction costs associated with the outsourcing decision, using an external
vendor to supply the component is comparatively efficient over in-​house
production.
In the following, we first explicate and then critically examine the
assumptions underpinning an efficiency analysis of governance alternatives.
Failing to understand the assumptions and the associated boundary
conditions, the designer may apply the efficiency logic in contexts in which it
is not applicable.

The Assumptions of Comparative Efficiency Analysis

A comparative efficiency analysis hinges on three implicit assumptions,

summarized in table 5.1. One is that the transacting parties enter the trans-
action voluntarily. If there are power asymmetries, efficiency analysis of
the transaction loses its relevance (Ketokivi and Mahoney 2020). In the
case of power asymmetries, whether the component is produced in-​house
or purchased from an external vendor will be decided by the compara-
tively more powerful party who will likely base the decision on its own cost
implications. A powerful buyer may simply impose the supply of a compo-
nent to an external vendor and force the vendor to absorb as much of the
cost of transacting as possible. However, this shifts focus from sustainable
efficiency to myopic efficiency, and as we have established at the beginning
of this book, the latter is not of interest to our exposition. Efficiency analysis
 Nonprofit and Public Organizations  151

Table 5.1  The Three Assumptions Required for a Comparative Efficiency

Analysis

Assumption Description

Voluntary The parties to the economic exchange enter the relationship

participation voluntarily; participation is neither expected nor imposed.
Instrumental Governance choice is guided by the common objective of
outcomes organizing the relationship efficiently.
Commensurate The trade-​offs made in alternative governance choices can
outcomes be assessed in efficiency terms; where a governance choice
involves multiple outcomes, they must have a commensurate
metric so that an analysis of net gains is possible.

starts with the premise that the exchange parties are interested in the effi-
ciency of the relationship, not just their respective income statements and
balance sheets.
The second assumption is that when the transacting parties evaluate alter-
native governance choices, neither the specific governance choices nor the
outcomes of the choices have intrinsic value. A case in point, an efficiency
analysis of the make-​or-​buy decision adopts the premise that there is nothing
intrinsically valuable about lower production costs or lower transaction
costs, both have only instrumental value. Indeed, this is the essence of the
notion of feasible alternatives. Instrumentality of outcomes is the reason why
the governance alternatives can be subjected to an explicit trade-​off calculus.
The instrumentality assumption becomes challenged in contexts in which
either the governance choices themselves or some of the outcomes of the
choices have intrinsic value. For example, some business schools might en-
dorse the principle that a respectable business school has its own faculty and,
consequently, readily dismiss the extensive use of visiting and adjunct faculty
as an option. If there are intrinsically valuable outcomes, comparative effi-
ciency analysis loses its relevance.
The third premise is that efficiency analysis in general and the idea of net
gains in particular assumes that the relevant outcome categories associated
with each governance option are commensurate, and consequently, how they
are traded off against one another becomes not only possible in principle (the
second assumption) but also analytically tractable. In the case of the make-​
or-​buy decision, the relevant cost categories are production costs and trans-
action costs. Even though the latter may be less salient than the former, the
152  Governance within and across Organizations

Table 5.2  Examples of Governance Decisions That Challenge the Efficiency

Approach

Governance decision Efficiency assumptions violated or questioned

Inmate discretion in (a) Inmates’ right to self-​determination and the

prisons correctional staff ’s safety are both intrinsically valuable and
incommensurate outcomes; and (b) inmates’ participation
in the organization is imposed.
Physical restraint of (a) The patient’s physical integrity and the hospital staff ’s
patients in psychiatric safety are both intrinsically valuable and incommensurate
care outcomes; and (b) patients object to being physically
restrained.
Hybrid teaching in (a) The outcomes for students in the classroom and
universities for those attending online are not commensurate; and
(b) faculty and administrators may value different
objectives.

two remain commensurate cost categories that can be traded off against one
another. Consequently, the idea of net gains becomes relevant: If savings in
one cost category offset increases in another, net gains result.
It is straightforward to think of examples of contexts in which at least some
of the assumptions are violated. Examples that involve public interest are
the most conspicuous ones. Table 5.2 gives three different contexts in which
governance-​as-​efficiency thinking is challenged. We discuss each briefly in
the following sections.

Inmate Discretion in Prisons

One of the key design decisions in prison governance is the amount of dis-
cretion given to inmates regarding behavior, language, wardrobe, visitors,
and so on. In governance terms, the question is about the inmates’ residual
rights of control. In his in-​depth analysis of prisons in the states of California,
Michigan, and Texas, political scientist John DiIulio (1987) elaborated on the
organizational details of prison governance in important ways. DiIulio (1987,
99) started at the idea that “[w]‌hether a prison (or a prison system) is safe,
humane, and treatment-​oriented, on the one hand, or violent, harsh, and un-
productive, on the other, may depend mainly on the character of its prison
governance.” Among other issues, DiIulio focused specifically on inmate dis-
cretion. Giving inmates more discretion and autonomy would privilege their
rights to self-​determination and possibly enhance their rehabilitation, but at
the same time, it might increase safety hazards for prison staff. DiIulio (1987)
 Nonprofit and Public Organizations  153

observed that prisons in Texas gave comparatively less discretion to inmates

than prisons in California and Michigan.
As table 5.2 shows, several assumptions of the efficiency approach are
violated: Inmate participation in the organization is imposed, constituencies
(inmates vs. staff) have strong ex ante preferences regarding specific govern-
ance options, and the constituency outcomes are both intrinsically valuable
and incommensurate. Therefore, the question whether there is an amount of
discretion that provides a comparatively efficient alternative is misguided.
Consequently, as political scientist James Wilson (1989, 360) noted in his dis-
cussion of decision-​making in such situations, “[h]‌owever thoughtful people
decide these matters, I doubt they will decide them on economic grounds.”

Physical Restraint of Psychiatric Patients

One of the most controversial topics in psychiatric care is the physical re-
straint of patients. In what is known as five-​point restraint, the psychiatric
patients’ ability to move is all but completely eliminated by placing them on
a gurney and physically restraining movement at all four limbs and the waist
or the chest. Those in the five-​point restraint cannot so much as scratch their
noses—​it is hard to think of a more serious invasion of a person’s right to
physical integrity. Just as in the context of prisons, the designers of restraint
policies readily understand that a comparative analysis of net gains cannot
possibly guide design decisions.
To be sure, efficiency is a relevant consideration in many design decisions
regarding the governance of healthcare organizations. However, in
discussions with a forensic psychiatrist who both made decisions regarding
the restraint of psychiatric patients in her own organization and consulted
her colleagues in other hospitals, it became clear to us that a comparative
efficiency analysis had no role in the formulation of policies and procedures
for physical restraint. Instead, ensuring system integrity was paramount.
Considerations of integrity evolved largely around the question of patient
rights, and one of the main objectives of both legislation and psychiatric care
was to reduce the use of physical restraint as much as possible. The reason the
issue is relevant in governance decisions is because legislation in and of itself
has proved in sufficient (Keski-​Valkama et al. 2007). Both institutional and
private-​ordering measures are required.
In psychiatric care, staff safety is obviously an important concern, but it
can usually be achieved by merely secluding the patient. Physical restraint is
to be applied only in the extreme situation where the patient’s own safety is in
154  Governance within and across Organizations

jeopardy due to an inadvertent or deliberate self-​harm hazard.7 Furthermore,

in the event the patient is physically restrained, the process is made deliber-
ately inefficient in the sense that not only is the patient under constant obser-
vation by a healthcare professional but also a formal report on the patient’s
condition must be submitted every hour. There are serious accountability
consequences for prolonged restraint without cause, and the formal hourly
reports are used in ex post evaluations of each restraint episode. Finally, all
patients are systematically briefed after being restrained. All these principles
and procedures are in place to ensure system integrity.

Hybrid Teaching at Universities

The COVID-​19 pandemic forced all institutions of higher education to create
policies and practices for organizing online instruction. As universities
returned to the conventional in-​class instruction in late 2021 and early 2022,
they faced two options. One was to return exclusively to in-​class, face-​to-​
face instruction; the other was to adopt a hybrid format where some students
would be in the physical classroom and others attended online. Can this
choice be subjected to a comparative efficiency analysis?
As faculty members in higher education, we both observed that some
constituencies answered in the affirmative and, accordingly, adopted an ef-
ficiency perspective in analyzing which option should be preferred. The
premise was that instruction would be efficient when a maximal number of
students could attend class. This premise led to the conclusion that the hy-
brid option would be preferred, as it would allow everyone, including those
not physically present, to attend class.
We have reservations against subjecting hybrid teaching to a comparative
efficiency analysis. This is because we do not think the outcomes for different
constituencies are commensurate. Whereas those unable to attend class in
person predictably value online participation, the hybrid format may lead
to negative learning outcomes to those in the physical classroom. As fac-
ulty members who taught numerous hybrid classes in the 2020–​2021 and
2021–​2022 academic years, we found it a pedagogical impossibility to give
equal attention to those in the classroom and to those attending online. The
predictable, and unfortunate, adjustment that many faculty members made
was that they switched from a conversational, dialectical approach to more

7 In retrospective surveys and interviews, some restrained psychiatric patients expressed beliefs
that the use of physical restraints ultimately protected their own safety; others expressed anger, fear,
and distrust toward staff (Wynn 2004).
 Nonprofit and Public Organizations  155

conventional lecturing. We submit that the two pedagogical approaches—​

dialectical versus lecturing—​led to different kinds of learning outcomes that
cannot be compared in efficiency terms. Consequently, the designer must
use judgment (not efficiency analysis) to decide whether it is appropriate to
forgo the benefits of interactive learning in favor of a large-​scale lecture-​like
format.

Governance as Integrity and the Probity Hazard

How should the designer approach design decisions in situations in which

the notion of comparative efficiency is inapplicable? How do designers facing
decision situations illustrated in table 5.2 choose among the options available
to them?
Since this book is about efficient organizing, the examples in table 5.2
fall outside the scope of any prescription that could be derived from the
Efficiency Lens. Accordingly, the main purpose of the discussion here is not
to make recommendations but, instead, to explicate the boundary conditions
of the comparative efficiency approach. This said, organization economists
may have something relevant to offer to the designer even in circumstances
where comparative efficiency is no longer the central objective.8
In his discussion of public organizations, Williamson (1999, 340)—​one
of the main architects of the governance-​as-​efficiency approach—​noted
that there are contexts in which governance-​as-​efficiency must yield to
governance-​as-​integrity. Instead of efficiency, the designer must place in-
tegrity, or probity, as the central objective. Accordingly, governance choices
should be compared to one another not in terms of their comparative effi-
ciency but in terms of their potential to avoid probity hazards. The prescrip-
tion is to choose the governance option that leads to comparatively lower
probity hazards.

8 Some organization economists suggest that even in contexts where some of the assumptions re-
quired for efficiency analysis are not met, there may be governance decisions that can be subjected
to an efficiency analysis. Even in the prison context, there are governance decisions in which the ef-
ficiency assumptions hold. For example, organization economists Oliver Hart, Andrei Shleifer, and
Robert Vishny (1997) suggested that the question whether the operational privatization of a prison
(delegating prison management to a private contractor while maintaining public oversight) can be
subjected to an efficiency analysis. However, before such an analysis is conducted, the designer must
ensure the assumptions of a comparative efficiency approach are indeed met.
156  Governance within and across Organizations

That a comparative efficiency analysis is inapplicable does not mean

that the Efficiency Lens and all its key concepts are irrelevant. Just the op-
posite: Nothing prevents the designer from analyzing, for example, the
implications of separating (or not separating) management from oversight
or management from risk in terms of system integrity. To be sure, efficiency
is not the only reason why designers contemplate separation of powers. The
designer merely has to acknowledge that the objective of separation (or
nonseparation) is different in contexts that require high probity.
A case in point, in the context of polity, the separation of powers into the
three coequal branches of government—​the legislative, the judicial, and
the executive—​has more to do with the integrity of the system of govern-
ment than its efficiency. In fact, polity represents a context in which some
inefficiencies may be deliberately designed (Williamson 1999, 318) into the
system to avoid the hazard that arises from, say, making hasty decisions. How
bicameral legislatures pass legislation offers an illustrative example of delib-
erately slowing down the legislative process. The existence of not only trial
courts but also appellate and supreme courts is an example of deliberately
slowing down the judicial process. That the bicameral legislative process
and the three-​tiered judicial process are inefficient is obvious; however, this
inefficiency is not a cause for concern, because in these contexts, probity is
more important than its efficiency. There are contexts in which organizations
should be deliberately designed to be economically inefficient.
At the same time, as we noted in c­ hapter 3, there are aspects of the legis-
lature where efficiency is a relevant concern. It is important to bear in mind
that efficiency is not something that either is or is not relevant to an organi-
zation; it may be relevant to some transactions and relationships but not to
others. Again, the unit of analysis in the efficiency approach is the relation-
ship, not the organization in its entirety. Consequently, the designer should
not discard the efficiency approach entirely simply because it does not apply
in a specific decision situation. In fact, we propose that the ability to identify
those decision situations where efficiency applies and when it has to yield to
probity considerations is a very valuable skill for the designer.

The Importance of Analyzing Assumptions

The preceding discussion emphasizes the crucial role that various
assumptions have in all decision-​ making. In general, making implicit
assumptions explicit always enhances transparency. However, in addition
to making the implicit explicit, we also counsel the designer to rigorously
 Nonprofit and Public Organizations  157

analyze the assumptions to determine if they are applicable in the specific

decision situation. The designer should neither accept nor reject an assump-
tion without explicit analysis. This is an important reminder, because it may
be tempting for a constituency to seek a privileged position or bargaining
power by suggesting that a certain outcome has intrinsic value and, there-
fore, should not be subjected to an efficiency analysis. Yet after closer inspec-
tion, something that may seem like an intrinsically valuable outcome can
be thought of in instrumental, efficiency terms. Also, sometimes even that
which is intrinsically valuable can have instrumental antecedents.
A case in point, we might be tempted to suggest that personal safety is in-
trinsically valuable. This is a reasonable position, but at the same time, even
a superficial look into contexts in which safety is relevant reveals that there
are many aspects of safety that not only can be subjected to a comparative ef-
ficiency analysis but can also be addressed in contractual terms. Insurance is
the obvious example. Purchasing an insurance policy is an economic trans-
action by which we exchange a portion of our economic wealth for peace
of mind. Even though we might consider having peace of mind intrinsically
valuable, it has contractual antecedents worth the designer’s attention.
As another example of safety, consider job security. Employees, particu-
larly those who commit to specificity, might understandably consider job
security as intrinsically valuable. However, as we have hopefully established
in the preceding chapters, employee commitments to specificity (and the
resultant need for job security) can often be addressed in contractual, effi-
ciency terms, and provisions for job security can be embedded in both in-
dividual employment contracts and governance structures more generally.

Summary

Examination of nonprofit and public organizations offers many insights into

governance. We see the implications and conclusions as twofold.
One conclusion is that although the choice of the organizational form is
a central decision, much of the governance action resides in the details and
the dynamics within the chosen form. Therefore, even though the choice
of form is consequential, the designer must not exaggerate its importance.
Upon choosing the organizational form, much of the designers’ work is still
ahead of them. For example, the main difference between a for-​profit and a
nonprofit organization is that the former has a residual claimant but the latter
158  Governance within and across Organizations

does not. This is a relevant decision criterion, but its implications may not be
as consequential as one might think. For example, the choice between having
and not having a residual claimant is not a choice between seeking or not
seeking a surplus, the main implications are how the surplus (the residual) is
governed.
In general, due to the high variance observed within forms, we find the
categorization of organizations into for-​profit, nonprofit, public, and pri-
vate less useful. To be clear, this does not mean the concepts are not useful.
Just the opposite, public and private are essential concepts if we wish to ex-
plicate the general governance principles of, say, an operationally privatized
prison: management is privatized, oversight remains public. It is therefore
not the use of the concepts public and private but the notion of the private
prison (categorizing an organization) that invites confusion. Discussions of
privatization more generally tend to involve exaggerated claims about how
privatizing a public organization increases efficiency. For these discussions
to become more tractable, those who participate in them should be explicit
about what they mean by privatization and what main problem privatization
is aimed at addressing.
Organization scholars Barbara Levitt and James March (1988,
325) suggested that learning in organizations is superstitious “when the sub-
jective experience of learning is compelling, but the connections between
actions are outcomes are mis-​specified.” The problem with giving too much
attention to the form may lead to superstitious learning. Specifically, be-
cause organizational forms are more salient to the observer than governance
microstructures, the casual observer may incorrectly ascribe organizational
outcomes to the organizational form instead of the characteristics of the gov-
ernance microstructure that are always hidden from plain sight.
The second insight that arises from this chapter is that it brings clarity to
the boundaries of efficiency thinking. To this end, we have in this chapter
explicated the implicit assumptions that underpin the analysis of compara-
tive efficiency. The goal of explicating the implicit is to provide the designer
with the requisite tools to identify the contexts in which a comparative effi-
ciency analysis is applicable. Here, it is crucial to maintain the proper level of
analysis. Specifically, the notion that context matters is not to be understood
as meaning that there are organizations where efficiency considerations are
not applicable; the point is that there are specific governance decisions where
 Nonprofit and Public Organizations  159

this may be the case. Designers are faced with numerous design decisions
both at the founding of the organization and over time. A central design skill
is the ability to selectively apply comparative efficiency analysis to govern-
ance decisions in which it is warranted. To this end, making the assumptions
underpinning comparative efficiency explicit is useful.
 PART III
GOV E R NA NC E A ND
T HE ORGA N IZ AT IONA L
L IF E C YC LE

Organizations face different challenges as they progress through different

stages of their development and maturity. In this third part of the book, we
examine the issues organizations face in the founding (­chapter 6), growth
(­chapter 7), and mature (­chapter 8) phases of their life cycles. We pinpoint
unique challenges associated with each specific phase.
Chapter 6 focuses on the governance questions that an organization’s
founders face. The topic is important, because founders often assume that
questions regarding organization design and governance can be addressed
in the months and the years after the founding, when the organization has
become more established. However, there are crucial questions that require
the designer’s attention even before the organization’s founding. This chapter
addresses both questions that can and should be addressed and are ex ante
contractible (i.e., before the founding) and those that are subject to various
ex post adjustments.
As organizations expand, their governance changes in fundamental ways.
The most consequential changes have to do with the gradual separation of
management, oversight, and risk. Chapter 7 focuses on the challenges and
the dilemmas that separation creates for the designer. The separation dy-
namic tends to make the organization more vulnerable, which is pertinent
particularly in situations in which the organization becomes the target of a
takeover attempt. In c­ hapter 7, we also examine the measures the designer
can take to protect its stakeholder relationships.
As organizations mature, their designs and governance structures become
institutionalized, and their modification becomes challenging. Chapter 8
162  Governance and the Organizational Life Cycle

focuses on the challenges that the persistence of governance choices presents

to the designer. Persistence of governance structures is a potential cause for
concern in intraorganizational relationships and transactions in particular.
Accordingly, c­ hapter 8 complements ­chapter 3 by a closer examination of the
trade-​offs associated with internal transactions.
 6
The Startup Organization

Master’s students in business administration who had just launched startup

firms asked in class when they should start to think about organization de-
sign and governance in their firms. We gave them the only intellectually
honest answer: It was probably already too late. The foundation of efficient
organization is laid in the months and weeks before the founding, not after it.
The students’ misconception likely arose from aspiring entrepreneurs
thinking of organization design and governance in terms of organizational
structures, reporting relationships, profit-​and-​loss structures, and so on.
These aspects of design can indeed be addressed after the organization’s
founding. To be sure, a small startup need not contemplate whether to or-
ganize as a multidivisional firm or as a matrix; many decisions regarding the
organization’s macrostructure do not become relevant until months, perhaps
even years, after the founding. Startup organizations tend to be fluid in their
structures, which makes them efficient.
However, organization design and governance extend beyond structural
choices, and there are many essential questions that merit the designer’s early
attention, because initial choices have far-​reaching consequences. For ex-
ample, once the designer of an industrial startup has decided that the form
of the organization will be a limited liability company with restricted aliena-
bility of residual claims (see ­chapter 2), the organization is likely committed
to this form for as long as it exists, or at the least, subsequent incremental
governance decisions and adjustments will be constrained by the founda-
tional decisions. Consequently, foundational decisions must be made with
conscious foresight, which can work toward imprinting the organization
with efficient governance principles and avoiding costly ex post adjustments.
Examining organization design and governance issues of the incipient
organization is instructive in three distinct ways. One is that it helps the
designer understand which aspects of governance are ex ante contractible
and which must be addressed through ex post adjustment. For example,
cofounders of a high-​technology startup firm may be able to agree ex ante
that the organization’s objective is to seek growth, increase firm value, and

Efficient Organization. Mikko Ketokivi and Joseph T. Mahoney, Oxford University Press. © Oxford University Press 2023.
DOI: 10.1093/​oso/​9780197610282.003.0006
164  Governance and the Organizational Life Cycle

ultimately, go public through an IPO; what is not contractible is the time-

line. The designer must carefully analyze contractibility and think of effi-
cient safeguards for the noncontractible aspects of cooperation in particular.
Ex ante contractible aspects are comparatively simpler because they can
be addressed, by definition, by ex ante contractual safeguards. However,
even there the designer must avoid a false sense of security: Not all ex ante
agreements are enforceable and, consequently, may ultimately require ex post
adjustment.
The second way in which examining startup organizations is useful is
that it helps the designer understand what kinds of ex ante private-​ordering
measures are feasible in the first place. Feasibility is not merely a matter of
contractibility but also one of understanding the limits the institutional en-
vironment imposes on private ordering. For example, most institutional
environments uphold the principle that shareholders of limited liability
companies have the right to sell their shares. Even though the law permits
some private-​ordering constraints, strict restrictions will likely be contested
by the courts. In general, contracting parties cannot privately agree to some-
thing that jeopardizes someone’s legal rights. However, what is considered
illegal is not merely a matter of the letter of the law but more broadly its in-
terpretation and application in specific cases. For example, strict restrictions
on the transfer of shares would mean the contracting parties are effectively
“contracting for market failure.” In the case of a contractual dispute, the party
arguing that the strict restriction should be enforced may have difficulty con-
vincing the judge presiding over the dispute that the court should uphold a
contract for an intentional market failure. The designer must be sufficiently
informed to avoid introducing into contracts clauses that rest on a false sense
of security about enforceability.
The third way in which an analysis of organizations at their founding is
instructive is that it can help the designer better understand the relation-
ship between the ex ante and ex post aspects of governance decisions. Even
though the latter are always conditioned by the former, there are situations in
which ex post decisions may not only modify but also sometimes even annul
or reverse an ex ante condition. These situations do not present a problem as
long as all parties to the contract agree. However, in many settings the deci-
sion to reverse an ex ante condition may be the prerogative of a subset of the
contracting parties, which turns attention to residual rights of control. An en-
during question in governance is how much discretion the board of directors
should have in modifying corporate bylaws, and which modifications
 The Startup Organization  165

require shareholder approval. In Efficiency Lens terms, which aspects of ex

post contracting are matters of oversight delegated to the board of directors?
We start this chapter with an example of a sports equipment industrial
startup, in which one of the authors was both cofounder and chairperson
of the board for the first three years. The purpose of the example is, on the
one hand, to provide an illustration of the three benefits of analyzing startup
organizations. On the other hand, we want to establish how fundamentally
context-​dependent startup governance is. Governance choices are not only
dependent on how the designer formulates the main problem but also on
how idiosyncratic institutional environments (most notably law) impose
constraints on private ordering.

The Case of the Sports Equipment Startup

In August 2015, one of the authors of this book became a cofounder and
chairperson of the board of an industrial startup that designed and man-
ufactured sports equipment as well as developed and built the proprietary
production technology used in production. In the two months preceding
the founding, four prospective cofounders got together to discuss the cen-
tral issues and challenges. A number of prospective product designs were
complete, along with a few prototypes. The prototypes were handmade, and
not even a rudimentary production system had been built. The prospective
cofounders agreed that the common premise of not separating management,
oversight, and risk would provide a useful starting point for thinking about
governance (Figure 6.1).
The first challenge would be to build a production line and scale it up to
a point where the firm could generate sales sufficient to secure a positive
month-​to-​month cash flow. The prospective cofounders concluded that the
firm would have the requisite technical expertise to build the production

Figure 6.1  The startup viewed through the Efficiency Lens

166  Governance and the Organizational Life Cycle

system, if needed. The greatest challenge for the startup was the financing of
assets. Would they be financed through debt or equity?
We commonly think of decisions of debt versus equity financing and
leverage as financial management decisions where the cost of capital is of
central importance. But as Williamson noted, these decisions not only have
governance implications, but they also are governance decisions: It is useful
to regard “debt and equity as governance structures rather than as financial
instruments” (Williamson 1988, 579).
The choice of debt versus equity financing has a number of important
organizational ramifications that link to oversight in particular. In firms
that rely on equity financing, the role of the board of directors is crucial in
securing the rights of the providers of equity capital and the continuing
supply of financing when needed. In a debt-​financed firm, in contrast,
the rights of the financier are stipulated in the formal contract (the loan
agreement), effectively eliminating the need for additional governance
interventions at the board level. If the organization defaults on its debt, the
creditor has several options and safeguards available. More generally, firms
that rely on debt financing tend to organize based on formalization (rule fol-
lowing); a stronger reliance on discretion is found in equity-​financed firms
(Williamson 1988, 581).
However, the financing of assets in and of itself was not construed as the
main problem, because there was a more fundamental question that had to
be addressed before the financing decision could be contemplated: What
kind of production technology would be used? This question would, in
turn, have to be considered simultaneously with the decision of whether the
products would be produced in-​house or by an external supplier—​the ques-
tion was fundamentally one of organizational boundaries. The prospective
cofounders concluded that the main problem should be formulated as a dis-
criminating alignment (see ­chapter 4) of the financial, the technological, and
the organizational, the central question being, “How would the startup en-
sure that the three are in sync with one another?”
The cofounders weighed the different options and concluded that compet-
itive advantage would be sought based on product differentiation. The aim
of the startup would be to introduce a product with a drastically different
structure than the incumbents’ products. The cofounders figured that trying
to compete in a highly consolidated market against large incumbents with
massive scale and scope economies and market power would be challenging.
Trying to enter such a market without significant product differentiation
 The Startup Organization  167

seemed like bad strategy. What must the startup offer that the established
brands that dominated the market did not already offer? The strategic de-
cision to introduce a drastically new product had immediate technological
consequences, which in turn led to a number of fundamental governance
decisions.
The immediate technological consequence was that the startup would have
to develop not only the product but also the production system. Design and
production of some parts and subsystems could be outsourced to external
suppliers, but the startup would have to design and build other subsystems
internally as well as integrate all subsystems—​including those purchased
from external vendors—​into the overall production system; this would re-
quire considerable investments in engineering.
The upside of in-​house production would be that the startup would main-
tain important residual rights of control with regard to production decisions
(see ­chapter 2). In their attempt to be forward-​looking, the cofounders
concluded that in-​house production would confer important advantages par-
ticularly in the growth phase of the startup. Specifically, if the firm designed
and built the production system itself and was in charge of system integra-
tion, it would have control over the entire system. This control would make
the scaling of production easier. The only thing the startup would have to en-
sure was the reliable supply of outsourced parts and subsystems. Achieving
a reliable supply was deemed not to be a problem, because contractual
relationships with technology suppliers were straightforward to formalize
due to low specificity. As an example, consider the supply of aluminum or
steel molds used to make some of the parts. All the startup would need to
provide to the supplier were the technical specifications, which the supplier
would feed into a general-​purpose computerized-​numerical-​control (CNC)
machine1 to produce the mold. No long-​term relationships would be re-
quired, and multiple suppliers would be available for each outsourced sub-
system. It would make no sense for the startup to carry CNC machines on its
balance sheet.

1 The CNC machine is a general-​purpose electromechanical device (e.g., a lathe) that can pro-
duce products for a variety of end uses using diverse materials, such as metals, plastics, or wood.
All the CNC machine needs are instructions from a computer on the dimensions of the product
to be produced. In many industries, an established, competitive network of large and small CNC
machinists offers manufacturing services to a variety of buyers. In addition to there being a large
number of CNC machinists, there are also many manufacturers of CNC machines, which ensures a
competitive market throughout the value chain.
168  Governance and the Organizational Life Cycle

The downside of the decision to build a unique production system was

that most of it would involve either internally developed or engineered-​to-​
order, special-​purpose technology that exhibited a high degree of physical
asset specificity; the system would be a dedicated design, and thus the pro-
ductive assets would have no redeployability. Unique, special-​purpose assets
are notoriously difficult to finance with debt, unless the firm can offer other,
more liquid assets as collateral. Absent such assets, the cofounders concluded
that if the startup was to be committed to a differentiated product, equity fi-
nancing would be the only feasible option.
Given these initial conditions, what kinds of contractual arrangements
would ensure an efficient startup organization? What kind of foresight would
have to be incorporated? We discuss these issues in the following sections by
comparing and contrasting what the law (the institutional pillar) stipulates
and what the cofounders agreed privately (the contractual pillar).

The Institutional and the Contractual Pillars in the Case

of the Sports Equipment Startup

Startup firms tend to incorporate as closed corporations (Fama and Jensen

1983a), that is, limited liability partnerships with an intentionally limited
number of shareholders. Because there is no market for shares, it is consid-
erably more difficult for cofounders to divest; often, the only feasible way to
divest is to try to sell to another cofounder. There are usually not many out-
side investors willing to buy the cofounders’ shares in the early stages of a
startup in particular. Furthermore, it may be a bad idea for a startup to create
an organization where new investors enter, and old investors exit, frequently.
Consequently, cofounders may want to intentionally limit the alienability of
residual rights.
Because ownership in a closed corporation is both unavoidably and de-
liberately “sticky,” cofounders are well advised to agree on the specific rules
by which equity will be governed, to make the organization credible in the
eyes of risk and to ensure effective oversight. This gives rise to a common
contractual pillar—​the shareholders’ agreement. Indeed, the shareholders’
agreement constituted the most important contractual arrangement at the
founding of the sports equipment startup.
Even though the shareholders’ agreement is a matter of private ordering,
the institutional pillar provides a useful starting point, because it establishes
 The Startup Organization  169

the broader context in which private ordering takes place. Cofounders

should obviously be aware of the requirements that corporate law imposes
on startup governance. However, two other issues that merit the designer’s
attention are less obvious: (1) Should some of the defaults in the law be mod-
ified or overridden? (2) What other governance issues not covered in the law
should be incorporated into ex ante contracting?
The sports equipment company was incorporated in Finland; therefore,
the central institutional pillar was the Finnish Limited Liability Companies
Act of 2006. The cofounders used the shareholders’ agreement both to over-
ride some of the defaults in the law and to agree on issues not covered by
the law. Table 6.1 summarizes selected issues for illustration. The table does
not constitute a complete list of all the issues contained in the shareholders’
agreement; we have chosen for illustration a number of specific issues that
either have more general appeal or that effectively illustrate important
differences between the institutional and the contractual pillars.

Board Composition
Predictably enough, as the prospective chairperson of the board approached
potential investors, several of them asked not only who would assume opera-
tional responsibility but also how the board of directors would be assembled.
It became clear that the board would have to have a substantial equity stake
in the startup, and that it would make little sense to appoint any outsiders to
the board at the inception. Cofounders quickly converged to the idea that a
board of directors of at least three members would be selected from among
the cofounders, and that all board members would have to have at least a
10 percent equity stake in the startup. Expressed in the terminology of the
Efficiency Lens, the cofounders wanted to avoid the excessive separation of
oversight and risk. Having a board with only a minimal equity stake may
work in a large corporation, but in a small startup, separating oversight from
risk may immediately threaten the credibility of the organization in the eyes
of both current and prospective providers of equity. Providers of equity con-
stitute the most important stakeholder of a high-​technology startup whose
success hinges on the successful development and productive use of unique,
high-​specificity production technology. If the startup cannot secure the req-
uisite funding, the startup fails, and discussion of all other stakeholder issues
becomes a moot point.
Attention to board composition was relevant not only from the point of
view of securing initial financing. The forward-​looking cofounders knew
170  Governance and the Organizational Life Cycle

Table 6.1  The Institutional and the Contractual Pillars in the Sports Equipment
Startup

Issue The Institutional Pillar The Contractual Pillar

(Finnish Limited Liability (Shareholders’ Agreement)
Companies Act)

Composition of the A minimum of two members Three to five persons; a block

board of directors at least one of whom resides of 30% of shares can appoint
in the European Economic one member
Area (required); appointed by
shareholders (default)
Procedures for Freely alienable (default); right Freely alienable only among
trading shares of squeeze-​out (default) partners; tag-​along and drag-​
along clauses regulate transfer
of shares
Loyalty and care Board members are subject to All shareholders are subject
the fiduciary duty of loyalty to the fiduciary duty of loyalty
and care to the organization but not care
(required)
Non-​compete and None Strict non-​compete and
confidentiality confidentiality clauses
clauses
Contributions clauses None None
Dividend policy Shareholders have the right to No minority dividend
minority dividend (default)
Dispute settlement None Binding arbitration in
Helsinki, in Finnish
Procedures for raising Shareholders decide (default); No additional stipulations or
equity regular issue requires simple overrides of defaults
majority (default); directed
issue requires qualified
majority (default)
Defining majority 1/​2 for majority and 2/​3 for No additional stipulations or
qualified majority (default) overrides of defaults

from the beginning that the startup would likely require multiple rounds of
equity financing. For this to be successful, the firm would have to maintain
credibility in the eyes of the providers of equity. It turned out the firm needed
a total of five rounds of equity financing in the first three years.
Finnish law requires a limited liability company to have a board of
directors. Further, the board must have as members a minimum of two nat-
ural persons at least one of whom resides in the European Economic Area (all
EU countries, Norway, Iceland, and Liechtenstein). The law further stipulates
 The Startup Organization  171

that the decision to appoint board members belongs to shareholders. This

legal requirement means that employees will have a formal oversight role in
the corporation only if they either are also shareholders or if shareholders
choose to appoint an employee representative to the board. Therefore, even
in cases where employees have a legitimate reason for having board represen-
tation, they cannot make a legal claim for inclusion—​legitimate and legal are
not synonyms.
The cofounders decided that a two-​person board would be too small and
that a large board would be problematic. Consequently, in the shareholders’
agreement they supplemented the requirement of a two-​person minimum by
requiring the board to have three to five members. They also supplemented
the default that directors are elected by shareholders by allowing a block of
30 percent to appoint a board member. Without this override, all prospec-
tive board members would have to be supported by a simple shareholder
majority.
Some organizations may delegate the selection of board members to the
board itself; such boards are known as self-​perpetuating boards. In a self-​
perpetuating board, boards are usually not allowed to increase board size by
adding new members at their discretion; they are merely allowed to select
replacements for outgoing members. Such autonomous boards are found
in nonprofit settings (Hansmann 1980). In the sports equipment startup, a
self-​perpetuating board made no sense: All board members would always be
selected in the annual shareholders’ meeting, with the exception that a voting
block of 30 percent could appoint a board member.

Procedures for Trading Shares

The law not only places no limitations on the ownership of shares, but the
very spirit of the law is also to ensure unrestricted transfer of wealth. At
the same time, unrestricted trading of shares may present a problem in
a high-​risk startup because it may invite instability and expose the startup
to unwanted takeovers. The resulting hazard may discourage prospective
shareholders from investing.
The shareholders’ agreement of the sports equipment startup placed sig-
nificant restrictions on the trading of shares, particularly if a trade would re-
sult in the introduction of a new shareholder. Consequently, the cofounders
decided that existing shareholders could trade shares among themselves
without restrictions; some cofounders accumulating more decision and
voting power in the organization over time was not deemed problematic.
172  Governance and the Organizational Life Cycle

The cofounders further thought that categorically prohibiting the intro-

duction of new shareholders would not be in the best interest of the organiza-
tion. If one cofounder identified an attractive prospective shareholder but no
additional equity financing was needed at the moment, why would existing
shareholders be prohibited from selling some of their shares to the new en-
trant? The cofounders decided that this option should be made possible but
that it would be done on a pro rata basis. This decision led to the introduction
of a tag-​along clause in the shareholders’ agreement. The tag-​along clause
stated that if a shareholder identified a prospective buyer, all shareholders
would be entitled to “tag along” into the sale based on their ownership share.
For example, if a prospective buyer of one thousand shares was identified and
the existing shareholders each owned 20 percent, each shareholder would be
entitled to “tag along” by offering to sell a maximum of two hundred shares.
Importantly, it is generally well advised to make “tagging along” optional;
including a clause that could effectively force the dilution of ownership is
thought to be bad governance. The tag-​along clause effectively introduces a
put option to existing shareholders.
Another set of rules for trading shares must be in place in the potential
acquisition of the startup by another firm. Since many startups are inter-
ested in being acquired by a larger firm at some point, exit conditions must
be stipulated. An exit usually means that the acquiring firm purchases all the
target’s outstanding shares. The law provides protection against involuntary
sale of shares unless ownership is highly consolidated. For instance, Finnish
law stipulates that a shareholder with more than 90 percent of all shares has
the right “to squeeze out” the remaining minority shareholders at a fair price.
The default gives any shareholder with an equity stake of more than
10 percent the option to effectively block an acquisition. To provide a safe-
guard against such an economic holdup problem by a minority shareholder,
shareholders’ agreements often incorporate a drag-​along clause. The drag-​
along clause states that in the event a third party makes a good-​faith offer to
acquire the firm, a majority (simple or qualified) of shareholders can “drag
along” the rest to the sale. In the sports equipment startup, the shareholders’
agreement contained a drag-​along clause based on a simple majority. The
drag-​along mechanism effectively functions as a call option to purchase
shares from existing shareholders.
Restrictions on the trading of shares can be viewed as a contractual mech-
anism that safeguards against one shareholder benefiting from the transfer
of their shares at the expense of other shareholders. Without the tag-​along
 The Startup Organization  173

clause, each shareholder would effectively face a unique market for the
startup’s shares; more powerful and better-​informed shareholders might be
able to use this to their advantage. Without the tag-​along safeguard in place,
more passive prospective shareholders might be reluctant to invest. Without
the drag-​along clause, shareholders would expose themselves to a holdup
problem, as a minority shareholder could effectively block even an economi-
cally attractive acquisition offer.

Beyond Limited Liability

The question of who is expected to be loyal to the organization and why
requires explicit attention in startups. The law assigns the legal fiduciary
duty of loyalty and care only to the board of directors. In contrast, aside
from the requirement to pay for their shares, shareholders do not have to
work, attend meetings, make decisions, be loyal to the organization, or pro-
vide additional financing. This feature goes to the heart of the principle
of limited liability: Purchasing shares must not be burdened by further
obligations.
Shareholders’ agreements can, and often do, impose additional obligations
and restrictions on shareholders, effectively partially compromising the prin-
ciple of limited liability. The designer must exercise caution here and craft all
additional obligations carefully to secure the credibility of the organization
in the eyes of both current and potential future shareholders. To the extent
the organization’s viability hinges on successfully raising equity in multiple
financing rounds, all additional obligations pose a hazard as they limit the
pool of prospective investors.
In the shareholders’ agreement of the sports equipment startup, the
cofounders agreed that the fiduciary duty of loyalty would be extended to
all shareholders. This was achieved by requiring all shareholders to act in the
best interest of the organization. Moreover, shareholders would be prohib-
ited from entering into organizational arrangements in which they would be
in competition with the startup.
How about the fiduciary duty of care? Should it be extended to all
shareholders as well? This is a more complex question, because the duty of
loyalty can largely be fulfilled by passively refraining from doing something.
In contrast, fulfilling the duty of care requires actively doing something.
Extending the duty of care to shareholders implies that all shareholders would
be required to put in at least some work, if only to ensure that good decisions
are made. This requirement would effectively lead to a contributions clause
174  Governance and the Organizational Life Cycle

of some kind. Contributions clauses are some of the most contested issues in
shareholders’ agreements.
After careful deliberation, the cofounders decided neither to extend the
duty of care to shareholders nor to include any contributions clauses, be-
cause these would likely not result in net gains. In fact, one central cofounder
and another key investor who joined the firm in the fourth financing round
would not have contributed to equity if the shareholders’ agreement had in-
cluded a contributions clause.
The designer must always consider the option that some investors only
want to invest in equity with no role, formal or informal, in management or
oversight. Consequently, embracing the democratic idea that “everyone also
put in the work” may be misplaced. As always, an analysis of net gains must
be conducted, although we are doubtful that a contributions clause would re-
sult in net gains in startups that rely heavily on equity financing.

Dividend Policy
Designers of startups that seek to become attractive targets for an acquisi-
tion must think carefully about how they manage a potential economic sur-
plus. In many startups, cofounders agree that the startup should retain all
its earnings and that the shareholders’ payday would be the day the firm
is acquired by another firm. Consequently, the startup might choose not
to pay dividends at all. If this is the case, it may be best to write it into the
shareholders’ agreement.
The importance of an explicit dividend policy is particularly important
in Finland, where the law contains an interesting idiosyncrasy not found in
many other jurisdictions. Specifically, Finnish law goes further than many
others in protecting the rights of minority shareholders. One of these rights
is the principle of minority dividend, which states that a block of shareholders
representing a minimum of 10 percent of all shares can force at least one half
of the profits for the fiscal year to be distributed as dividend. In the sports
equipment startup of eight cofounders with roughly equal equity stakes, the
right to minority dividend would mean that any one shareholder could force
a dividend.
The minority dividend is a default that can be overridden by an ex
ante agreement of all shareholders. In the sports equipment startup, the
cofounders unanimously agreed that all shareholders, current and future,
would have to forgo the right to the minority dividend. The startup would
have to be able to retain all its earnings to enable sufficient growth.
 The Startup Organization  175

Settling Disputes
The law is silent on dispute resolution, which effectively means the contracting
parties decide the most appropriate form of settling disputes. Most startups
choose binding arbitration as the preferred method. Arbitration has three
potential advantages over litigation: It is cheaper, faster, and has an expert
presiding over the dispute. The cofounders of the sports equipment startup
decided that disputes would be settled in arbitration. An additional stipula-
tion was that arbitration would take place in Helsinki and that the language
of the arbitration procedures would be Finnish.
In retrospect, the author who was one of the cofounders has concluded
that the decision to default to the common practice of binding arbitration
was ultimately ill informed. In a conversation with a legal expert about five
years after the shareholders’ agreement had been signed, the author learned
that in Finland, arbitration is in fact often much more expensive than lit-
igation. This point merits attention because this may not be merely an
idiosyncrasy of the Finnish context. In his review of arbitration as a dispute-​
resolution mechanism, Stipanowich (2010, 1) noted that “[o]‌nce promoted
as a means of avoiding the contention, cost and expense of court trial,
binding arbitration is now described in similar terms—​‘judicialized,’ formal,
costly, time-​consuming, and subject to hardball advocacy.” Lord Michael
Mustill, a British barrister and judge, expressed the concern in slightly more
vivid terms by describing arbitration as having “all the elephantine labori-
ousness of an action in court, without the saving grace of the exacerbated
judge’s power to bang together the heads of recalcitrant parties” (cited in
Stipanowich 2010, 23).

Minor Issues
There were a number of additional procedural and technical issues that the
shareholders’ agreement had to address. For example, the law protects the
dilution of any shareholder’s equity by permitting a directed issue of shares to
new shareholders only in exceptional circumstances. By default, equity must
be raised through a regular issue of shares where existing shareholders are
entitled to buy the newly issued shares on a pro rata basis. Under Finnish
law, a directed issue requires both a qualified majority of two-​thirds of
shareholders and a positive affirmation from the board that a regular issue
is infeasible. The cofounders were satisfied with what the law stipulated
and could not find any compelling reasons why the authority to raise eq-
uity should be delegated to the board of directors as a general rule. Instead,
176  Governance and the Organizational Life Cycle

the cofounders decided that potential delegation would be deliberated in

a shareholders’ meeting on a case-​by-​case basis. Shareholders ultimately
delegated the authority to raise equity to the board in every financing round.
Another technical issue was the definition of majority and qualified ma-
jority. The cofounders saw no reason to deviate from the default definition of
majority as fifty percent of shares plus one share and qualified majority as two
thirds of shares plus one share. The concern was that raising either threshold
might cause delays in decision-​making.

Summary of the Sports Equipment Startup

The preceding discussion and table 6.1 highlight the fact that the ability to
raise equity financing was both foundational and critical to the sports equip-
ment startup. Consequently, all the foundational governance decisions would
have to be derived from and justified by this main problem. Let us briefly dis-
cuss three governance implications of this main problem formulation.
First, restricting alienability of residual claims signals stability and long-​
term commitment, and convinces the shareholders that they need not
worry about the arbitrary introduction of new investors by one or a few
shareholders. Note that this restriction is an ex ante decision where the ob-
jective is to establish credibility in the eyes of prospective cofounders: What
kind of a governance structure will convince prospective investors to make
a wager in the organization? It is understandable that an investor may not
be interested in a “revolving-​door startup” where individual investors can
enter and exit without any oversight or joint decision-​making, or where a
minority shareholder can effectuate substantial changes to the ownership
structure by unilaterally introducing new shareholders. Note that although
neither tag-​along nor drag-​along clauses remove the residual rights from the
shareholders, they do effectively transfer some of the residual rights of con-
trol from shareholders as individuals to shareholders as a collective.
Second, the company was founded with a total of eight cofounders, each
with roughly an equal number of shares. The cofounders were aware that
having an unusually large number of shareholders at the outset not only
constituted a trade-​off but that it also ran against conventional wisdom. The
most obvious problem was that any decision that required shareholder ap-
proval would have to be supported by at least five of the eight cofounders.
Furthermore, the larger the founder base, the more complex the coordination
 The Startup Organization  177

and the communication. At the same time, the cofounders agreed that a
broad ownership base made sense given the main problem of securing the
requisite equity financing both at the outset and in foreseeable future fi-
nancing rounds. The startup could have had fewer founders, but this would
have meant that it might not have been able to raise equity from its existing
shareholders in future financing rounds. The cofounders decided to tackle
the challenge of having to manage a relatively large number of shareholders
already at the founding as an ex ante problem. This was successful in the
sense that in the subsequent four rounds of financing, only one new share-
holder was introduced.
The potential problem of cumbersome decision-​making due to a large
number of cofounders was alleviated by an active board of four members
who collectively owned a majority of the shares. In fact, the four-​member
board was deliberately assembled from cofounders who had a somewhat
higher equity stake than the others. This governance arrangement meant that
the board could de facto make decisions that belonged to all shareholders.
Instead of calling a shareholders’ meeting, the board could simply reach out
to all shareholders and propose a consensus decision. The board would in-
form the shareholders that it had reached a unanimous decision on an issue,
which meant that the proposal had already secured majority approval. If
all shareholders agreed to the proposal that everyone knew would pass an-
yway, the decision could be written and filed expediently as a unanimous
shareholders’ decision, without having to call a shareholders’ meeting.
Third, governance decisions strongly echoed the idea of limited liability.
Because all commitments above and beyond putting money at risk upfront
and being loyal to the organization might alienate prospective investors,
the shareholders’ agreement did not contain contributions clauses or
commitments to further financing. Moreover, the board of directors took
shareholder interests as its primary objective. This priority did not mean that
there were no other stakeholders; indeed, just the opposite. Many employees
would be required to commit to developing firm-​specific skills, which meant
that they had a legitimate residual interest as well. However, this was not
something that the cofounders considered so central that it would require
the board’s attention at the beginning. Again, the main problem was that
the organization would have to secure a steady flow of equity financing until
the firm could be financed by revenue. Had debt financing been possible,
the board could have been given an altogether different primary task. As we
mentioned earlier, in equity-​financed firms, discretion in decision-​making is
178  Governance and the Organizational Life Cycle

central. Here, it is precisely the board of directors whose task is to use discre-
tion by deliberating on the key financing issues. If financing were organized
through debt, the board could direct its attention away from financing is-
sues, which in debt-​financed firms are more an issue of compliance and rule
following. Understanding the intimate linkage between finance and govern-
ance is central (Williamson 1988).

Understanding Context Dependence

In the case of the sports equipment startup, the board attended primarily
to shareholder interests because the main problem was defined in terms of
raising equity. The main problem is always organization specific, and conse-
quently, so are all the governance choices derived from it. In a setting where
assets can be financed through debt or where the startup can quickly start
to rely on revenue financing, there is no need to define the main problem in
equity-​financing terms. Consequently, the board of directors can be assigned
a very different oversight role. In sum, context dependence enters into gov-
ernance decisions through the contextually idiosyncratic main problem.
Another way in which context dependence enters governance decisions
is through the institutional environment. Although many of the topics and
issues in table 6.1 generalize to other settings, every environment exhibits
idiosyncratic aspects as well. For example, not all jurisdictions require the
limited liability company to have a board of directors. In Spain, one of the
founding decisions the designer of a limited liability company (sociedad
limitada) indeed faces is whether the company will have a board of directors.
In Finland, the law requires limited liability companies to have a board of
directors. Another idiosyncrasy of the Finnish legal environment is the prin-
ciple of the minority dividend whereby a voting block of 10 percent can force
a dividend for the fiscal year. The designer must incorporate the demands of
the idiosyncratic institutional environment into governance decisions.

The General Case of Startup Governance

The sports equipment startup example presents one specific case of startup
governance and offers an illustration of its context dependence. Even though
all governance decisions are fundamentally context dependent, there are a
number of broader issues that merit attention. In this section, we discuss
three topics that we see as generally applicable: (1) ex ante contractibility,
 The Startup Organization  179

(2) enforceability of private ordering, and (3) assignment of ex post decision

rights.

Ex Ante Contractibility

Think back to your first summer job delivering newspapers, mowing lawns,
or selling ice cream (see c­ hapter 4). We venture to guess your contract with
your employer required neither additional contractual safeguards nor ex
post adjustments. More generally, many relationships the organization has
with its constituencies are ex ante contractible and, therefore, can also often
be safeguarded ex ante. The more the pertinent contracting issues can be
addressed and formalized ex ante, the more the contract acquires the charac-
teristics of a complete contract.
Not everything about a relationship is contractible. Noncontractibility
typically arises when the desired outcomes of contracting are uncertain.
For example, a high-​technology firm can use employment contracts to hire
R&D experts to work on various development projects. Whereas the firms’
managers and the hired R&D experts can further formally agree on the scale
and the scope of the R&D efforts, the outcomes of these uncertain innova-
tion efforts tend to be noncontractible. For example, the contracting parties
cannot (or at least probably should not try to) contract for the number of
patents filed or the amount of revenue the new products and services devel-
oped will create. These are issues that must be addressed ex post. Similarly,
a team of cofounders may agree that their startup will seek to be acquired
by a large incumbent, but the exact timeline and the sale price are ex ante
noncontractible.
Noncontractible does not mean nonaddressable. When an issue is
noncontractible, it simply means that instead of approaching the governance
of the relationship in terms of a formal ex ante contract, the designer must
think of ways in which emergent issues will be addressed through ex post ad-
justment. In the following, we discuss two examples.

Ex Post Adjustment in Risky R&D Projects

In ­chapter 3, we presented R&D collaboration as an example of a situation
in which efficient contracting is relevant. Specifically, to manage uncertain
interfirm collaboration, the designer must determine whether a joint eq-
uity alliance offers a more efficient approach to adapting to uncertainty than
180  Governance and the Organizational Life Cycle

collaborative contracting. This question effectively presents two governance

options as alternatives: one emphasizes an ex ante contract and the other ex
post adjustment.
That collaborative contracting presents the ex ante alternative is
straightforward. Those engaging in R&D collaboration agree to an ex ante
contract that stipulates the rights and the responsibilities of both parties.
In case something unexpected that is not covered by the contract emerges,
the contracting parties must adjust the contract through ex post negotia-
tion. The ex post adjustments effectively create a new ex ante contract to
be enforced in the future. Even though adjustments are made ex post, the
general idea is to address problems by ex ante specification.
Adjusting formal contracts becomes cumbersome in collaborations
beset with high uncertainty and ambiguity. Instead of seeking “to con-
tract the noncontractible,” the collaborating parties may decide to
choose a governance structure that assigns the decision rights over
ex post adjustments to a separate legal entity. In R&D collaboration, a
joint equity alliance constitutes such an entity. In a joint equity alliance,
considerations that are ex ante contractible are embedded in the alliance’s
founding documents. However, instead of trying to anticipate the po-
tentially numerous ex post adjustments ex ante, the designer assigns the
decision rights over the requisite adjustments to the alliance, more spe-
cifically, to its board of directors. The board will convene whenever an
issue requires attention and then either addresses the issues directly or
delegates them to the alliance’s management. Under conditions of high
uncertainty, this governance arrangement can be more efficient than col-
laborative contracting.

Ex Post Adjustments in Determining Share Prices in Unlisted Companies

How do shareholders of unlisted companies determine the price at which
shares are bought and sold? In contrast with publicly traded companies
where shares have an unambiguous market price, determining the share
price of a closed corporation requires an explicit valuation procedure. Even
though the share price at any future date is ex ante noncontractible, there
are ways in which a forward-​looking designer can address noncontractibility
ex ante. This merits attention, because as legal scholar John Ghinger (1974,
225) noted, the way shareholders’ agreements approach noncontractibility
leaves much to be desired:
 The Startup Organization  181

It is ironic that lawyers who are extremely careful in the conception and
drafting of transfer restrictions and death buy-​out provisions are content to
permit the most significant provisions of the agreement, those dealing with
purchase price and payment thereof, to speak loosely in terms of “book
value,” “annual installments,” and the like.

When the buyer and the seller enter the transaction voluntarily, the req-
uisite ex post adjustment will occur autonomously as the buying and the sel-
ling parties negotiate the price—​no additional safeguards are required. But
how should the organization safeguard situations in which transfer of shares
is involuntary? For example, a shareholders’ agreement may stipulate that
shareholders who violate the shareholders’ agreement must relinquish their
shares. However, instead of adopting purchase price provisions based on po-
tentially vague terms such as fair value, an informed shareholders’ agreement
might stipulate the procedure by which shares will be valued in the case of
an involuntary transfer. Agreeing on the procedures would likely not solve
all problems, but it would make the ex post decision of share price determi-
nation easier compared to a situation in which the shareholders’ agreement
stipulated purchase “at fair value.” The valuation of young firms is particu-
larly challenging because cash flows are uncertain, and the book value of as-
sets is likely not a useful measure. Agreeing on the basic rules by which the
value of the underlying business would be calculated paves the way toward
an informed analysis.
Even though all ex post adjustments are ultimately a matter of reacting to
unexpected events as they occur, giving them ex ante attention by agreeing
on the procedures that will be followed in the case an adjustment is required
has one substantial advantage. Specifically, when the contracting parties ad-
dress potential ex post adjustments ex ante, they share a common interest
of defining what is fair from the point of view of the organization. Once an
actual dispute occurs, entrenched positions quickly emerge as the disputing
parties feel compelled to defend their own positions. It is more efficient to
agree on the rules of conflict resolution before disputes occur, that is, when
the transacting parties view potential (not actual) disputes from behind
what philosopher John Rawls (1999) called the veil of ignorance. When
the contracting parties do not know how they themselves might be af-
fected, “they are obliged to evaluate principles solely on the basis of general
considerations” (Rawls 1999, 118).
182  Governance and the Organizational Life Cycle

Enforceability of Private Ordering

Overzealous and uninformed reliance on private ordering may lead to unen-

forceable contracts. To be sure, safeguards that provide a false sense of security
can be significant sources of inefficiency. The contractual hazard is pronounced
in situations in which young, inexperienced entrepreneurs launch their first
startup. Strapped for cash, the cofounders may choose to work out the founda-
tional contractual arrangements without the assistance and expertise of legal
counsel. However, the decision to avoid ex ante transaction costs may be my-
opic and, consequently, result in considerable ex post transaction costs.
Whenever contracting parties agree to do something that is inconsistent
with the law, the law prevails. Although this point may seem self-​evident,
sometimes determining what is and what is not consistent with the law is
not straightforward. Misguided entrepreneurs may think that all they need
do is ensure that they know “what the law says.” In reality, the letter of the
law often provides merely broad guidelines, which will then be interpreted
in specific cases.
The designer must exercise caution whenever private ordering seeks to
adjust or to override either the legal rights or the legal obligations of the
contracting parties. Uninformed designers may engage in excessive (and
wasteful) private ordering that ultimately infringes on the legal rights of the
contracting parties. A common misconception is to think that as long as all
contracting parties enter into the agreement voluntarily, the contract will be
enforced by the courts as well. Two examples illustrate that this assumption
may not hold.

Contracting for Market Failure? Limits on the Right to Buy and

Sell Property
It is understandable that the cofounders of a high-​risk startup firm will want to
safeguard against the intrusion of unwanted new shareholders. The designer
may also want to decide whether individual shareholders can exit the firm by
taking advantage of private information and selling all shares to a third party,
leaving other shareholders “stuck” with the new investor. Although some
restrictions may be desirable, the designer must tread carefully not to exces-
sively limit the transferability of stock. Ghinger (1974, 215) noted that when
courts are asked to assess the reasonableness of restrictions placed on the
transfer of shares, “it is certain that any absolute prohibition against transfer
will fail.” Indeed, absolute prohibition would effectively be the equivalent of
 The Startup Organization  183

contracting for market failure: There might be a situation in which a buyer

and a seller have agreed on a price at which the shares would be transferred
from the seller to the buyer, but a contractual restriction prevented the trans-
action, effectively inducing a market failure.
In the case of limiting transferability of shares, limitations based on the
principle of equal treatment of all shareholders are usually not only legally
sanctioned but also efficient, forward-​looking governance. Equal treatment is
at the heart of tag-​along and drag-​along clauses described earlier. Tag-​along
clauses ensure that no shareholder is allowed to benefit from private infor-
mation at the expense of the others; drag-​along clauses prevent an individual
shareholder from holding the others hostage. Tag-​along and drag-​along
clauses help build startup credibility in the eyes of particularly those who join
the firm simply as investors, that is, residual claimants who seek no role in
management or oversight. Courts tend to view both tag-​along and drag-​along
clauses as reasonable and justified restrictions on the transfer of shares.

Can the Parties Privately Contract Not to Sue One Another?

Many shareholders’ agreements contain a clause that requires disputes to be
resolved through private ordering, most commonly by binding arbitration.
The sentiment is salient, but the implications are not. Does this mean the
contracting parties have abdicated their right to sue one another no matter
what? What if there is a reason to suspect one contracting party has acted il-
legally? The issue is complex because making a promise not to sue and, sym-
metrically, receiving assurances that one will not be sued, are both legally
problematic. If one party decides not to honor the agreement not to sue, will
the courts enforce the mutual commitment to private ordering? They may not.
A shareholders’ agreement by which shareholders commit to alternative
dispute resolution should not be viewed as a legally binding obligation to
use private ordering in all disputes. For example, by signing a shareholders’
agreement with an arbitration clause, shareholders do not relinquish their
right to sue the board of directors for violations of their fiduciary duty, nor do
they preempt shareholders from filing a police report if they have a reason to
believe that the board has committed a criminal act. In the United States for
instance, the First Amendment to the Constitution protects “the right to pe-
tition the Government for a redress of grievances.” Any private contract that
limits this right is profoundly problematic.
On the issue of what can be privately contracted, we counsel the designer
to avoid optimism and to err on the side of caution: If in doubt, assume the
184  Governance and the Organizational Life Cycle

safeguard is less effective than intended. For example, agreeing on binding

arbitration should be viewed as a sincere wish to settle conflicts privately, not
as a legally binding contract that the courts will uphold. Whether a specific
clause in a contract is enforceable is not up to the contracting parties, or some-
times even an arbitrator, to decide: “[A]‌rbitrators cannot arbitrate disputes
where shareholders seek the rescission of the very shareholders’ agreement
which empowers the arbitrators to intervene” (Brownlee 1994, 309).
Finally, legal scholar Owen Fiss (1984) offers an intriguing angle to private-​
ordering approaches that merits attention. Fiss (1984, 1085) argued against al-
ternative dispute resolution because it exhibited the same shortcomings as plea
bargaining:

Settlement is for me the civil analogue of plea bargaining: Consent is often

coerced; the bargain may be struck by someone without authority; the
absence of a trial and judgment renders subsequent judicial involvement
troublesome; and although dockets are trimmed, justice may not be done.
Like plea bargaining, settlement is a capitulation to the conditions of mass
society and should be neither encouraged nor praised.

The proposition that there are situations in which justice may be more im-
portant than peace is worth the designer’s consideration.

Assignment of Ex Post Decision Rights

Whenever the future is uncertain and contracts materially incomplete, the

need for ex post adjustments arises. When an issue is not ex ante contractible,
the designer must decide how to allocate the residual rights of control effi-
ciently (Grossman and Hart 1986, 696). For example, who should have the
authority to make the following adjustment decisions?

1. Increase the size of the board of directors.

2. Appoint a board member.
3. Change the way by which disputes are resolved.
4. Adopt antitakeover provisions (see c­ hapter 7).
5. Adjust the way by which fair value of equity is determined in a closed
corporation.
 The Startup Organization  185

All these ex post adjustment decisions, and scores of others, require the
designer’s attention at the founding of the organization. Even though the
specific adjustment decisions are made ex post, the designer must create
the requisite ex ante rules that govern how these decisions will be made
and by whom. In Efficiency Lens terms, principles of oversight regarding
adjustments must be delineated ex ante to guide ex post management
decisions.

Which Residual Rights Are the Board’s Prerogative?

In limited liability companies, ex ante decisions regarding residual rights of
control commonly involve deciding which issues require shareholder ap-
proval and which are the prerogative of the board of directors. Giving the
board too little, or too much, power can lead to significant inefficiencies. On
the one hand, if boards have too little power, decision-​making becomes cum-
bersome. Stringent limits on the board’s ability to act unilaterally also send
a mixed signal to board members: Why does the designer grant a group of
persons a central fiduciary role only to strip them of the authority to make
decisions? Does the designer trust the board or not? Moreover, if the board
has only limited discretion, will it ultimately be held responsible for some-
thing over which it has no control? Here, understanding the essentially
fiduciary (not contractual) role of the board is crucial—​the board cannot pri-
vately contract out of its legal duties to the organization. On the other hand,
if the board has too much power, entrenchment and agency problems may
pose serious hazards. We return to the topic of entrenchment when we dis-
cuss antitakeover provisions in c­ hapter 7.

Summary: Understanding Governance by Contract

This chapter on incipient organizations complements ­chapter 3, which

addressed contractual relationships within and across organizations. In this
chapter, we emphasize the importance of viewing the incipient organization
through the lens of “governance by contract” (Fisch 2018). To this end, the
designer has three tasks:

1. identify which issues are ex ante contractible and for which efficient ex
ante safeguards can be implemented;
186  Governance and the Organizational Life Cycle

2. identify which issues are not ex ante contractible but require ex post ad-
justment; and
3. allocate the residual rights of control regarding potential ex post
adjustments.

We caution against excessive optimism regarding the efficiency of ex ante

contracting. A significant hazard arises from mistakenly thinking that the ex
ante contractible is also ex post enforceable. Not falling prey to a false sense
of security requires an understanding of the limits that the institutional en-
vironment sets on private ordering. This issue likely requires legal exper-
tise and experience in the specific jurisdiction in which the organization is
embedded. Even though spending precious shareholders’ equity on lawyer’s
fees at the founding of a startup sounds unappealing, these fees likely pale in
comparison with the ex post adjustment costs that arise when the contracting
parties realize that their contracts are materially unenforceable.
We similarly counsel against overlooking the importance of conscious
foresight regarding ex post adjustments. Specifically, the general principles by
which ex post adjustments are made should be considered matters of ex ante
agreement. Although designers cannot provide ex ante solutions to ex post
problems, they can, and should, agree on which procedures will be followed
and who exercises the residual rights of control in these procedures. When
residual rights of control are poorly defined, a common problem we have
witnessed is that shareholders adopt the attitude that the board of directors
is directly responsible only to them and that they can give the board advice
with the expectation that the advice is always to be followed. However, the
idea that the board (the fiduciary of the organization) is accountable only to
shareholders (merely one of the stakeholders of the organization) is problem-
atic. This notion further becomes untenable when ownership is fragmented,
and consequently, shareholder interests are not only heterogeneous but also
potentially conflicting. In addressing the potentially volatile situation of con-
flicting interests, both the designer and the board can take comfort in the fact
that there is nothing in corporate law that establishes the board’s exclusive
accountability to shareholders. The board is well advised to give attention to
shareholders, but this must be done without compromising the two essen-
tial characteristics of the fiduciary role: independent judgment and responsi-
bility jointly to all residual claimants (Blair and Stout 2001).
 7
The Expanding Organization

As organizations expand, they experience two dynamics, which can both be

described using the Efficiency Lens. The more intuitive of the two is what we
label the growth dynamic, which refers to the increasing scale and scope of
management, oversight, and risk:

1. When the organization grows, planning and coordination challenges

become more complex. To adapt to the expanding scale and scope,
the organization decentralizes, adds vertical layers to its structure,
establishes lateral cross-​unit linkages to enhance collaboration, and
so on. Furthermore, international expansion and product diversifica-
tion may prompt the organization to adopt a multidimensional matrix
structure with complex reporting relationships. These are examples of
the expanding scale and scope of management.
2. As the organization grows, it needs both to deepen and to broaden the
mechanisms by which it exercises oversight over management. The
organization adds more members to its board of directors, amends
its bylaws, and implements oversight functions at all levels of the or-
ganization: The board of directors exercises oversight over the top
management team, top management team over divisional manage-
ment, divisional management over functional management, and so
on. As Jensen and Meckling (1976, 309) aptly noted, principal-​agent
relationships exist at every level of management in all organizations,
which also gives rise to the need for oversight at all levels. Finally, as a
closed corporation goes public through an IPO, the institutional envi-
ronment starts to impose more elaborate disclosure requirements on
the organization. These are examples of the expanding scale and scope
of oversight.
3. As a limited liability company raises more equity to finance its oper-
ations, the scale of its investor risk increases. Scope of risk expands as
well through various fundamental transformations that convert some
employee groups, key suppliers, and large customers into stakeholders

Efficient Organization. Mikko Ketokivi and Joseph T. Mahoney, Oxford University Press. © Oxford University Press 2023.
DOI: 10.1093/​oso/​9780197610282.003.0007
188  Governance and the Organizational Life Cycle

over time (see c­ hapter 4). Commitments to specificity through relation-​

specific investments lead to bilateral dependency that broadens the
pool of constituencies with a residual interest. These are examples of
the expanding scale and scope of risk.

The growth dynamic has been thoroughly covered in the published lit-
erature on organization design, most notably under the rubric of organiza-
tional redesign. Since we have little to add to the conversations on the general
challenges of the expanding scale and scope of organizations, we direct the
reader to excellent treatments of this topic in the published literature.1
In this chapter, we focus on the more elusive separation dynamic. An
expanding limited liability company offers an illustration. As the company
expands, the scale of its oversight increases, which is manifested by the ad-
dition of members to the board of directors. However, because the new
members are increasingly likely to be outsiders (particularly if the firm is
heading toward an IPO), the expansion of oversight tends to be associated
with the separation of oversight from management. An obvious manifes-
tation of separation is the declining proportion of insiders on the board of
directors. Moreover, as the number of individuals and collectives who bear
risk increases, not everyone can be afforded (or even wants) an oversight
role. This effectively separates oversight from risk. Finally, as those who make
the most important strategic decisions bear a smaller portion of the wealth
consequences of their decisions, management and risk become separated as
well. Figure 7.1 illustrates the mutually reinforcing role of the growth and the
separation dynamics. Specifically, expansion of management, oversight, and
risk occurs in directions that leads to increasing separation.
The reason we focus in this chapter on the separation dynamic is twofold.
One is that it has received less attention than the growth dynamic in the or-
ganization design and governance literature. The other reason is that we find
the separation dynamic to be in many ways more foundational to governance
than the growth dynamic. Indeed, the separation of ownership and control
has stood at the very foundation of corporate governance for nearly a century.
In their classic The Modern Corporation and Private Property, legal scholar

1 Excellent practitioner-​oriented texts on the managerial challenges of organizational growth are

organization design scholar Jay Galbraith’s two books Designing Matrix Organizations That Really
Work (Galbraith 2009) and Designing Organizations (Galbraith 2014). The rich literature on scaling
up organizations is also relevant; we recommend in particular The Founder’s Dilemmas (Wasserman
2012). More academically oriented works include the 1959 classic The Theory of the Growth of the
Firm (Penrose [1959] 1995) and Scale and Scope (Chandler 1990).
 The Expanding Organization  189

Figure 7.1  The growth dynamic and the separation dynamic of the expanding
organization

Adolf Berle and economist Gardiner Means (1932, 7) noted that “[t]‌he sepa-
ration of ownership from control produces a condition where the interests of
owner and of ultimate manager may, and often do, diverge, and where many
of the checks which formerly operated to limit the use of power disappear.”
The separation dynamic continues to be essential to governance because
it presents the designer with a dilemma, succinctly expressed by Grossman
and Hart (1980, 42): “In all but the smallest groups social choice takes place
via the delegation of power from many to few. A fundamental problem with
this delegation is that no individual has a large enough incentive to devote
resources to ensuring that the representatives are acting in the interest of the
represented.” The reason this fundamental problem constitutes a dilemma is
that although separation of powers and delegation are necessary, monitoring
the actions of those to whom decision-​making is delegated becomes cum-
bersome with increasing separation. A case in point, when those who make
the most important decisions no longer bear the wealth consequences of
their decisions (management separates from risk), how can those who bear
risk ensure adequate oversight? Furthermore, how do those who bear risk
ensure that those to whom oversight is delegated genuinely adopt a fiduciary
role? We unfortunately have plenty of evidence that the separation of powers
gives rise to various hazards.
Understanding separation as a dynamic phenomenon is crucial because
separation of powers is not how organizations are born. In an incipient or-
ganization, the small group of individuals who make the most important
190  Governance and the Organizational Life Cycle

decisions tend to exercise oversight over their decisions and be the principal
risk-​bearers. Even the largest of organizations started small:

Cofounders Larry Page and Sergey Brin started in their Stanford dorm rooms
by building an internet search engine that they brought to market as Google.
The “two Steves”—​Jobs and Wozniak—​started by building a computer circuit
board, Apple I, and selling Jobs’ VW microbus and Wozniak’s calculator to
begin funding its production. (Pollman 2019, 166)

The separation dilemma emerges over time as the organization expands and del-
egation becomes necessary. Importantly, the necessity for separation emerges
not only due to the increasing scale and scope of the organization but also as
something that is mandated by the organization’s environment. For example,
before a closed corporation can be listed on a stock exchange, it must clearly
separate oversight from management by having the majority of board members
be independent. The separation dynamic therefore offers further insight into
the interplay of the contractual and the institutional pillars.

Addressing the Separation Dilemma

In the seven-​year period spanning from the founding to the IPO, Tesla Motors
raised a total of $200 million of equity in a total of six funding events (Series
A through F). In the later financing rounds, Tesla’s ownership was opened to
external investors. Concurrently with the broadening ownership base, the
composition of Tesla’s board of directors expanded from a board of three
cofounders—​Marc Eberhard, Elon Musk, and Marc Tarpenning—​to a board
with nine members. Predictably, the addition of board members occurred in
lockstep with the addition of shareholders. For example, in Series B, Valor Equity
became a shareholder and its owner Antonio Gracias a board member; in Series
C, Draper Fisher and VantagePoint became shareholders, both appointing a
board member. The first independent board member was not appointed until
Series F. Concomitantly with Series F, Tesla also established an audit committee,
a compensation committee, and a nomination committee to its board. Various
board committees were established in preparation for the IPO.2

2 Nasdaq corporate governance guidelines require that a corporation listed on the Nasdaq stock
exchange have audit and compensation committees that consist of exclusively independent directors.
Having a nomination committee is optional.
 The Expanding Organization  191

Figure 7.2  Tesla’s expansion viewed through the Efficiency Lens

Viewing Tesla’s expansion through the Efficiency Lens is straightfor-

ward. As Tesla expanded, it followed a simple rule: Gradually separate
management, oversight, and risk from one another. There was no material
separation of the three either during Tesla’s founding or during the first
years of its expansion. To be sure, the scale and the scope of management,
oversight, and risk increased: The scale and the scope of Tesla’s opera-
tions were increased, more board members were added, and more equity
was raised. At the same time, a close examination of the Series A through
F issues and the concomitant changes in Tesla’s governance show that
during the early years of Tesla’s expansion, management, oversight, and
risk remained concentrated around a small number of individuals and
investors. Separation did not become necessary until Tesla headed to its
IPO in 2009. Figure 7.2 summarizes Tesla’s path from the founding to the
IPO, viewed through the Efficiency Lens.
192  Governance and the Organizational Life Cycle

Why Going Public Requires Separation

When a limited liability company goes public, it starts to offer its shares to an-
yone willing to pay the market price. As we have mentioned earlier, the word
public is a misnomer in the sense that limited liability companies remain pri-
vately owned even when shares are publicly traded. It seems misdirected to
use the word public to describe an organization when the word is merely a
characteristic of the institutional context (the stock exchange) in which pri-
vate transactions are executed.
At the same time, there are two senses in which the word public, although
not strictly speaking descriptively accurate, does provide a useful label. One
is that publicly traded securities are no longer an exclusively private concern.
For one, although the exchange of publicly traded securities involves private
actors, trading takes place in a centralized institutional setting—​the stock
exchange. In the stock exchange, securities are further traded at a market
price. The task of the stock exchange, as an organization, is to ensure price
integrity. A common threat to price integrity arises from asymmetric infor-
mation, that is, corporate insiders being better informed about the future
prospects than outsiders. Asymmetric information should give the stock
exchange an incentive to regulate, monitor, and disclose to the public espe-
cially those transactions that involve organizational insiders. Regulation,
monitoring, and disclosure can aptly be described as activities that benefit
the public. Asymmetric information is not a problem in a closed corporation
where those who trade shares are insiders and where a tag-​along clause in the
shareholders’ agreement (see ­chapter 6) offers a safeguard to the compara-
tively less-​informed shareholders. The word public is useful also in the sense
that publicly traded corporations are subject to more public scrutiny than
closed corporations. For example, legislation places significantly more strin-
gent disclosure requirements on publicly traded firms.
For the stock exchange, as an organization, probity (see ­chapter 5) is es-
sential. It is one thing for an individual corporation to lose credibility by
failing to maintain its integrity as an organization, but it is quite another if
this happens to an organization such as a stock exchange. The stock market
is an indispensable source of risky financing to companies seeking funding
for highly specific assets and projects. Its failure can have devastating ripple
effects on industries, even entire societies. Proper functioning of the stock
exchange is not only a matter of getting the prices of the traded securities
right, but also a matter of the integrity of the organization. The Supreme
 The Expanding Organization  193

Court described the importance of probity of the stock exchange: “It requires
but little appreciation of the extent of the Exchange’s economic power and of
what happened in this country during the 1920’s and 1930’s to realize how es-
sential it is that the highest ethical standards prevail as to every aspect of the
Exchange’s activities.”3
Establishing and maintaining integrity activates both the contractual
and the institutional pillars. Stock exchanges must not only be sufficiently
regulated by the authorities (the institutional pillar), but they also have to
engage in self-​regulation (the contractual pillar) by monitoring those whose
shares are traded on the exchange. To this end, stock exchanges have explicit,
elaborate rules that listed companies must follow. Many of these rules ad-
dress the characteristics of the securities themselves, but for our purposes,
the rules regarding corporate governance are the most relevant. Some of the
rules warrant attention here because they shed light on the changes that must
occur in corporate governance at an IPO.

The Importance of Independent Oversight

When management, oversight, and risk expand in scale and scope, there
is a natural drift toward greater separation of the three. In addition, the
organization’s environment starts to impose various independence and
separation-​of-​powers requirements. For example, Nasdaq and New York
Stock Exchange (NYSE) require that the majority of board members be in-
dependent. Independence involves a number of both formal and informal
requirements. Not having an employment relationship is a good example
of a formal, “bright-​line” requirement of independence. To qualify as in-
dependent, neither the board member nor his or her family members can
have had an employment relationship with the company within the last three
years.4
However, establishing independence is not simply a matter of ticking the
proper boxes to establish formal independence; the essence of independence
resides in directors using independent judgment. The main concern is that the

3 Silver v. N.Y. Stock Exch., 373 U.S. 341 (1963).

4 Just to show how detailed the requirements are, Nasdaq Rule 5605 provides an exhaustive list of
people who are considered family members: “a person’s spouse, parents, children, siblings, mothers-​
and fathers-​in-​law, sons-​and daughters-​in-​law, brothers-​and sisters-​in-​law, and anyone (other than
domestic employees) who shares the person’s home.”
194  Governance and the Organizational Life Cycle

nonindependence of a director may interfere with the exercise of independent

judgment. Yet, independent judgment is so foundational that we promote it as
the most important tool in addressing the separation dilemma: Those exercising
oversight in the organization must genuinely embrace their fiduciary role.
The importance of implementing a sound process for selecting board
members cannot be overstated. In his pioneering analysis of boards of
directors, legal and management scholar Myles Mace (1971) made a number
of observations that should give us pause. Mace’s central conclusion was
that boards were often unduly influenced and controlled by CEOs, which
raised the concern that oversight and management might not be sufficiently
separated. For example, CEOs might be able to exert influence on the selec-
tion of board members and often also set the board’s agenda, particularly
if the CEO also chaired the board. Mace’s (1971, 205–​06) conclusion re-
garding board members’ ability to monitor management was equally discon-
certing: “A few board members ask discerning questions. Most do not.”
The issues Mace identified may not pose a hazard in closed corporations,
but they can be toxic to the credibility of an open, publicly traded corporation.
Many of the corporate scandals at the turn of the millennium—​Enron, Tyco,
WorldCom, and many others—​had the common denominator of insufficient
de facto separation of powers and inadequate oversight. Insufficient separation
of management and oversight is untenable in situations where the majority of
risk-​bearers are members of neither management nor oversight and yet, bear
the brunt of the financial consequences of bad management decisions.5
An alternative would be to ensure that the largest shareholders have rep-
resentation on the board, thus keeping oversight and risk in lockstep as the
organization expands. This was indeed the case in Tesla’s expansion where
each financing round prior to the IPO occurred with the simultaneous addi-
tion of new board members representing the incoming investors. However,
whereas keeping risk and oversight in lockstep is feasible prior to the IPO,
it is no longer an option for open corporations whose ownership becomes
unavoidably fragmented. With the exception of Elon Musk who owns about
20 percent of Tesla’s shares, the largest individual shareholders own just a few
percent stake.

5 At Tesla, this concern is alleviated because CEO Elon Musk is Tesla’s largest shareholder with
23.1 percent of all outstanding shares (as of June 30, 2021). This means that at Tesla, management and
risk are not as clearly separated as they are, say, at Apple, whose CEO Tim Cook owns only 0.02 per-
cent of all shares (as of December 28, 2020), which is more typical of large, open corporations. Due
to a lower separation of management and risk, Tesla is an organizationally exceptional open
corporation.
 The Expanding Organization  195

Why should someone with, say, a 5 percent stake in the corporation get to
appoint a board member? This issue leads to what we view as the most fun-
damental question in board composition: Does stakeholder representation
on boards enhance efficiency? We have covered this issue in ­chapter 4 in the
discussion of the dilemma of stakeholder participation. In the following, we
extend this discussion by introducing another complicating factor—​the sep-
aration hazard.

The Separation Hazard

Consider a small startup firm where five cofounders contribute the requisite
equity, three form the board of directors, and one board member becomes a
full-​time employee. The startup, as an organization, is best described as a co-
operative agreement among a small group of individuals all of whom have a
residual interest and bear residual risk. The concern for agency problems is neg-
ligible, because for all practical governance purposes, principals act as their own
agents. If the board holds the voting majority of shares, every board meeting is
de facto a shareholders’ meeting. Concerns for asymmetric information are fur-
ther alleviated by the tag-​along clause and potential economic holdup problems
are preempted by the drag-​along clause in the shareholders’ agreement.
When management, oversight, and risk are not materially separated, eve-
ryone involved in the organization is readily incentivized to do what is best
for the organization. This alignment does not mean that everyone wants
exactly the same thing; indeed, cofounders often disagree on issues. But
conflicts among known individuals with everyone committed to risk are
more likely to be manifestations of honest disagreements than opportunistic
attempts at seeking private benefits at the expense of other cofounders.
The separation of management, oversight, and risk involves a number of
fundamental changes to the organization. One immediate consequence of
separation is that principal-​agent relationships emerge throughout the or-
ganization. Consequently, those in charge of oversight at various levels of the
organization find themselves monitoring individuals who have merely an
employment relationship with the organization; the fraction of employees
who are also residual claimants rapidly declines. As the organization expands
further, the board finds itself no longer controlling the majority of votes at
the shareholders’ meeting. Finally, before a private limited liability company
196  Governance and the Organizational Life Cycle

can go public, the majority of its board members must be independent. These
phenomena are familiar in expanding organizations.
When an organization expands, it no longer exhibits many of the charac-
teristics of what we colloquially ascribe to an organization: unity of purpose,
stable and unambiguous participation, and, most importantly, a clear dis-
tinction of what is “inside” versus “outside” the organization. As the organ-
ization expands, organizational boundaries become elusive as individuals
and collectives participate in the organization’s activities in various roles and
in varying degrees. Some are merely passive investors entitled to residual
payments; others are full-​time employees who are guaranteed fixed payments.
Here, it is useful to think of the organization as the interconnection of diverse
contracting relationships the organization has with its constituencies, that
is, as a nexus of contracts. In the nexus-​of-​contracts thinking, the organiza-
tion itself is viewed merely as a “fictional entity” that consists of “a complex
process in which the conflicting objectives of individuals (some of whom
may ‘represent’ other organizations) are brought into equilibrium within a
framework of contractual relations” (Jensen and Meckling 1976, 311).
Invoking the nexus-​of-​contracts analogy can be instructive in developing
an understanding of the contractual aspects of the organization in general and
the separation dynamic in particular. The nexus-​of-​contracts analogy invites
the designer to view the organization and its constituencies as a collection of
diverse contractual relationships where distinct rights and responsibilities are
defined for the transacting parties (fig. 7.3). The heterogeneity observed in
these rights and responsibilities is a direct consequence of the separation dy-
namic. More generally, the nexus-​of-​contracts analogy invites us to think of the
organization not in terms of entities but in terms of contractual relationships.
Some of the contracts in the nexus are written, formal, ex ante contracts;
buyer-​supplier contracts, employment contracts, and loan agreements are
representative examples of formal contracts based on private ordering. Other
relationships should be considered contractual even in the absence of a
formal contract; the residual claimancy of shareholders is the most salient ex-
ample. Finally, some contracts in the nexus are more accurately described as
obligations that originate in the institutional environment; tax liabilities are a
good example. All these examples describe relationships between the organ-
ization and one of its constituencies, which can all be usefully examined in
terms of the rights and the obligations they bestow on the exchange parties.
Note that the board of directors does not appear explicitly in figure 7.3.
This deliberate omission is consistent with our premise that the relationship
 The Expanding Organization  197

Figure 7.3  The organization as a nexus of contracts

between the organization and those in charge of central oversight should not
be viewed in contractual terms. In the nexus-​of-​contracts analogy, the main
task of the board of directors is to ensure that the nexus in its entirety is effi-
cient. Accordingly, it is useful to think of the board of directors as a trustee
of the nexus itself: The board’s role is to engage in “other-​regarding behavior”
(Blair and Stout 2001, 404) to ensure the efficiency of all contracts the legal
entity has with its constituencies. Given this fiduciary role, it is straightfor-
ward to see why the board of directors should remain independent and not
become a stakeholder in the organization.6
When decision-​making powers, monitoring, and risk-​bearing are con-
centrated in a small number of actors, the nexus is comparatively simple.

6 The idea that the board of directors should be considered a fiduciary of the entire organization as
opposed to being merely a representative or an agent of shareholders is not novel. Already in 1932,
legal scholar Merrick Dodd (1932) argued that in specializing in various organization-​specific tasks,
employees may de facto invest in the organization in a way that is analogous with the investments
made by shareholders. Commitments to specificity create a residual interest that requires the
designer’s attention. Dodd’s ideas have been further expanded and elaborated by Clark (1985) and
Blair and Stout (1999, 2001), among many others.
198  Governance and the Organizational Life Cycle

Delegation is trivial as those involved in the organization are entitled to

participate in the central decisions and bear the economic consequences of
their choices. But when the organization grows and powers start to separate,
some delegation of decision rights becomes necessary. Especially when risk
becomes materially separated from management, executives will be able to
make decisions on behalf of the risk-​bearers without their explicit approval.
Without the requisite safeguards, this can give rise to organizational hazards.
Another issue that adds to the separation hazard has to do with asymmetric
information arising from the separation of management and oversight. What
are the chances that a very busy director who spends roughly one day a month
attending to the organization’s issues is able to effectively monitor the decisions
of a manager who, in stark contrast, spends all the days in the month managing
the organization? Mace’s (1971, 205–​06) observations regarding the reality
of boards are just as relevant today as they were in 1971: Board members are
very busy executives whose motivation and time to serve as directors of other
companies is highly limited. Because of this limitation, many of those tasked
with oversight lack even the basic competence of being able to ask management
discerning questions.
In sum, when the organization expands its scale and scope, its members
specialize in different roles, decision-​making powers are delegated, and di-
verse principal-​ agent relationships emerge throughout the organization.
Consequently, the organization becomes fragmented in a number of ways. In
the following section, we examine the implications of such fragmentation in the
context of ownership changes. Whereas changes in ownership may occur in all
organizations, they are particularly relevant in expanding organizations that are
more likely to experience recurring, both incremental and radical, changes in
ownership. Furthermore, although most changes in ownership are desirable
and benefit all constituencies, others involve various degrees of “hostility” that
affect the participants of the nexus in drastically different ways. A forward-​
looking designer must think of proper safeguards in case there are “hostile”
buyers in the market for organizations.

Changes in Ownership in the Market for Organizations

Just like contracting parties exchange products and services with one an-
other in markets for products, organizations can be bought and sold in the
market for organizations. Mergers, acquisitions, takeovers, and leveraged
 The Expanding Organization  199

buyouts are examples of exchange where the object of the exchange is an or-
ganization, or a part of it.
It is useful to think of the acquisition of an organization by another as the ac-
quisition not of assets but of central decision rights. This way of thinking is par-
ticularly fitting in contexts in which the target is an organization where fixed
assets are not central; a professional service firm is a representative example.
When a large law firm acquires a smaller firm, it does not acquire ownership
of the smaller firm’s lawyers but does obtain the rights to plan and coordinate
their activities. However, as we discussed in c­ hapter 1, even in the case of or-
ganizations with fixed assets, the essence of ownership resides not in the title
to the assets but in the rights to make decisions regarding their use; both deci-
sion rights and residual rights of control. Furthermore, because the market for
organizations tends to be especially relevant in the context of limited liability
companies, we typically speak not of the market for organizations, but of the
market for corporate control (Manne 1965; Jensen and Ruback 1983).
Expansion is often a sign of success, or at least, potential for success. Firms
with a promising future are more likely to attract potential buyers. Moreover,
because expansion is also associated with increasing organizational fragmenta-
tion and hazards, the potential buyer may also be able to take over the organ-
ization without the approval of some of its stakeholders. The designer’s task is
to ensure that the decision to be acquired is made by the explicit approval of the
organization’s key fiduciary, its board of directors. The designer’s task is there-
fore to ensure that in the event of an acquisition, the fiduciary has the requisite
bargaining power to negotiate with the buyer to act in the best interest of the
organization.

Wanted vs. Unwanted Acquisitions

To be sure, the market for corporate control can serve a useful purpose
(Manne 1965). If organizations are traded between voluntary participants on
an open market, the ownership and control of organizations are more likely
to end up in the hands of those who value them the most. For example, the
leadership of one firm may look at a smaller competitor and conclude that its
value-​creating activities can be improved or that economies of scope can be
realized by the merger of the two companies.
In discussing changes in ownership, we often make the distinction between
“friendly” and “hostile” takeovers. A takeover is “friendly” when the acquirer
200  Governance and the Organizational Life Cycle

approaches the target’s board of directors in good faith; in a “hostile” takeover,

the acquirer bypasses the target’s board of directors and makes offers to purchase
shares directly to the target’s shareholders. If the “hostile” buyer is successful in
purchasing a majority of the shares, it can proceed to replace the target’s board
of directors (and subsequently management) without the target’s approval.
We submit that directing attention only to the buyer’s motivations and actions
results in an understanding of acquisitions that is partial at best and misguided
at worst. To understand an acquisition in its entirety, we must also incorporate
the motivations and the actions of those who negotiate on behalf of the target.
Just like the representatives of the acquirer, the representatives of the target may
or may not act in good faith. Instead of making its decisions by incorporating
the interests of the entire nexus of contracts, the target’s board may engage in
self-​serving and entrenched behaviors. To incorporate both the target and the
acquirer side to the analysis, we propose that the friendly/​hostile distinction be
replaced by the wanted/​unwanted distinction (table 7.1). A takeover is wanted
when both the acquirer and the target negotiate voluntarily and in good faith; if
either party is an unwilling participant or either side acts in bad faith, the take-
over is unwanted.
We present the wanted/​unwanted distinction with an important caveat. The
designer should not think of takeovers as categorically being either wanted or
unwanted. No matter how unwanted an acquisition is, at least 50 percent of the
target’s voting shareholders must deem it wanted, otherwise the acquisition will
not take place. Similarly, no wanted acquisition enjoys the unconditional ap-
proval of all the target’s and the acquirer’s constituencies; Tesla’s acquisition of
SolarCity (see ­chapter 2) is a perfect example. The same caveat applies to the
friendly/​hostile distinction. In a hostile takeover, the acquirer exhibits hostility
primarily toward the target’s board of directors, not its shareholders, employees,
or any other constituency. Furthermore, disregard might be descriptively a better
word than hostility to describe the buyer’s behavior toward the target’s board.7

7 We do not know where and how the word hostile entered into the antitakeover lexicon, al-
though Orts (1992, 24) offered one plausible account in the US context. According to Orts, political
constituents in the so-​called rustbelt states (e.g., Pennsylvania, Ohio, and Illinois) believed that cor-
porate takeovers posed a threat to the jobs in their states. These perceived threats fueled public meas-
ures to address takeovers, most notably antitakeover legislation in the form of corporate constituency
statutes. Perhaps not surprisingly, Pennsylvania—​not, say, Florida or Hawaii—​was the first US state
to adopt a constituency statute. However, protection against takeovers is for all practical purposes a
matter of private ordering because constituency statutes permit but do not obligate directors to con-
sider broader stakeholder interests. Incorporating a constituency statute into corporate governance
is ultimately a matter of board discretion (Bebchuk and Tallarita 2020).
 The Expanding Organization  201

Table 7.1  A Comparison of Wanted and Unwanted Takeovers

Issue Wanted takeover Unwanted takeover

Mutual interest Both the acquirer and the The target is an unwilling
target act voluntarily participant
Whom does the The target’s board of directors The target’s shareholders
acquirer approach?
Bargaining power The target’s board of directors The target’s board of directors
has sufficient bargaining power lacks bargaining power or is
bypassed altogether
Whom does the The target’s entire nexus of The target board’s and
target’s board contracts, those with a residual management’s entrenched
represent? interest in particular interests
Reason for The target’s board concludes The acquirer secures a voting
accepting an offer the offer is beneficial to those majority of shares without the
with a residual interest and target board’s approval
the board’s justification is
supported by a shareholder
majority
Reason for rejecting The target’s board concludes Either the target’s board is
an offer the offer is not in the best entrenched and rejects the
interest of those with a residual offer out of self-​interest, or
interest the acquirer fails to acquire a
voting majority of shares

The wanted/​ unwanted distinction elaborates the friendly/​ hostile dis-

tinction in two important ways. Specifically, whether a friendly or a hostile
takeover attempt should be considered wanted or unwanted depends on
the motivations on the target’s side. A hostile takeover attempt may well be
wanted if the target’s board has entrenched itself and rejects all offers, in-
cluding ones that would be beneficial to its residual claimants. The target’s
board might resist a takeover attempt simply because board membership
may be a significant source of income to the board members. In a takeover,
the target’s board is often dismissed in its entirety, which means loss of in-
come to its members. Similarly, a friendly takeover attempt may exhibit char-
acteristics of an unwanted takeover if the target’s board lacks bargaining
power and, consequently, has trouble negotiating on behalf of the target’s re-
sidual claimants.
202  Governance and the Organizational Life Cycle

Protection against Unwanted Takeover Attempts

We counsel the forward-​looking designer to address the hazard of unwanted

acquisitions already at the founding of the organization. To this end, the de-
signer has two related objectives. One is ensuring that those exercising pri-
mary oversight in the organization authentically internalize their fiduciary
role. The importance of independent judgment becomes pronounced in
situations in which the organization becomes the target of unwanted acqui-
sition attempts. Safeguards aimed at forcing all potential acquirers to nego-
tiate with the target’s board of directors provide protection against unwanted
acquisitions.
The second objective is to ensure that when oversight and risk are separated,
the board of directors maintains credibility in the eyes of the organization’s
residual claimants. In the case of takeovers, the importance of those residual
claimants with decision and control rights—​the shareholders—​is para-
mount. Unwanted takeovers become impossible in situations in which a
simple shareholder majority refuses to consider purchase offers that have not
been approved by the target’s board of directors.

Formulating the Main Problem

Just like all governance decisions, protection against unwanted takeovers

requires the formulation of a main problem. Let us start at the general
premise that the designer should safeguard those contractual relationships
where vulnerability is present and suggest that this premise applies to
changes in ownership as well. The key question therefore becomes, “Which
relationships in the nexus become vulnerable in the event of an unwanted
takeover attempt?”
Formal contracts are more likely to survive acquisitions. When one
firm purchases another, it commonly purchases all the contracts in the
nexus: buyer-​supplier contracts, employment contracts, loan agreements,
and so on. There is no immediate reason to assume that these formal con-
tractual relationships will become jeopardized when ownership changes. In
fact, the effect may be just the opposite, particularly in situations in which
a comparatively smaller growing firm is purchased by a large incumbent.
The acquisition may lead to better job security for employees, larger order
volumes for suppliers, and lower probabilities of defaults on loans. Unless
 The Expanding Organization  203

those with a formal contractual relationship also have a residual interest in

the target, no additional safeguards are needed.
The situation is more complex with constituencies that have a residual
interest in the target. In the event of an acquisition, this residual interest
is either eliminated altogether or transformed into a residual interest in
the acquiring organization. For example, in an all-​stock acquisition, the
target’s shareholders exchange their shares in the target firm for shares in
the acquiring firm. For this exchange to be in the best interest of the target’s
shareholders, the target’s board must have sufficient bargaining power to
negotiate on behalf of the target’s shareholders. Not surprisingly, the main
problem in protection against takeovers has conventionally been framed in
shareholder terms: The organization should implement those antitakeover
provisions that enhance the shareholders’ bargaining power in the event of a
takeover attempt.

Addressing the Main Problem

Suppose the designer decides that protection against unwanted takeovers

is desirable. What are the design tools available to implement protection in
the case of unwanted takeover attempts? In the following, we discuss two
examples used to safeguard against hostile takeovers: staggered boards and
the poison pill.8

Staggered Boards
The hostile acquirer’s central objective is to replace the target’s board of
directors as rapidly as possible. The longer it takes to replace the board after
an acquisition, the less economically attractive the target becomes to the hos-
tile buyer. Consequently, an intuitive delay tactic for the target is to make the
replacement of the board more difficult. This impediment can be achieved
by adopting a staggered board. In a staggered board, only a certain portion

8 There are numerous other antitakeover measures, such as dual class shares, supermajority
merger approval provisions, fair-​price amendments, reduction in cumulative voting provisions, and
anti-​greenmail provisions. Since covering all these in detail is outside the scope of this book, we rec-
ommend the following scholarly work on the topic (in chronological order): Grossman and Hart
(1980), DeAngelo and Rice (1983), Jensen (1986), Hirshleifer and Sheridan (1992), Sundaramurthy,
Mahoney, and Mahoney (1997), Bebchuk, Coates, and Subramanian (2002), Bebchuk (2003), Liu
and Mulherin (2019). To complement these academically oriented texts, Larcker and Tayan (2015)
offer an excellent text to practitioners.
204  Governance and the Organizational Life Cycle

of board seats are opened for reelection at the annual shareholders’ meeting.
Staggered boards are a common and effective form of protection adopted in
the pre-​IPO stage. In their empirical study of hostile bids, Bebchuk, Coates,
and Subramanian (2002) found that not a single hostile bid was successful
against well-​structured staggered boards in the five-​year period examined.9
The way staggering provides antitakeover protection is straightforward.
However, whether any governance measure is ultimately efficient requires
a broader consideration of the net impact. An undesirable consequence of
staggering is that it effectively protects board members from being rapidly
replaced. This protection is problematic, because any measure that makes a
person, or a group of persons, more difficult to replace facilitates entrench-
ment. A staggered board may have the unintended consequence that the
board may be reluctant to reject all acquisition offers, including ones that
might be beneficial to shareholders.
The dilemma that staggering presents effectively highlights the impor-
tance of having an independent board that fully embraces its fiduciary
duty. Anything that leads to the board developing a stakeholder rela-
tionship with the organization presents a hazard because the board may
start acting in ways that is beneficial to the board as a stakeholder, not
to the entire nexus of contracts. In their study of antitakeover provisions
and shareholder wealth, Sundaramurthy, Mahoney, and Mahoney
(1997, 239) found that “the market reacts less negatively to antitakeover
provisions adopted by boards with a chairperson who is not the CEO
than to antitakeover provisions adopted by boards chaired by the CEO.”
Expressed in Efficiency Lens terms, the less management is separated
from oversight, the more skeptically those bearing risk will interpret the
board’s actions to engage in protection. We also know from research on
board litigation that boards chaired by the CEO are sued more often by
shareholders than boards that are not chaired by the CEO (Kesner and
Johnson 1990). Entrenchment becomes particularly hazardous when the
prerogative to adopt antitakeover measures is delegated to the board of
directors, which is the case in the poison pill defense.

9 The principle of staggering is not limited to corporations; it is employed in many contexts in

which rapid replacement of a decision-​making body is undesirable. For example, only one-​third of
the seats in the US Senate are up for election in a given election year; senators have a six-​year term,
with one-​third of the seats up for election every two years.
 The Expanding Organization  205

The Poison Pill

A common antitakeover provision available at the discretion of the target’s
board of directors is that of making the target’s shares less attractive in the
eyes of prospective hostile buyers. The mechanism is known as the poison
pill. The name is misleading because the poison pill acts more like a dilutor
than a toxin; the poison pill dilutes the acquirer’s equity in the event of a hos-
tile takeover attempt. The target’s board may activate the poison pill when-
ever an acquirer purchases a threshold stake of the target’s stock (usually
between 10 and 20 percent of outstanding shares) without the target board’s
approval (Sundaramurthy et al. 1997).
There are two variants of the poison pill. A flip-​in poison pill gives the
shareholders of the target firm the right to purchase additional shares in the
target firm at a deep discount after a hostile acquirer purchases a threshold
stake. This effectively floods the market with the target’s shares, thus diluting
the acquirer’s stake in the target. A flip-​over poison pill gives the shareholders
of the target firm the right to purchase shares in the acquiring firm at a dis-
count. Both variants have the same effect of diluting the acquirer’s post-​
acquisition equity either in the acquirer (flip-​over) or the target (flip-​in) firm.
The poison pill constitutes a strong form of protection because it effec-
tively blocks hostile bids completely. We are not aware of any instances of the
acquirer “swallowing the pill”—​purchasing a threshold stake in the target,
triggering the pill, absorbing the dilution, and then continuing to buy shares
until a voting majority has been attained.

A Broader View of Residual Interest

As both staggered boards and the poison pill illustrate, residual interest in
antitakeover conversations is usually considered only from the point of view
of shareholders: An antitakeover provision should be adopted if it benefits
the shareholders.
Defining residual interest in shareholder terms may be contrived in
contexts where other stakeholders have a legitimate residual interest due to
their commitment to organization-​specific skills (in the case of employees)
or relation-​specific investments (in the case of suppliers and customers). As
we suggested in ­chapter 4, commitments to specificity should be considered
analogous with investments in equity, and therefore, deserve to be supported
by credible commitments from the organization. We thus propose that the
206  Governance and the Organizational Life Cycle

focus in antitakeover conversations and designer’s decisions should not be

limited to shareholder wealth but should incorporate residual interests more
broadly.
To address the broader implications of takeover defense from the point of
view of those committing to specificity, let us turn to the ex ante challenge
the designer faces: How does the designer ensure that the organization’s
stakeholders are willing to put something at stake in the organization by
committing to specificity? How will employees, suppliers, and customers
who are asked to commit to specificity react if they find that the organization
has not adopted safeguards against unwanted takeovers?
Consider the case of the employees who are asked to commit to specificity
by learning and further developing organization-​specific skills, for example,
by engaging in firm-​specific R&D and innovation activities. These employees
may understandably view all takeovers as potentially hazardous to their
careers. A potential consequence of insufficient antitakeover provisions is
that it lowers the employees’ willingness to commit to specificity. In the case
of a growing high-​technology firm, this may further lead to an underinvest-
ment in innovation, which can have devastating consequences if the firm’s
competitive strategy is innovation-​based. Decisions regarding the extent to
which one is willing to commit to specificity are understandably affected by
uncertainty regarding postacquisition employment and compensation (Dey
and White 2021).
If the ability of the organization to create value hinges on sufficient levels
of commitment to human capital specificity, the designer can consider
antitakeover provisions as a way of signaling credible commitments to-
ward high-​specificity employees. In this signaling function, antitakeover
provisions could lead to important economic bonding effects with high-​
specificity employees and, consequently, secure adequate levels of invest-
ment in human capital specificity by employees.
More generally, mergers and acquisitions tend to be more disruptive,
and a greater cause for concern, to those constituencies with long-​term
relationships with the organization. To the extent these relationships are
crucial for organizational viability (as they often are), the designer should
think of antitakeover provisions as credibility instruments toward all those
stakeholders with a residual interest. In competitive markets and contexts in
which switching costs are low, the concern is alleviated (Cremers, Nair, and
Peyer 2008).
 The Expanding Organization  207

The nexus-​ of-​

contracts analogy is useful in exploring the broader
ramifications of takeover hazards. Because different constituencies have dif-
ferent kinds of contractual relationships with the organization, they tend to
be affected differently by changes in ownership. Management scholars Andrei
Shleifer and Lawrence Summers (1988) pointed out that making choices
regarding antitakeover provisions by focusing exclusively on shareholder
wealth does not necessarily result in a comparatively efficient solution for the
organization. Furthermore, Shleifer and Summers’s suggestion that share-
holder gains may come at the cost of breaching some of the organization’s
long-​term contracts with its constituencies should give the designer pause;
the trade-​off gives rise to both an ex post and an ex ante problem. If the
constituencies whose contracts are breached have committed to specificity,
takeovers may lead to serious ex post disruptions; the organization’s failure to
signal credible commitments to long-​term relationships may in turn lead to
the ex ante problem of underinvestment in specificity.
The more recent literature on takeover defenses has highlighted the im-
portance of their bonding effect. Finance scholars William Johnson, Jonathan
Karpoff, and Sangho Yi (2015, 329) empirically corroborate the hypoth-
esis that “takeover defenses can help to bond the IPO firms explicit and
implicit commitments to its stakeholders, including customers, suppliers,
and strategic partners.” The researchers specifically found that having large
customers, dependent suppliers, and strategic partners was not only asso-
ciated with a more extensive deployment of takeover defenses but also with
longevity of post-​IPO relationships.

Summary: Expanding Organizations,

Changing Governance

Most theories and models of governance are, if only implicitly, aimed at large,
established organizations (Pollman 2019). The purpose of this chapter on
expanding organizations and the previous chapter on incipient organiza-
tions has been to shed light on the idiosyncratic governance aspects of young
and expanding organizations.
Research literature on organizational expansion tends to focus on the
managerial (as opposed to governance) challenges that growth poses for the
organization. For example, what did it take for Tesla to go from an annual
production of 50,000 automobiles in 2015 to 500,000 in 2020? Even though
208  Governance and the Organizational Life Cycle

questions of scaling are plainly essential in all organizations, we posit that it

is ultimately not growth but the associated separation of management, over-
sight, and risk that requires the designer’s attention. This is because the sep-
aration dynamic presents the designer with a more profound governance
challenge than the growth dynamic. To be sure, international expansion,
product diversification, and scaling of organizations are far from straightfor-
ward. At the same time, the managerial challenges of organizational growth
are broadly and deeply covered in the published literature, and we have little
to add to these conversations. The challenges that the separation dynamic
presents are in our view both more challenging and less discussed.
We present the separation dynamic as an inevitable consequence of the
growth dynamic. As the organization expands, the number of organizational
subunits and vertical levels increases. This increase gives rise to multiple
principal-​agent relationships within the organization, effectively separating
management from oversight throughout the organization, not merely at the
top. Just like the board of directors exercises oversight over the top manage-
ment team, so does the top management team over division management,
division management over functional management, functional management
over functional subunits, and so on.
Viewing the organization as a nexus of contracts brings the governance
challenges associated with the separation dynamic effectively into focus.
In this chapter we have chosen corporate takeovers as an illustration of the
challenges associated with the separation dynamic. Corporate takeovers
constitute a salient example of situations in which different stakeholders
are affected in different ways, and when focusing on the interests of just
one of them runs the risk of diverting attention from efficient organizing.
Yet, we propose that efficient organizing is ultimately in the best interest of
both the target and the acquirer. In general, because shareholder returns are
postappropriation measures of performance, they tend to be poor proxies
for organizational efficiency and value creation. As Shleifer and Summers
(1988, 37) importantly pointed out, sometimes takeovers can create value
for the shareholders but simultaneously destroy value for the organization
due to their disruptions to efficient contracting with other stakeholders. The
disruptions pose a hazard in situations in which commitments to specificity
are crucial for the organization’s viability. Underinvestment in specificity
often means underinvestment in R&D and innovation.
Consistent with the general approach taken in this book, we empha-
size the context dependence of the challenges associated with expanding
 The Expanding Organization  209

organizations. As always, we counsel the designer to start at defining the main

problem. If the main problem in the context of antitakeover provisions is de-
fined in terms of all residual claimants (not just shareholders), the designer
must conduct an explicit analysis to determine which constituencies have a
residual interest. If employee commitments to specificity are negligible and
relationships with suppliers and customers largely transactional arm’s-​length
relationships, then shareholders can be viewed as the sole constituency with
a residual interest. Consequently, decisions regarding takeover defenses can
be structured around the effect on shareholder wealth. However, to the ex-
tent that constituencies other than shareholders have a legitimate residual
interest, the designer should incorporate these interests into the analysis. If
it is indeed the case that privileging shareholders means breaching implicit
or explicit contracts with those who have committed to specificity, the inef-
ficiency implications of takeovers may offset shareholder gains. As always,
the designer must analyze the ramifications of all governance alternatives in
their entirety.
The bonding effects of takeover defenses are more difficult to quantify,
which is a likely reason academics have preferred the more easily measured
outcome of shareholder wealth. Here, we might observe that we academics
live in a world where it is both possible and permissible to simplify problems
by abstracting out that which is not quantifiable, seek an exact answer, and
then finish by noting that real-​life organizations are more complex than what
our models assume.10
In stark contrast with academics, designers live in a world where not
being able to address the problem in its entirety has real consequences and
in which offering caveats and excuses of intractability or immeasurability
is ineffective. Readily measurable or not, breach of contract with those who
have committed to specificity can have serious ex ante and ex post efficiency
consequences in the organization. We are, however, encouraged by the more
recent research literature that addresses outcomes more broadly than just

10 In fact, the use of words such as optimization or maximization is a telltale sign that the academic
is drawing conclusions based on a simplified model. Optimization and maximization are infeasible
in just about any real-​life decision situation. When was the last time a discussion of “maximizing
shareholder wealth” led to any practically relevant insight? Also, it is important to debunk the notion
that the law requires maximization of shareholder wealth; we are not aware of a single instance of
the word maximization (or any of its variants) appearing in corporate law. We further concur with
Jensen (2001, 11) who noted that whenever academics use the word maximization, they are actually
referring to the more general (and realistic) objective of value seeking: “It is not necessary that we be
able to maximize, only that we can tell when we are getting better—​that is moving in the right direc-
tion.” Jensen’s position is consistent with the pursuit of comparative efficiency.
210  Governance and the Organizational Life Cycle

in terms of shareholder wealth (see the excellent and thorough review by

Mulherin, Netter, and Poulsen 2017 for more details).
In situations where residual interest is not limited to shareholders, the
board’s role as the organization’s fiduciary becomes pronounced. In an or-
ganization where the board acts as a representative of shareholders or has it-
self been made a residual claimant through equity-​based compensation, the
organization may adopt antitakeover provisions that are inefficient.
 8
The Institutionalized Organization

In the early 2000s, one of the authors consulted a large multinational corpo-
ration whose division managers had long been critical of the bloated corpo-
rate headquarters and the annual corporate management fees imposed on
their divisions. The situation finally came to a head when one division man-
ager analyzed the corporate fees in detail and found, among other things,
that his division was effectively paying the corporation $2/​page for the cen-
trally organized photocopying services. The competitive rate would have
been somewhere around $0.05/​page. How can paying a 4,000 percent pre-
mium for a service possibly go unnoticed in an organization?
There are several plausible explanations for the persistence of the mani-
festly inefficient organization of photocopying services. One is complacency.
Just like humans are creatures of habit and become set in their ways, organ-
izations develop established ways of structuring and managing their con-
tractual relationships. The multinational in question had always contracted
with the providers of photocopying services centrally, and the cost had been
allocated to the divisions. At the time the centralized service was introduced,
the chances are it served an efficiency purpose. Whatever led to the exorbi-
tantly high cost must have occurred incrementally over time through a series
of progressive distortions (Williamson 1975, 118). Furthermore, if in partic-
ular the cost at the outset was acceptable, the progressive distortions never
garnered further attention from the designer, even when there was a reason
to believe a better option might be available. As sociologist Everett Hughes
(1939, 283) put it, a “once technically useful means of achieving some known
end persists as an accepted and even sacred practice after better technical
devices have been invented.”
A second plausible explanation is that the high cost of photocopying
services was known but nobody had the incentive to do something about
it. Indeed, one of the primary drawbacks of organizing activities internally
is that incentive intensity is comparatively lower than in interorganizational
transactions. We venture to guess that division managers were not

Efficient Organization. Mikko Ketokivi and Joseph T. Mahoney, Oxford University Press. © Oxford University Press 2023.
DOI: 10.1093/​oso/​9780197610282.003.0008
212  Governance and the Organizational Life Cycle

incentivized to improve the efficiency of photocopying services within their

divisions, they had more important issues on their agenda.
A third possibility is that even if a division manager did identify the inef-
ficiency problem, conducted a comparative efficiency analysis, and voiced
the inefficiency concern to corporate management, suggestions for im-
provement might simply have been ignored. Organizing shared services is
often one of the central functions of the corporate headquarters, and divi-
sion heads calling into question the efficiency of centralized services im-
plicitly undermines the corporation’s value-​adding role. Consequently,
even if improving efficiency of shared services did enter the division
managers’ agenda, it might not necessarily have entered that of corporate
management.1
A fourth and final possibility is that the problem was one of collective ac-
tion. In order to challenge the centralized provision of photocopying serv-
ices, it would not be sufficient for just one division manager to challenge
the status quo. Instead, there would have to be a division manager who is
sufficiently motivated to rally support from all divisions. The idea of the di-
vision heads in a large multinational joining forces to challenge corporate
headquarters on the issue of overpriced photocopying services sounds al-
most comical.
No matter what the root cause of the problem was, all plausible accounts
are variations on the familiar theme of “organizations assuming a life of their
own.” Due to cognitive, economic, organizational, and political factors, gov-
ernance structures tend to persist over time even if they no longer serve an
efficiency purpose. Even when a more efficient alternative emerges, it may
not be implemented due to various inertial forces. The designer must under-
stand that in large, established organizations, intraorganizational transacting
in particular may be subject to inefficiencies that are difficult to overcome.
This unearths the dark side of institutions and institutionalization, the main
topic in this chapter.

1 We find it plausible that the excessively high cost of $2/​page was in large part caused by the
photocopying services fee including cost categories that had little to do with the services provided.
Corporations are known for sometimes engaging in various “creative” accounting practices when
they allocate costs to divisions. In the specific case of the multinational, the corporate headquarters
had about four times as many employees as corporations of similar size with a similar corporate par-
enting role would have (see Goold and Campbell 2002a, ­chapter 6). Since the headquarters did not
generate any revenue, all its costs had to be allocated to the divisions. The chances are the vast ma-
jority of photocopying costs was simply corporate overhead.
 The Institutionalized Organization  213

Institutions and Institutionalization

The word institution is like the word strategy or culture in that it can mean
different things in different contexts. At the same time, the word institution
is unlike the word strategy or culture in the sense that it can invoke both posi-
tive and negative connotations; that something has been institutionalized can
be both a source of comfort and a cause for concern.
The positive connotation of institutions stems from the order and the
stability they confer. This positive view is succinctly expressed by the def-
inition of institutionalization as “the emergence of orderly, stable, socially
integrating patterns out of unstable, loosely organized, or narrowly technical
activities” (Broom and Selznick 1955, 238). In the spirit of this definition, we
might offer the New York Stock Exchange and the Securities and Exchange
Commission as two central economic institutions both of which have an in-
dispensable role in the functioning of the US economic system. Their role
becomes evident during times of conflict when adversaries seek to under-
mine one another’s central institutions.
We can, however, think of institutions in the more abstract sense as well.
The limited liability company (an organizational form), an independent
board of directors (a governance principle), and the employment contract (a
formal agreement) are three central institutions that work toward supporting
economic business activities. The taxi ride example in ­chapter 1 is also re-
plete with “social microinstitutions” of all kinds. For example, how does the
taxi driver know that a person waving one’s hand at a passing taxi indicates
that the person needs transportation? How does the person in need of trans-
portation in Madrid know that the green light on the taxi’s roof sign means
that the taxi is neither busy nor off duty? We know because there are institu-
tionalized ways of nonverbal communication in the specific context. In the
most general sense, we can define institution as any commonly agreed-​upon
way of organizing an activity.
In all these examples, the word institution acquires an exclusively positive
meaning; institutions contribute to the efficient functioning of the economic
system. Efficiency arises from the fact that institutions make behaviors pre-
dictable, they facilitate communication, and so on. In fact, many institutions
have been deliberately designed with such efficiency in mind. Just imagine
how hopelessly inefficient it would be to buy and sell shares of the General
Motors Company if the New York Stock Exchange and the Securities and
Exchange Commission did not organize and exercise key oversight over the
214  Governance and the Organizational Life Cycle

trading of GM’s nearly one and a half billion shares. In this sense, institutions
breed efficiency.
At the same time, institutions may also be sources of inefficiency.
Sociologist and legal scholar Philip Selznick (1957, 16–​17) noted that the in-
stitutionalization of an activity, structure, or routine involves an “infusion
of value beyond the technical requirements of the task at hand.” We suggest
that there are two distinct ways of interpreting the notion of “infusion of
value.”2 The positive interpretation suggests that as an activity or an organi-
zation institutionalizes, it is no longer an instrument toward some technical
end but, instead, becomes considered intrinsically valuable in its own right.
Importantly, the positive interpretation suggests that institutionalization is
largely beneficial to the organization; Selznick (1996, 271) noted that insti-
tutionalization is the very process by which organizations develop their dis-
tinct characters and distinctive competences.
Both acknowledging and embracing the positive interpretation, we also
wish to give voice to an alternative, less flattering interpretation of institu-
tionalization. The negative interpretation holds that over time, institutions
may become a source of inefficiency. This is because “infusion of value” may
also mean that governance choices are either no longer called into ques-
tion or simply have become embedded in the organization in ways that
make changes prohibitively costly. If in particular the objective is to main-
tain an organization that adapts to changes in its environment, it is easy to
see how embeddedness hinders adaptation and constitutes a threat to effi-
ciency. For example, if the organization has adopted a functional structure
that has become infused with value over time, there is a probability it will not
be replaced by a divisional structure even if the latter were deemed compar-
atively efficient.
The institutionalization of governance structures is in many ways anti-
thetic to the prescription of remaining relentlessly comparative. In c­ hapter 6,
we sought to establish that the relentlessly comparative designer should take
nothing for granted; the governance needs of a small startup organization
are very different from those of an expanding or a mature organization. In
their insightful article on organization redesign, management scholars
Christopher Worley and Edward Lawler (2006, 19 [emphasis added]) noted

2 It did not occur to us until the writing of this chapter that we, the two authors of this book, read
this passage from Selznick in more or less opposite ways. One of us interprets the “infusion of value”
as positive and desirable, and the other, as negative and undesirable. Both interpretations warrant the
designer’s attention.
 The Institutionalized Organization  215

that even though designers often speak of the importance of adaptation, “the
truth is that most businesses have organized themselves in ways that inher-
ently discourage change.” In our view, the discouragement stems from the
persistence of governance structures.
The challenge that institutionalization presents to organizational effi-
ciency is succinctly presented and summarized by sociologists John Meyer
and Brian Rowan (1977). In their classic analysis of organizational structures,
Meyer and Rowan suggested that formal organizational structures tend not
to serve the purpose of efficiency; instead, they are meant to signal legitimacy
to the organization’s key constituencies. When Meyer and Rowan (1977,
340) looked at the structure of an organization, they saw “myth and cere-
mony,” not efficiency. Although myths and ceremonies may serve important
purposes, efficiency tends not to be one of these purposes.3 Two examples
illustrate the point.

The Manager as an Institution

When the first author of this book chaired the board of directors of an indus-
trial startup, he quickly learned that the CEO is an institution not only in es-
tablished firms but also in startups. As he reached out to a number of external
investors to raise equity, one of the first questions many prospective investors
asked was, “Who is your CEO?” Note that the question was not whether the
startup had a CEO or not—​that the firm had a CEO was something prospec-
tive investors took for granted.
During the first two rounds of financing, the company did not have a CEO,
because the board saw no immediate efficiency reasons for appointing one,
and the law did not require it. But the board soon realized that without a
CEO, attempts at raising equity would be impossible. Therefore, although the
board did not see any efficiency reasons for appointing a CEO, it had no op-
tion but to appoint one in order to appear legitimate in the eyes of external
investors. In sum, it might not have been efficient for the startup to have a
CEO, but it certainly seemed appropriate.

3 The position that the objective of formal organizational structures is not efficiency (as defined
in this book) but legitimacy is both well established and compelling. There is a rich research tra-
dition on the topic in the sociology literature, starting with the foundational works of the German
sociologist Max Weber. In the organization research literature, the edited volumes by Powell and
DiMaggio (1991) and Greenwood et al. (2008) provide excellent summaries of this research and the
key arguments. Selznick’s (1957) classic book Leadership in Administration also merits reading.
216  Governance and the Organizational Life Cycle

More generally, we tend to take it for granted that firms have managers.
Indeed, that the startup must have a CEO is an instance of the more general
notion of the institutionalization of the manager as an organizational role.
At the same time, as we suggested in c­ hapter 2, whether managerial work is
performed by individuals whose formal position is that of a manager is not
something to be taken for granted but, instead, something to be analyzed. We
have on multiple occasions observed how rigorous discussions on manage-
ment are hamstrung by the fact that the formal organizational position of the
manager has been institutionalized. Discussions on the future of manage-
ment (and managers) seldom lead to actionable insights. In the provocatively
titled article “First, Let’s Fire All the Managers,” strategy scholar Gary Hamel
(2011, 51) aptly noted that “we are all prisoners of the familiar.”
The idea that a startup firm must have a CEO ultimately suggests that the
de facto foundation of an organization is both technical and ideological.
A case in point, the founding of a startup exhibits many features of a ritual
where many ideas are taken for granted. For example, one common assump-
tion embedded in many shareholders’ agreements is that arbitration is pref-
erable to litigation because the former is cheaper and less time-​consuming.
Consequently, cofounders ritualistically introduce to the shareholders’
agreement a clause of binding arbitration as the preferred method of dis-
pute resolution. At the same time, experience has shown that the taken-​for-​
granted assumptions of less costly and less time-​consuming arbitration are
contestable (Stipanowich 2010).
To sociologists, the institutionalization of both organizations and our
thinking about them is a familiar phenomenon. Meyer and Rowan (1977,
344) elaborate:

Ideologies define the functions appropriate to a business—​such as sales,

production, advertising, or accounting; to a university—​such as instruc-
tion and research in history, engineering, and literature; and to a hospital—​
such as surgery, internal medicine, and obstetrics. Such classifications of
organizational functions, and the specifications for conducting each func-
tion, are prefabricated formulae available for use by any given organization.

Fully acknowledging Meyer and Rowan’s position as understandable and

well argued, we alert the designer to the potential inefficiencies embedded in
the “prefabricated formulae,” particularly when the assumptions embedded
 The Institutionalized Organization  217

in them may be either incorrect or become obsolete over time. Taken-​for-​

grantedness may ultimately transform into a liability.

R&D Spending as an Institution

The second example of institutionalization concerns R&D spending and

innovation. For over two decades, politicians and policymakers in the
European Union have lamented the low levels of innovation in European
firms. The aim of the so-​called Lisbon Strategy, launched in 2000, was to make
Europe “the most competitive and dynamic knowledge-​based economy in
the world.”4 The strategy called, among other things, for more investment in
knowledge and innovation. A key metric for gauging investment in innova-
tion is Gross Domestic Expenditure on R&D (GERD), defined by OECD as
“the total expenditure (current and capital) on R&D carried out by all resi-
dent companies, research institutes, university and government laboratories,
etc., in a country.”5 The objective of the Lisbon Strategy was to raise R&D
spending in the European Union to 4 percent of the gross domestic product
(GDP). The implicit one-​size-​fits-​all prescription was that all EU member
countries, regardless of their industrial structures and histories, should raise
their R&D spending to 4 percent of their national GDP.6
Even though the ways in which organizations can innovate are numerous,
equating innovation with R&D is a long-​standing institution both among
policymakers and strategy researchers (Sawhney, Wolcott, and Arroniz
2006). The reason R&D spending is appealing is that it provides a general,
unambiguous accounting metric for an elusive, multidimensional concept.
Furthermore, because R&D spending is something companies must disclose
in their annual reports, the metric is also publicly available information,
which makes the it easy to monitor not only at the country level but also at
the level of industries and even individual organizations.

4 Lisbon European Council, Presidency Conclusions, March 23–​24, 2000 (www.europ​arl.eur​opa.

eu/​summ​its/​lis1​_​en.htm), accessed August 11, 2021.
5 OECD Data (data.oecd.org/​rd/​gross-​domestic-​spending-​on-​r-​d.htm), accessed August 11, 2021.
6 By all accounts, the Lisbon Strategy has failed. While GERD in the EU has increased from
1.7 percent in 2000 to 2.1 percent in 2020, it is not only nowhere near the 4.0 percent target, but it
is also considerably below the OECD average of 2.5 percent (source: OECD Data, data.oecd.org/​
rd/​gross-​domestic-​spending-​on-​r-​d.htm, accessed August 11, 2021). One contributing factor to the
problem may be that most executives in Europe we have talked to have never heard of the Lisbon
Strategy. If the objective is to encourage R&D investment in firms, a policy that is invisible to corpo-
rate managers has little hope of being effective.
218  Governance and the Organizational Life Cycle

Not surprisingly, R&D spending as a proxy for innovation receives less

support from executives and practitioners. In an interview with Fortune
magazine in November 1998, Steve Jobs gave what is perhaps the best-​
known, and most scathing, evaluation: “Innovation has nothing to do with
how many dollars you have. When Apple came up with the Mac, IBM was
spending at least 100 times more on R&D. It’s not about money. It’s about the
people you have, how you’re led, and how much you get it” (Kirkpatrick and
Maroney, 1998, 90).
It is hard to disagree with Mr. Jobs. At the same time, critique from even
arguably the most prominent corporate executive of all time has done next
to nothing to dissuade the use of R&D intensity—​annual R&D spending di-
vided by annual revenue—​as the most common metric of innovativeness in
economic research on innovation. In some contexts, the very definition of
a high-​technology company is based on R&D intensity: To qualify as a high-​
technology company, one must have an R&D intensity of at least 10 percent.
The R&D intensity example unveils a central characteristic of institutions.
As taken-​for-​granted ways of thinking, they are incredibly resilient in the
face of even the most rational and justified critique. We encourage the de-
signer to challenge such resilience whenever it threatens efficient organiza-
tion, no matter how appropriate the principle or action that is protected may
be regarded.

Remediableness and Adjustment Costs

The general prescription in comparative efficiency analysis is to modify a

given governance structure when a superior alternative emerges. However,
in determining whether the new alternative can be implemented with net
gains, one must incorporate various adjustment costs into the efficiency
analysis. Even though the new alternative might offer a comparatively effi-
cient governance structure, the benefits it offers may not always offset the ad-
justment costs associated with its implementation. The QWERTY keyboard
layout offers perhaps the best-​known example.
The QWERTY layout was designed in the 1800s for mechanical
typewriters. One of the problems was that the mechanical arms on which
the keys were mounted would jam if typing was too fast (David 1985). In an
attempt to slow down typing, the layout of the keyboard was intentionally
designed to be inefficient. To this end, the letters would be positioned such
 The Institutionalized Organization  219

that the person typing would have to use the comparatively weaker little and
ring fingers on the commonly used letters, and the comparatively stronger
index and middle fingers on the less commonly used letters.
The advent of computers effectively eliminated the problem of typing too
fast. Yet, the efforts to develop and adopt more efficient keyboard layouts
have been, for all practical purposes, nonexistent. Most of us think that
the QWERTY layout is good enough, and consequently, simply do not see
the point of searching for better alternatives. Many of us simply take the
QWERTY keyboard for granted and are not compelled to search for better
alternatives.
However, there is another angle to the QWERTY story that tends to re-
ceive less attention. Specifically, what is the adjustment cost associated with
switching to another keyboard layout, and can this cost be recovered through
improved efficiency? Williamson (1999, 316) elaborated: “[I]‌mplementation
costs need to be included in the efficiency calculus. It is fanciful to treat two
modes ‘as if ’ they were de novo entrants if, in fact, one has incurred ini-
tial setup costs and has durable, nonredeployable assets in place while the
other has not.” Consequently, a seemingly inefficient governance choice may
persist because even though a more efficient alternative exists, it cannot be
implemented with net gains.7
Whether driven by taken-​for-​grantedness, complacency, convenience, or
lack of high-​powered incentives to make internal inefficiencies agenda items
for management or oversight, it behooves the designer to understand the
consequences of various efficiency distortions present in intraorganizational
relationships in particular. In the following sections, we examine the
implications of institutionalization for organizational efficiency by taking a
closer look at two biases: (1) persistence bias and (2) internal transaction bias.
In ­chapter 3, we approached the question of contracting within and
across organization in terms of the costs of transacting. For example, in
contemplating the make-​or-​buy decision, the designer should incorporate

7 Whether alternative keyboard layouts offer efficiency advantages that offset the adjustment cost
is debatable. David (1985, 332) cited experimental studies conducted in the US Navy in the 1940s
that suggested the Dvorak keyboard layout to provide a superior alternative even after accounting for
adjustment costs. In a strong rejection of David’s argument, economists Stan Liebowitz and Stephen
Margolis (1990) challenged Dvorak’s superiority on two grounds. One was that experiments other
than the Navy study suggested that the Dvorak keyboard’s comparative efficiency in typing speed
was either marginal or nonexistent. The other was that the idea of a new dominant design taking over
an incumbent based on technical efficiency considerations alone was based on “a sterile model of
competition” (Liebowitz and Margolis 1990, 22). Echoing the second point, we propose that ignoring
adjustment costs results in a sterile analysis of comparative efficiency.
220  Governance and the Organizational Life Cycle

both production costs and transaction costs into the analysis. In the fol-
lowing, we revisit the issue by taking a closer look at the advantages and the
disadvantages of specifically internal transactions. We further focus on the
advantages and disadvantages of internal transactions in the context of es-
tablished organizations where governance structures have institutionalized
over time. The general advantage of intraorganizational transacting is that it
has various informational benefits over interorganizational transacting; the
general disadvantage is that persistence of internal governance structures
may give rise to efficiency distortions over time. The disadvantages shed light
on some of the more elusive aspects of transaction costs found in institution-
alized organizations.

Persistence Bias

Organizational macrostructures tend to persist over time, and re-

design is challenging. One source of the challenge is resistance to
change: Organizational redesigns are politically volatile as they tend to create
winners and losers as some subunits and individuals gain and others lose
power (Goold and Campbell 2002b). However, another reason structures
persist is because they simply become taken for granted or adjustment costs
grow large. Here, we highlight this more elusive source of persistence: The
problem in taken-​for-​grantedness is not resistance but the designer’s inability
either to see that redesign is needed or that a new structure with net gains
can be implemented. To explore this idea in detail, let us examine the four
stages through which growing corporations go as they expand in their scale
and scope: (1) ad hoc structure; (2) functional structure; (3) divisional struc-
ture; and (4) matrix structure (e.g., Bernstein and Nohria 2016). Each stage
is supported by a central taken-​for-​granted assumption about its efficiency.
Table 8.1 summarizes the four stages, their taken-​for-​granted assumptions,
and their sources of inefficiency.
Most companies start small, and although some governance issues require
attention already at the outset, a formal organizational structure for the divi-
sion and coordination of tasks tends to be viewed as unnecessary. Instead, the
organization’s overall design is best described as an ad hoc structure: There
are no formal organizational positions and reporting relationships either do
not exist or are managed informally. Organizational members are assigned
to their respective tasks based on either the members’ own preferences or
 The Institutionalized Organization  221

Table 8.1  Taken-​for-​Grantedness in Organizational Structures

Structure Taken-​for-​granted efficiency Source of inefficiency

premise

Ad hoc Small organizations should Low routinization and lack of

remain flexible instead of adopting prioritization
unnecessary formal structures
Functional Creating organizational subunits by Functions have low-​powered
pooling similar tasks economizes incentives and prioritization
on specialization of projects and products is
cumbersome
Divisional Creating organizational subunits Replication of similar tasks in
with self-​contained assets and multiple subunits (e.g., each
outputs promotes accountability division has its own R&D
and enables high-​powered function)
incentives
Matrix Lateral structures and cross-​unit Compromises high-​powered
sharing leads to economies of scope incentives of the divisional
structure

ad hoc rules. The taken-​for-​granted efficiency premise is that small organ-

izations must remain flexible and that adopting formal structures would be
unnecessary (and therefore wasteful). In the incipient organization, this as-
sumption is often reasonable.
As the organization grows, similar tasks must be repeated, and prior-
itization of projects and activities becomes essential. Consequently, the
inefficiencies of the ad hoc structure become amplified to the point that
they exceed the benefits that flexibility has to offer; the ad hoc structure is no
longer comparatively efficient.
In their first attempt at implementing a formalized structure, designers
often find themselves grouping together tasks that are similar in content and
that require similar expertise (Galbraith 1971). Accordingly, employees who
create revenue are assigned to the sales/​marketing department, those who
convert raw materials into final products to the operations department, and
those who create new products to the R&D department. Grouping similar
tasks to form organizational subunits gives rise to the familiar functional
structure. Functional tasks tend to be self-​contained (e.g., operations can
largely be run without inputs from marketing or R&D) and grouping or-
ganizational members with similar functional skills into the same depart-
ment has many informational advantages. The key concept is economies of
222  Governance and the Organizational Life Cycle

specialization, which confers many benefits. The taken-​for-​granted assump-

tion is that pooling a “critical mass” of like-​minded employees into the same
organizational subunit facilitates economies of specialization.
The drawback of the functional structure is that functions tend to have
low-​powered incentives. This problem stems from the fact that although
functional tasks are self-​contained, value creation is not; it is generally im-
possible to organize individual business functions as profit centers. For ex-
ample, running a production unit as a profit center would require that both
the revenue and the cost of the unit are salient. Although costs may be salient,
profit is not. The revenue of the production unit would depend on the price at
which finished products are transferred to the sales/​marketing department.
Because transfer prices are often arbitrary, so is the production unit’s rev-
enue. The solution is to organize the production unit as a cost center, but the
trade-​off is that the incentive intensity of a cost center is lower than that of a
profit center (see ­chapter 2).
Another problem with the functional structure, which often emerges over
time when the organization begins to expand not only in scale (volume) but
also in scope (number of products), is that prioritization of outputs becomes
cumbersome. How does the functional structure incorporate the fact that
some products are strategically more important than others? In the func-
tional structure, those in charge of products or product groups (the product
managers) must negotiate separately with all functional departments for
their contributions.
Just as in the case of the ad hoc structure, the drawbacks of the functional
structure intensify over time. With the broadening scope, low-​powered
incentives and lack of prioritization may become so significant that the
organization’s structure must be redesigned to implement high-​powered
incentives and clear accountability for those in charge of the organization’s
outputs. The structural solution that achieves this is the divisional structure,
where the grouping of individuals is no longer based on tasks but on outputs.
In the divisional structure, high-​powered incentives are created for divi-
sion management by assigning profit-​and-​loss (P&L) responsibility to the
divisions.
Switching to a divisional structure does not mean the functional struc-
ture is eliminated, as each division within the divisional structure may create
a functional substructure within their divisions. The task of the divisional
functions is to serve the needs of the division and its products. The cen-
tral assumption in the divisional structure is that creating subunits that are
 The Institutionalized Organization  223

self-​contained in terms of their tasks, assets, and outputs enhances accounta-

bility and creates high-​powered incentives for divisions to create value.
Divisional structure has the disadvantage of redundancy and replication,
which becomes an efficiency liability in organizations where the outputs of
different divisions share many similarities. Reconsider briefly Volkswagen
Group’s divisional structure (­chapter 3). The functional needs of Volkswagen,
Audi, Seat, and Škoda (four of the largest divisions) are appreciably sim-
ilar. Specifically, each division must design, produce, and sell passenger
automobiles. Due to this significant overlap, it would make little sense to
have all four divisions manage their unique and isolated business functions,
product development in particular. The need to share resources and to lev-
erage competences across divisions gives rise to the matrix structure.
In the matrix structure, there is no single overarching dimension that
defines the macrostructure of the organization; instead, there are multiple
dimensions of strategic importance. The dimensions commonly found in
matrix organizations are product, function, and geography; in some contexts,
there may also be a separate customer dimension (Galbraith 2009). The ma-
trix structure has the advantage of capitalizing on economies of scope by
enhancing collaboration and the transfer of competences and information.
The problem with the matrix structure is that it compromises the high-​
powered incentives of the divisional structure and adds administrative com-
plexity. To the extent that complexity and lower-​powered incentives become
sources of significant inefficiencies, the matrix structure may not be compar-
atively efficient.

Why Persistence Causes Delayed Responses

The four structural solutions and their respective strengths and weaknesses
are well known and documented in the organization design literature. Our
focus in this chapter is on why they tend to persist and what the designer
can do about it. In this context, by persistence we mean a situation in which
a macrostructure other than the one currently in place presents a compara-
tively efficient alternative, but the designer fails to recognize this. The transi-
tion from a functional to a divisional structure provides an example.
It is well established that the functional structure works well if the scope of
the organization’s outputs is narrow and there is no compelling need to pri-
oritize products. But at what point has the scope become sufficiently broad
224  Governance and the Organizational Life Cycle

for the divisional structure to provide a comparatively efficient alternative to

the functional structure? Insofar as persistence is concerned, we propose this
question as the designer’s primary challenge. Furthermore, an inability to ad-
dress the question due to complacency becomes a liability.
Suppose the organization has been expanding the scope of its outputs but
has held on to a functional macrostructure for ten years. Seeing the divisional
structure as potentially more efficient may elude the designer who, quite
simply, views the now-​familiar functional structure as appropriate. Again,
we do not refer here to the common phenomenon of the functionally “siloed”
organization in which functional managers develop entrenched positions
and become protective of their own functions. Instead, we refer to a situa-
tion in which the designer takes the functional macrostructure for granted,
effectively forgoing a continuous and relentless comparison of alternative
macrostructures. The problem is inadvertent complacency, not deliberate
subgoal pursuit. To be sure, the failure of a functionally structured organi-
zation to divisionalize may also be caused by the entrenchment of functional
management, which may make adjustment costs so high that the divisional
structure cannot be implemented with net gains. The only feasible (and de-
cidedly costly) remedy to entrenchment might be the dismissal and replace-
ment of functional managers by new managers who are willing to enter a
divisionally structured corporation.

Persistence of the Internal Organization

To be sure, exchange relationships and cooperative value-​creation activi-

ties both within and across organizations develop taken-​for-​granted char-
acteristics over time, making persistence a potential source of inefficiency
in all exchange relationships. However, persistence likely presents a com-
paratively more significant challenge to internal organization where taken-​
for-​grantedness may be more prevalent and adjustment costs high. As
management scholar Peter Drucker (1973) famously noted, no organization
wants to abandon anything. In contrast, cross-​organizational relationships
tend to be subjected to periodic evaluation and active renewal decisions.
Many notable economists such as Friedrich Hayek (1945) have noted that the
fundamental advantage that markets (i.e., interorganizational transactions)
have over firms (i.e., intraorganizational transactions) is their ability to adapt
autonomously over time.
 The Institutionalized Organization  225

Consider the decision whether a company should have an internal legal

department or whether it should contract with an external law firm. Legal
scholar David Wilkins (2012) suggested that three different factors explain
the increasing use of in-​house counsel in US corporations: (1) the eco-
nomic argument suggests that in-​house counsel lowers legal costs; (2) the
substantive argument suggests that internal lawyers give better guidance be-
cause they have more intimate knowledge of the organization; and (3) the
professional argument suggests that as company employees, inside lawyers
are more likely to take the organization’s long-​term interests into account.
Wilkins (2012, 260) further suggested that these three arguments have be-
come embedded in “the ideology of the in-​house counsel movement [that] is
now broadly accepted throughout the US legal profession.”
The drawback of internal counsel is that internal lawyers are employees
whose contracts are more difficult to terminate than those of outside counsel.
To invest in internal counsel is to commit to a fixed amount of legal expertise
within the organization. Much like having an internal division that produces
and supplies parts to another division, having an internal counsel becomes
a sunk cost. Moreover, the incentive intensity of the internal counsel will be
lower than that of the external lawyer. The billable-​hour method and billable
targets used in law firms tend to create high-​powered incentives to use the
lawyer’s time as efficiently as possible.
With internal counsel, billable targets are not as relevant (if at all), which
reduces incentive intensity. With lowering incentive intensity, “corporate
counsel may feel pressure to accentuate their legal capabilities in order to
be seen as valuable and non-​redundant” (Jenoff 2012, 733). Furthermore,
when in-​house lawyers work for just one client (i.e., their employers), there
may be added pressure to broaden the lawyer’s role from an advocate and a
legal advisor to mediator, compliance officer, even a business team member
(Jenoff 2012, 732). This broadening scope of activities may confer benefits on
the one hand, but on the other hand, the comparatively high levels of special-
ization of the legal profession may become a liability. When lawyers work as
compliance officers or business team members, their skills are not likely in
their best or even next-​best use, and yet, the cost to the organization remains
the same. Given the organization has decided to hire internal lawyers, it must
find them something to do.
The in-​house counsel example is useful because it highlights the problem
of persistence in relationships that involve contractual relationships with
human capital. In the case of technological assets, persistence tends to be less
226  Governance and the Organizational Life Cycle

problematic due to depreciation. Designers need not worry to the same de-
gree about technological assets persisting, because technological assets be-
come obsolete over time as they lose their value through depreciation. The
persistence of fixed assets does not occur inadvertently but, instead, requires
active reinvestment decisions. In these reinvestment decisions, the designer
can engage in an explicit deliberation over whether reinvestment serves an
efficiency purpose. In stark contrast, organizational subunits, structures, and
employment contracts do not depreciate in the accounting sense and, conse-
quently, can persist without explicit attention, deliberation, and active rein-
vestment decisions. At the same time, subunits, structures, and contracts can
depreciate in the sense that their ability to create value declines over time.
What makes such organizational (as opposed to technological) efficiencies
particularly vexing is that complacency on the part of the designer may cause
them to persist for long periods of time.
The use of in-​house counsel is an illustration of the general challenge that
persistence bias causes specifically within organizations. In the following sec-
tion, we discuss the general topic of trade-​offs in internal transactions. Just as
it is important for the designer to engage in a comparative analysis of intra-​
versus interorganizational contracting (see ­chapter 3), understanding the
trade-​offs associated with intraorganizational transacting is essential.

The Trade-​Off of Internal Transactions

Further building on the premise that the problem is one of complacency,

taken-​for-​
grantedness, and adjustment costs, we turn to governance
problems that are in the economics literature discussed under the rubric of
transactional distortions (Williamson 1975, 118). Here, we adopt the label ef-
ficiency distortion, as it is a descriptively better title for our purposes: We are
specifically interested in how, over time, some aspects of the organization’s
governance may become “distorted” in that what once was comparatively ef-
ficient no longer is so. We further highlight the largely inadvertent drivers
of distortions; they occur as governance choices gradually shift out of align-
ment with the characteristics of the relationship and the environment in
which the relationship is embedded.
To examine the trade-​off associated with internal transactions, consider
the situation in which an organization consists of multiple subunits (e.g.,
divisions) that transact with one another. For example, organization design
 The Institutionalized Organization  227

scholar Jay Galbraith (2002a, ­chapter 8) described the front/​back model, an

organizational structure found in many large technology corporations that
provide solutions of large scale and scope to their corporate clients. The
front/​back hybrid is a special case of the multidivisional firm that consists
of market-​facing divisions (the front structure) that host the customer op-
erations, and technology-​facing divisions (the back structure) where in-
dividual products and services are developed and maintained. Galbraith
(2002b, 201) discussed the networking and telecommunications equipment
company Nokia Networks as an example of the front/​back hybrid. The task
of the front structure at Nokia Networks was to provide network solutions
to massive cellular network operators such as AT&T, Vodafone, and France
Télécom (now Orange). The back structure, in contrast, was responsible for
developing the individual technologies (e.g., IP networks, data center in-
frastructure, cyber security, and operations support systems) that the front
assembled into network solutions. The central organizational challenge in
the front/​back hybrid is how to connect the front with the back.
Why are the components developed and the solutions assembled by in-
ternal units of the same organization? What advantages does the integrated
front/​back hybrid offer over the alternative where one firm creates cus-
tomer solutions by purchasing the requisite components and subsystems
from external vendors? In the following, we discuss the advantages and the
disadvantages of transacting internally.

The Advantages of Transacting Internally

What is the price of a gallon of milk? A smartphone? A flight from Madrid

to Chicago? In many instances, acquiring information on prices is a simple
matter of discovering the market price. If a product or a service can be bought
at an unambiguous market price, the advantages of internal transacting be-
come elusive. Indeed, why would an organization choose to internalize an
activity it could easily contract an external supplier to execute at a specific,
predetermined market price?
There are two general situations in which prices are not a simple matter of
discovery. One is when the object of exchange is unique, and consequently,
there simply is no market; the other is when the object of exchange creates
value only when it is integrated into a broader product or service to create
value. For example, how do the automotive seating manufacturer Adient and
228  Governance and the Organizational Life Cycle

the final assembler of automobiles (e.g., Volkswagen) determine the price at

which Adient supplies car seats to Volkswagen’s final assembly? Even though
Adient’s seats may not be unique, there really is no competitive market that
provides salient market prices. What is more, car seats do not provide value
independently of other components, instead, they must be embedded within
a broader system of the automobile. When the value an individual product or
service provides is fundamentally contingent on other products and services,
its price cannot be based on the value it provides; cost-​based pricing offers
the obvious alternative. In such contexts, transacting internally may provide
an informational advantage over transacting in the market.
In cost-​plus pricing, the exchange takes place at a price that consists of the
seller’s cost and the seller’s margin. But what is the seller’s cost, and how is the
margin determined? If the object of the exchange is a component that is to
be integrated into a broader system, what is the cost of integration? Will the
buyer absorb the cost of integration, or should it be taken into consideration
when the seller’s margin is determined? Are the buyer and the seller aware of
one another’s costs, or are there information asymmetries? How are disputes
resolved?
Unique products or intermediate products that must be integrated into
final products obviously have a price. But these prices are not so much discov-
ered in the market as they are constructed by the transacting parties (Eccles
and White 1988). This process of constructing prices is facilitated if the
transacting parties are parts of the same parent organization. A corporate
parent that has access to the cost information of both parties is in a unique
position to mediate the process as an impartial facilitator. In economic
terms, when the transaction is intraorganizational, management may ful-
fill an important quasijudicial function (Williamson 1975, 30) and mediate
potential disputes. If the transaction were interorganizational, the dispute
might escalate more easily to a point where it strains the relationship. When
transacting internally, “the parties are more inclined to adapt cooperatively”
(Williamson 1975, 30). In addition, cooperative adaptation is mandated in
the sense that internal disputes must be handled internally—​the courts will
not settle such disputes.
In sum, the advantages of transacting internally arise from two related
sources. One is that informational asymmetries cause comparatively fewer
problems due to the presence of a parent that has access to information on
both sides of the transaction. The other advantage is that the organization
facilitates ex post adaptation when disputes arise.
 The Institutionalized Organization  229

The Disadvantages of Transacting Internally

In the front/​back hybrid, what are the technologies the front purchases
through internal transactions from the back, and where can it use its dis-
cretion and purchase from the market? Symmetrically, which products
and solutions does the back sell exclusively to the front, and when can it
sell to external customers? Should the parent organization mandate in-
ternal transactions whenever they are possible? Are internal transfer prices
set unilaterally by the parent or are the internal buyer and seller allowed to
negotiate?
The designer must understand that all policies that limit the internal buyer’s
and the internal seller’s discretion tend to lower incentive intensity. When
the internal buyer has a captive internal supplier, and vice versa, the incentive
intensities of both parties inadvertently and unavoidably decline. Unlike in
the case of deliberately designed low-​powered incentives (see c­ hapter 5), in-
ternal transactions tend to be associated with unintentionally lower incentive
intensities. In the front/​back hybrid, if those responsible for the back know
that their outputs are always guaranteed to receive demand from the front,
their declining incentive intensity may have adverse consequences such as
underinvestment in R&D. Specifically, if the back has a captive buyer who is
required to purchase technologies for customer solutions from the back, the
latter is immediately disincentivized to invest in R&D.8 Symmetrically, if the
front is guaranteed an internal supply of components from the back, its in-
centive to maintain its own purchasing competences declines.
Even though mandating internal transactions may be sensible at the
outset, it may result in internal cross-​subsidization that ultimately protects
nonviable capabilities. The phenomenon is sometimes called internal pro-
curement bias (Williamson 1975, 119).
Internal transactions become even more problematic if the inefficiencies
become salient to the transacting parties. For example, the problem of in-
ternal procurement bias is further exacerbated if an internal supplier realizes

8 If the back is grouped into profit centers and executive compensation is linked to the unit’s prof-
itability, there is an immediate disincentive to invest in R&D, especially if demand from the front is
guaranteed by the parent organization. But why would a front/​back hybrid assign P&L responsibility
to the back instead of the front? If solutions are modular and consist of interchangeable components
that have a market price (e.g., routers and switches in the telecommunications equipment context), it
may well make sense to assign primary P&L responsibility to components or products (the back) and
only secondary P&L to solutions and customer accounts (the front). Some front/​back hybrids cal-
culate revenue in both the product and the customer dimensions and implement a hierarchy of P&L
structures.
230  Governance and the Organizational Life Cycle

it could receive a higher price for its outputs if it sold them to an external cus-
tomer, but at the same time, is required by the corporate parent to trade only
internally. Symmetrically, an internal customer may realize it could obtain a
component at a lower price from an external supplier. Efficiency distortions
that arise from mandated internal transacting are a cause for concern, but
when these inefficiencies become salient to the transacting parties and are
not addressed, the viability of the entire organization is threatened.
Importantly, the problem is more fundamental than that of getting the
transfer prices right. If internal procurement bias has led to a situation in
which an organization is maintaining a nonviable competence through
cross-​subsidization, the obvious solution is to seek a viable competence in
the market and terminate cross-​subsidization.

Summary: Institutions as a Liability

Well-​defined institutions are the bedrock of all functioning societies and

economic systems; their importance cannot, and must not, be understated.
At the same time, the informed designer is well advised to give attention to
problems that arise from taken-​for-​grantedness that may follow from the
“infusion of value beyond the technical requirements of the task at hand”
(Selznick 1957, 16–​17).
To be sure, losing sight of “the technical requirements at hand” due to
taken-​for-​grantedness of an institutionalized governance structure may be-
come a liability in all exchange relationships. However, institutionalization
poses a hazard particularly in situations in which the relationship is not re-
lentlessly monitored for efficiency. This can occur especially in those internal
exchange relationships where depreciation does not provide the requisite
mechanism for eliminating obsolete, non-​value-​adding activities. Whenever
the exchange relationship is subjected to periodic review and evaluation, the
designer has the opportunity to be relentlessly comparative. Capital rein-
vestment decisions and regular contract renewal present such opportunities,
but in the case of internal organizational structures and processes, these
opportunities tend to be absent. This may give rise to various efficiency
distortions.
In ­chapter 3, we compared intra-​and interorganizational transactions in
an attempt at specifying the conditions under which each is comparatively
efficient. In this chapter, the objective has been to complement ­chapter 3 by
 The Institutionalized Organization  231

examining the trade-​offs associated with internal transactions. The well-​

established advantage of intraorganizational transactions is that they offer
various informational benefits. In situations where prices cannot be simply
discovered but must instead be constructed, the internal organization offers
several advantages. One is the existence of impartial general management
that can alleviate problems that stem from asymmetric information. Not
only is cooperation and adaptation easier within organizations, but inter-
nally transacting parties have also an incentive to cooperate, because disputes
must be resolved as matters of private ordering. To this end, the internal or-
ganization has central quasijudicial functions for internal dispute resolution.
The disadvantages of the internal organization are more elusive, which is
why we have devoted explicit attention to them here. The disadvantages be-
come pronounced in established organizations that have a broader set of in-
stitutionalized structures, activities, and exchange relationships within their
boundaries. To the extent that institutionalization within organizations leads
to inadvertent outcomes such as gradually declining incentive intensity, in-
ternal governance structures start to exhibit efficiency distortions.
Efficiency distortions alone do not imply that intraorganizational
relationships are comparatively inefficient and inferior to interorganizational
relationships. As in all governance decisions, our prescription to the designer
is to analyze the issue in its entirety. In the case of internal transactions, the
designer should consider the advantages jointly with the potential efficiency
distortions due to internal organization. The problem with the failure to in-
corporate efficiency distortions into the comparative analysis is that it makes
the designer complacent. Considering the benefits of internal transactions
while ignoring the drawbacks leads to bias specifically toward internal
transactions. Complacency may ultimately prove costly, as in the case of di-
vision managers discovering that their divisions were paying a 4,000 percent
premium for centralized photocopying services. If an efficiency distortion of
this magnitude can persist for years, one cannot help but wonder what kinds
of remediable inefficiencies are embedded in the taken-​for-​granted practices
of established organizations.
 Epilogue

Economist Frank Knight (1941, 252) prescribed to those interested in

behaving economically to “make their activities and their organization ef-
ficient rather than wasteful,” and that “[t]‌his fact deserves the utmost em-
phasis.” Building on this idea, the pioneering organization economist Oliver
Williamson (1988, 571) suggested that the resultant “organizational impera-
tive” is to design governance structures that align with the characteristics of
the relationships the organization has with its constituencies. The ultimate
objective of all the chapters and illustrations in this book has been to explore
the ramifications of this imperative.
Although we might not go as far as to suggest that efficiency deserves the
“utmost emphasis,” we maintain that it certainly deserves the designer’s at-
tention. In c­ hapter 1, we described efficient organization as analogous with
a healthy blood pressure: certainly a worthy objective, but hardly the most
important aspect of one’s health, let alone the reason for one’s existence. We
believe this analogy creates a useful connotation of how the designer can,
and should, think of efficient organization.
Since all organizations involve a multitude of diverse relationships, they
also involve a multitude of governance structures; indeed, each relationship
in which the organization is involved has its own governance structure. For
example, the organization has a relationship with all its employees, and each
relationship may exhibit both general and idiosyncratic features. Similarly,
governance structures with providers of capital must be structured in a way
that secures mutual credibility and continuing cooperation. Relationships
with those entitled to fixed payments (debt financing) must be structured
differently than relationships with those entitled to residual payments (eq-
uity financing). Governance, as a general term, refers to the totality of all
these structures that support and secure the organization’s relationships with
its constituencies.
The reason we refrained from using the term corporate governance in this
book is because we have sought to bring the level of analysis to the relation-
ship. Consequently, we highlight the notion of governance of contractual
relationships, a term that is well established in the contemporary literature
234 Epilogue

on organization economics. The distinction between corporate governance

and governance of a contractual relationship is not merely semantic; it has
profound implications for designers of organizations. One is that since the
totality of contractual relationships tends to be context-​specific and idiosyn-
cratic, so are governance structures. Therefore, mimicry and benchmarking
of other organizations may be misguided. Instead, the designer must rigor-
ously analyze the key relationships and choose governance structures ac-
cordingly. A common misconception is that the institutional environment
determines governance structures and that those facing a similar institu-
tional environment should have identical governance structures.
That all organizations have a multitude of contractual relationships
(and a multitude of governance structures) has another profound implica-
tion: Efficiency will always be relevant in one way or another. We are hard-​
pressed to think of an organization where not a single relationship in which
the organization is involved would not benefit from a comparative efficiency
analysis. Consequently, it behooves all designers to identify and carefully an-
alyze relationships where seeking efficient governance is relevant. Efficiency
arises from making trade-​offs in a way that results in a positive net effect, or
net gains, for the parties in the relationship.
Symmetrically, the designer must be mindful of relationships in which
comparative efficiency does not apply. Critically, there are relationships
in which the objective of positive net gains becomes infeasible, and the
governance-​ as-​
efficiency approach no longer applies. Even in these
relationships, the designer must still make a governance choice; however, the
choice can no longer be justified in efficiency terms.
The applicability of comparative efficiency analysis is not a question of
whether the organization is for-​profit or nonprofit, or whether it is public
or private. What is relevant is not the organizational form but the context
and the characteristics of the specific relationships. Comparative efficiency
analysis loses its relevance in relationships where the alternative governance
choices involve outcomes that cannot (or must not) be traded off against one
another in the conventional economic sense. The informed designer will
be able to identify such situations and supplant an economic analysis with
other forms of deliberation and judgment to determine what is appropriate,
not what is efficient. In many relationships in which comparative efficiency
is no longer operational, governance-​as-​integrity takes precedence over
governance-​as-​efficiency.
 Epilogue  235

Importantly, that integrity takes precedence over efficiency does not imply
rejection of the latter. Once integrity of a relationship has been secured,
an efficiency analysis may well be applicable to the same relationship in a
way that not only fosters efficiency but simultaneously further bolsters in-
tegrity. A case in point, in most contexts the separation of management
and oversight—​separation of powers—​tends to promote both integrity and
efficiency.
Problems arise when the quest for efficiency compromises integrity,
which can happen in two ways. One is that efficiency is considered in short-​
term, myopic terms; the other is that efficiency analysis is used outside its
boundaries of applicability. In both cases, the problem is not efficiency but
its uninformed application. We hope that this book has established that such
misapplication is remediable, and that comparative efficiency analysis, prop-
erly applied, can always either provide the foundation or a complement to
other organizational objectives. To this end, we emphasize two principles to
guide the way. One is the idea of other-​regarding behavior proposed by legal
scholars Margaret Blair and Lynn Stout. The other principle is adopting the
kind of an analytical mindset that organization economist Oliver Williamson
endorsed throughout his career: Have an active mind, be disciplined, and be
interdisciplinary.
We hope this book can serve as a catalyst in the proliferation of analytical,
other-​regarding designers.
 Glossary of Terms

This glossary of terms contains the definitions and the descriptions of the
central terms used in this book. However, instead of presenting conventional
formal definitions, our objective is to establish the relevance of each term in
the context of organization design and governance. To this end, this glossary
is constructed according to two principles.
The first principle is that for a concept to be relevant to organization de-
sign and governance, it must be explicitly contextualized. Consequently,
the definitions offered in the following are not meant as formal, universal
definitions; rather, they are presented specifically in the context of the ef-
ficiency approach adopted in this book. The definition of governance is an
illustrative example of context dependence. Our definition reflects the con-
tractual, private-​ordering aspects of governance. In contrast, those adopting
an institutional perspective (organizations such as the OECD, for in-
stance) might define governance from the point of view of compliance, not
contracting. Different definitions serve different purposes.
The second principle is inspired by the concept of nomological validity in
quantitative psychology: A concept does not acquire its full meaning until
it is considered in conjunction with other concepts with which it is related.
Consequently, in discussing the definition of a given concept, we may in-
voke several other concepts simultaneously. These other concepts are either
intimately related to the focal concept, crystallize its meaning, or provide a
useful contrast. Discussing residual claimant in conjunction with residual is
a good example. Specifically, the concept of the economic residual remains
merely a measure of net income until we ask whether someone can present a
legitimate claim for it. Introducing the residual claimant makes the concept
of residual relevant to governance.
As an example of one concept providing a contrast to another, transac-
tional contracting provides a useful contrast to relational contracting, as does
hazard to risk.
238  Glossary of Terms

Terms Included in This Glossary

Alignment (and Adjustment)

Contracting (Transactional vs. Relational, Complete vs. Incomplete)
Credibility (and Credible Commitments)
Dependency (Unilateral vs. Bilateral)
Design (and Designer)
Efficiency
Efficiency Distortion
Ex Ante vs. Ex Post
Failure
Feasibility (and Remediableness)
Fiduciary (and Fiduciary Duty)
Forbearance (and Quasijudicial Functions)
Fundamental Transformation
Governance
Hazard
Incentive Intensity
Institution
Main Problem
Management
Myopia
Organization
Oversight
Ownership
Private Ordering (vs. Legal Centralism)
Probity (and Probity Hazard, Sovereign Transaction)
Profit-​seeking vs. Nonprofit Organizations
Public vs. Private Organizations
Rationality (and Bounded Rationality, Self-​Interest vs. Opportunism)
Residual (and Residual Claimant, Residual Interest)
Risk (and Switching Cost)
Safeguard
Specificity (and Temporality, Dedicated Assets)
Stakeholder (and Stakeholder Analysis)
Transaction Cost
 Glossary of Terms  239

Value Creation vs. Value Capture (and Preappropriation vs.

Postappropriation)
Viability

Alignment (and Adjustment)

Some environments are more uncertain and change more rapidly than
others; some contracts are subject to more transactional hazards and risk
than others; some contracts are short term, others are long term. The contexts
in which organizations operate and in which contractual relationships are
embedded are highly diverse.
The foundational idea in efficient organization is that governance choices
must align with the context. It is incumbent upon the designer to discrim-
inate, that is, to derive the efficiency implications of different alternatives,
and then choose the one that is comparatively efficient. Williamson (1991,
277) used the term discriminating alignment to describe the outcome in
terms of aligning transaction characteristics with governance choice.
Importantly, because an organization has a multitude of relationships with
diverse constituents, discriminating alignment must be applied separately to
all relationships.
For example, if the purchasing manager of an industrial firm faces the
governance choice of either making a component in-​house or outsourcing it
to an external supplier, discriminating alignment calls for an analysis of the
efficiency implications of the two alternatives. If the part is something the
firm buys frequently, if its availability is subject to high uncertainty, and if
the exchange relationship involves relation-​specific investments, then a dis-
criminating alignment analysis likely leads the manager to a conclusion that
it is more efficient to produce the component in-​house to better safeguard
against transactional hazards.
Due to various inertial forces, organizations tend to be more stable than
their environments. The inability to adjust instantaneously without cost
means that over time, governance choices may drift out of alignment with the
demands of the organization’s environment. In the case of the industrial firm,
perhaps technological developments eliminate the need of relation-​specific
investments, and consequently, the use of an internal supplier becomes
less efficient than outsourcing. In this case, maintaining discriminating
240  Glossary of Terms

alignment prescribes governance adjustment, which is always associated

with a cost. Adjustment cost is a special case of an ex post transaction cost.
In considering adjustments due to misalignment, the designer must exer-
cise discrimination as well. Some misalignments are inconsequential in the
sense that even though realignment might enhance efficiency, the improve-
ment is so small that the organization may not even be able to recover the
associated adjustment cost. Use of discrimination must always involve the
analysis of net gains.

Contracting (Transactional vs. Relational,

Complete vs. Incomplete)

We distinguish between the formal/​legalistic and the informal/​collaborative

domains of contracting. A written contract that stipulates the responsibilities
of the contracting parties is a good example of the former. The formal do-
main links to law in the sense that disputes over formal contracts are often
settled in the context of jurisprudence.
Even though formal contracts are plainly essential, this book seeks balance
by incorporating also the informal and the collaborative side of contracting.
By making the formal/​informal distinction, we want to point to the ex ante
and ex post features of contracting. The notion of a formal contract implies
that the key aspects of contracting are determined ex ante (before the con-
tract is in force). Consider a simple employment contract: The formal ex
ante contract specifies the rights and the responsibilities of the contracting
parties. In case of a dispute, the most important document to which the
parties will refer is the formal contract (as well as the applicable laws, which
of course are in force ex ante as well).
Most employment contracts are transactional in the sense that the formal
contract is sufficient to safeguard the relationship. Employer-​ employee
relationships become more complex when employees become bearers of re-
sidual risk and develop a legitimate residual interest. These relationships are
no longer merely transactional but also exhibit relational characteristics. The
formal contract is still relevant but may no longer be sufficient in and of itself.
The relationship between a limited liability company and its shareholders
is a good example of a relationship in which a formal contract is inconse-
quential. Even though the relationship can still be considered contrac-
tual, there is no formal contract that stipulates the responsibilities of the
 Glossary of Terms  241

contractual parties, the duties of the corporation in particular. In the con-

tractual relationship between the corporation and the shareholder, the share-
holder is given nothing but a nebulous promise that the organization’s board
of directors will incorporate shareholder interests into its decisions. Unlike
in the case of the employment contract, there is no formal ex ante contract
that specifies in detail what incorporating shareholder interests means.
Furthermore, if the board fails in its attempts at producing returns to the
shareholder, the shareholder may have no recourse.
The formal and the informal are not mutually exclusive. Indeed, informal
contracting can sometimes complement formal contracts. Goldberg (1976,
428) elaborates:

While the [contracting parties] might want to go into considerable detail

at the formation stage concerning the rights and obligations of each party
given various contingencies, it will often prove too costly to specify the
precise terms of the contract and it will be desirable instead to use rough
formulae or mutual agreement to adjust the contract to current situations.
As the relational aspects of the contract become more significant, emphasis
will shift from a detailed specification of the terms of the agreement to a
more general statement of the process of adjusting the terms of the agree-
ment over time.

Consequently, the contract becomes more a general framework than an

explicit agreement (Williamson 2002b).
The notion of a complete contract pertains to contracting situations in
which the responsibilities of the contracting parties can be specified and
codified into the formal contract in a way that renders ex post adjustments
immaterial. In contrast, relational contracting characterizes situations where
the contracting parties must rely on the mutual agreement to adjust ex post.
Relational contracting is often found in transactions involving bilateral de-
pendency and specificity that require various additional safeguards to ad-
dress contractual risk.
Williamson (1996, 378) offered an extensive list of reasons why a complete
ex ante contract may be infeasible:

(1) not all the relevant future contingencies can be imagined, (2) the details
of some of the future contingencies are obscure, (3) a common under-
standing of the nature of the future contingencies cannot be reached, (4) a
242  Glossary of Terms

common and complete understanding of the appropriate adaptations to fu-

ture contingencies cannot be reached, (5) the parties are unable to agree on
what contingent event has materialized, (6) the parties are unable to agree
on whether actual adaptations to realized contingencies correspond to those
specified in the contract, and (7) even though both the parties may be fully
apprised of the realized contingency and the actual adaptations that have
been made, third parties (e.g., the courts) may be fully apprised of neither.

Credibility (and Credible Commitments)

Entering into exchange relationships involves various degrees of transac-

tional risk. The relationship is credible when the contracting parties volun-
tarily “make a wager” by committing to the exchange despite the risk involved.
Note that the question is not whether one party finds the other credible but,
rather, whether the conditions of the contract are such that they secure the
requisite cooperation of all exchange parties—​credibility is a property of the
arrangement, not the parties themselves. These conditions are met when all
parties have made sufficient mutual credible commitments, described by
Williamson (1983, 519) as “reciprocal acts to safeguard a relationship.”
Even though credibility is, for all practical purposes, synonymous with
trust, using a common term as a label for what is ultimately a complex issue
may be misleading:

[U]‌ser-​friendly terms [such as trust] do not encourage us to examine the

deep structure of the organization. Rather, we need to understand when
credible commitments add value and how to create them, when repu-
tation effects work well, when poorly, and why. Trust glosses over, rather
than helps unpack, the relevant micro-​analytic features and mechanisms.
(Williamson 1996, 216)

We agree with Williamson that the term mutual credible commitments

may be more conducive than trust in inviting the designer to explicate the
principles, structures, and actions necessary for securing collaboration
under transactional risk. This said, we ultimately find that what is essential is
elaborating the mechanisms by which risky relationships are safeguarded—​
choosing which labels to use to describe the mechanisms is of secondary
importance.
 Glossary of Terms  243

Dependency (Unilateral vs. Bilateral)

Two exchange parties are bilaterally dependent if the consequences of

the termination of the relationship are so significant that they require the
designer’s ex ante and ex post attention in both organizations. For example,
supplier switching costs on the buyer’s side and buyer switching costs on the
supplier’s side may both be sufficiently high for the parties to be motivated
to implement the proper safeguards against an unwanted termination of the
relationship.
If switching costs are high for one party and negligible for the other, de-
pendency is unilateral. Contracting under unilateral dependency is funda-
mentally different from contracting under bilateral dependency, because
additional safeguards would be redundant for the party that faces a negligible
switching cost. Unilateral dependency is often found in contracting situations
in which one of the parties is significantly larger and more powerful than the
other. In this book, we have little to say about transacting under unilateral
(or strongly asymmetric) dependency, other than perhaps making the un-
surprising observation that under unilateral dependency, the comparatively
more powerful exchange party sets the rules which the comparatively less
powerful party either accepts or rejects. In a one-​time exchange, accepting
unilateral dependency would be myopic, but with frequent exchanges, there
can be interproject and intertemporal spillovers. Under such conditions, the
dependent party, say, an original equipment manufacturer to Dell, could
gain reputational effects that increase its chances of striking deals outside the
dyadic exchange. Over time, the supplier may become more valuable to Dell,
which transforms the relationship to bilateral dependency.

Design (and Designer)

The most intuitive way of thinking about organization design is to think of

the organization’s structure. Is the organization structured by function (func-
tional organization)? Is the corporation a single-​business or a multibusiness
firm? If the organization is a matrix, what are its central dimensions? It is also
straightforward to think that organizations are structured the way they are
because that is how they were deliberately designed. To be sure, multibusiness
firms do not simply emerge over time without deliberate design decisions.
244  Glossary of Terms

In this book, we embrace the premise that organizations are deliberately

designed. We use the common label designer to refer to all those involved in
the design decisions of an organization. For example, in a limited liability
company, the three central designers are shareholders (who design by voting
in shareholder meetings), the board of directors (who design by making
decisions in board meetings), and top executives (who design by making
executive decisions). That there is a designer also carries the assumption
that the designer seeks to be rational in its decisions. In our exposition, ra-
tionality of the designer is aimed at building and maintaining an efficient
organization.

Efficiency

If there are two ways of organizing an activity, one should choose the one that
uses a smaller number of inputs to produce the output. Throughout this book,
efficiency is a comparative notion that invokes the known alternatives: Of
all the known and available options, the one that produces the least amount
of waste is comparatively efficient. All feasible options (the comparatively
efficient option included) are flawed in the sense that they produce at least
some waste.
The most salient example of waste is physical waste produced by a pro-
duction line. However, for the purposes of this book, various forms of organ-
izational waste are central. The free-​riding problem is a good example. In an
environment where individuals can minimize their effort without sanctions,
productivity (efficiency) is comparatively lower compared to an environ-
ment where free-​riding is “metered well” and free-​riding sanctioned: “If the
economic organization meters poorly, with rewards and productivity only
loosely coupled, then productivity will be smaller; but if the economic or-
ganization meters well productivity will be greater” (Alchian and Demsetz
1972, 779).
Ineffective communication can also sometimes constitute a form of sig-
nificant waste. This may be due to opportunistic behavior and the deliberate
dissemination of false or misleading information. However, there are also be-
nign sources. For example, consider the buyer-​supplier relationship of a final
assembler and a component supplier of a smartphone. Suppose that in this
relationship, the buyer and the supplier must engage in continuous collab-
oration and problem-​solving to ensure continuous innovation of products.
 Glossary of Terms  245

These processes involve various forms of communication: emails, video-

conferencing, phone calls, product team meetings, and the like. It is easy to
see how the requisite communication is more challenging (and potentially
wasteful) in situations in which the buyer and the supplier represent different
firms. An obvious reason is that within firms, classified information travels
with less friction and many problems can be solved using business judg-
ment and authority. In contrast, problem-​solving in cross-​firm relationships
requires (more time-​consuming) negotiation. Finally, the use of language
and various coding procedures (e.g., accounting practices) tend to be more
standardized within organizations, which facilitates intraorganizational
communication.

Efficiency Distortion

In established organizations, internal structures, processes, and principles

may give rise to inefficiency. For example, a production unit in an industrial
firm may use an internal supplier for some of the components needed in final
assembly. Over time, this arrangement may lead to internal procurement bias
in which “the existence of an internal source of supply tends to distort pro-
curement decisions [ . . . ] A norm of reciprocity easily develops, [for ex-
ample], I buy from your division, you support my project proposal or job
promotion” (Williamson 1975, 119, 120).
Another manifestation of efficiency distortion is internal expansion bias,
where the organization adopts “a compromise solution by which concessions
are made to subsystems rather than require them to give up essential
functions or resources” (Williamson 1975, 120).
A third example is persistence bias, where “sunk costs in programs and
facilities of ongoing projects insulate existing projects from displacement by
alternatives which, were the current program not already in place, might oth-
erwise be preferred” (Williamson 1975, 121).
Internal procurement bias, internal expansion bias, and persistence bias are
all examples of what is colloquially described as “an organization assuming a
life of its own.” A common denominator in the examples is that relationships
whose features might suggest a comparatively transactional, arm’s-​length
contracting approach, start to develop unnecessary relational characteris-
tics simply because the contracting parties are parts of the same corpora-
tion. Much like the fundamental transformation, efficiency distortions are
246  Glossary of Terms

emergent phenomena that are difficult to prevent. The designer’s task is to

think of both ex ante and ex post measures for addressing them.

Ex Ante vs. Ex Post

In the governance of contractual relationships, it is useful to distinguish be-

tween the time before (ex ante) and the time after (ex post) the relationship
begins. In the case of formal contracts, the signing of the contract separates
the two.
In general, the designer should seek to turn as many ex post problems as
possible into ex ante problems, because problems are usually more efficient
to preempt than to solve after they have occurred. However, it is important
to realize that many problems in relational contracting in particular cannot
be addressed ex ante, and instead, require various ex post adjustments. The
necessity of various ex post adjustments renders many complex, long-​term
contracts unavoidably incomplete.

Failure

In the context of contractual relationships, failure means that the requisite co-
operation of the exchange parties is not secured even though there is mutual
interest in the transaction. Contractual failure occurs when the contracting
parties fail to implement the requisite safeguards to address risk to which
the contracting parties would have to expose themselves. Contractual failure
is a special case of the more common notion of market failure, where both
demand and supply exist for a product or service, but the transaction fails
to take place because, for one reason or another (e.g., asset specificity, asym-
metric information, or externalities), supply and demand do not meet.

Feasibility (and Remediableness)

Comparative analysis of feasible alternatives stands at the foundation of all

the analyses, practical implications, and recommendations found in this
book. Focus on comparative analysis has a number of important implications.
 Glossary of Terms  247

The first implication is that a governance alternative should not be

compared to an infeasible ideal. One sometimes encounters governance
critics who play the devil’s advocate by pointing to the flaws of a given alter-
native. But no governance alternative is perfect; therefore, a person pointing
out flaws is merely stating the obvious. To merit the designer’s attention,
the critic should be compelled to present a feasible alternative that can be
implemented with net gains. To this end, designers must remain relentlessly
comparative in their approaches to governance decisions.
The idea of remaining relentlessly comparative invokes the notion of
remediableness. A governance problem is remediable only if there is another
alternative that can be implemented with projected net gains. Because gov-
ernance choices tend to affect the entire organization or the entire contrac-
tual relationship instead of just one part of it, the consequences of any given
choice must be analyzed in their entirety (Williamson 1996, 9).
A good example of the importance of projecting net gains is the make-​or-​
buy decision. Suppose that the designer of an industrial firm worries about
the high cost of producing a component internally and, consequently, turns
to outsourcing as the alternative. Failing to incorporate the increased trans-
action costs, the designer may conclude that because the internal production
cost of a component exceeds the external supplier’s asking price, outsourcing
is the comparatively efficient option. However, if the increases in transac-
tion costs exceed the production cost savings, switching from internal pro-
duction to outsourcing will not result in net gains. Considering the issue in
its entirety by incorporating transaction costs might lead the designer to
conclude that no net gains can be projected. Consequently, the uncomfort-
ably high cost of producing the component internally is not remediable by
outsourcing.
Identifying a better alternative after the fact often fails the feasibility cri-
terion. In the outsourcing example, suppose that after failing to incorporate
transaction costs, the industrial firm outsources the component and sells the
facility and the equipment used to produce the component to the supplying
firm. After realizing that the outsourcing option proved even more expensive,
reversing the outsourcing decision and repurchasing the production facility
may no longer be feasible. In general, remorsefully looking back at choices
with perfect hindsight tends not to be practically useful. Organizations must
live with the choices they make, which is why designers must emphasize con-
scious foresight in all decisions.
248  Glossary of Terms

Fiduciary (and Fiduciary Duty)

Constituencies such as employees, suppliers, and lenders have an unam-

biguous contractual relationship with the organization. In the contracting
process, the exchange parties negotiate the rights and the responsibilities of
each, which then become embedded in formal contracts (e.g., employment
contracts, buyer-​supplier contracts, and loan agreements).
Building on the idea of the fiduciary, discussed in the context of governance
by Blair and Stout (2001), we propose that whereas the relationship between
the organization and its management is contractual, the relationship with
those in charge of oversight—​most notably, the board of directors—​should
be thought of as fiduciary. Blair and Stout (2001, 404) aptly described the
fiduciary relationship as “other-​regarding” and “mediating.” Consequently,
a person with a fiduciary role should neither be a stakeholder nor represent
one, instead, those in a fiduciary role are “charged with the task of balancing
the sometimes-​conflicting claims and interests of the many different groups
that bear residual risk and have residual claims on the firm” (Blair and Stout
2001, 404). This quote succinctly captures the essence of the other-​regarding
role of the fiduciary.
It is in our view misleading to think of the board member’s relationship
with the organization in contractual terms, because the essence of the fidu-
ciary relationship is not about negotiable rights and responsibilities. The fidu-
ciary role is more appropriately thought of as a duty toward the organization.
Further, as Blair and Stout (2001, 406), and many others, have remarked, de-
spite common belief, there is nothing in corporate law that compels the fidu-
ciary to serve just one constituency, such as the shareholder:

Despite the emphasis legal theorists have given shareholder primacy in re-
cent years, corporate law itself does not obligate directors to do what the
shareholders tell them to do. Nor does corporate law compel the board to
maximize share value. To the contrary, directors of public corporations
enjoy a remarkable degree of freedom from shareholder command and
control. Similarly, the law grants directors wide discretion to consider the
interests of other corporate participants in their decision making—​even
when this adversely affects the value of the stockholders’ shares.

The beneficiary of the director’s fiduciary duty is not any specific stakeholder
group but the entire organization. Indeed, one of the important functions of
 Glossary of Terms  249

corporate law is to shield directors from shareholder control (Blair and Stout
2001, 406).

Forbearance (and Quasijudicial Functions)

Consider the case of a contract dispute between a buyer and a supplier. If

the two are separate firms not belonging to the same corporate parent, they
can as a last resort refer to the courts to settle the dispute through a court
ruling.
However, if the two are under unified ownership, reference to the courts
is not an option, because an organization would be effectively suing itself.
Instead, the organization must address the dispute as a matter of private
ordering. Within the organization, “contract law . . . is that of forbearance,
according to which internal organization is its own court of ultimate appeal”
(Williamson 1996, 378 [emphasis added]). How an organization structures
its internal principles and processes of forbearance is a foundational govern-
ance decision. To this end, the internal organization has a number of impor-
tant quasijudicial functions (Williamson 1975, 30).

Fundamental Transformation

Consider the situation in which a firm seeks competitive bids from a large
pool of candidate suppliers for a standard auditing service. As the buyer is
considering its options, it need not pay attention to the identities of indi-
vidual suppliers, because all suppliers are ex ante substitutable. Suppose the
buyer then chooses one of the candidates from the pool and enters into a
buyer-​supplier contract, which is renegotiated and renewed every twelve
months.
At contract renewal, is the buyer going to face a pool of substitutable
suppliers? This may no longer be the case because in the preceding twelve
months, the buyer has learned to collaborate with a specific auditor and to
adapt to unforeseen circumstances. Perhaps the buyer has also discovered
that the chosen auditor reliably delivers its services, thereby signaling a pos-
itive reputation effect. Instead of facing an ex post pool of equally attractive
suppliers, one stands out from the rest. A fundamental transformation has
occurred.
250  Glossary of Terms

Williamson (1985, 13) noted that the fundamental transformation has

“pervasive importance for the study of economic organization.” Whereas
at the outset the buyer was dependent on the pool of suppliers only in the
aggregate sense (it needed someone to audit its books), it has now become
dependent on a specific supplier in the sense that it is economically more ef-
ficient to continue with the same supplier. Symmetrically, the supplier now
prefers the specific buyer. As a result, the contractual relationship has been
infused with bilateral dependency. There is indeed something fundamental
about this transformation: Bilateral dependency exposes both contracting
parties to risk, and consequently, both parties should be incentivized to de-
sign the requisite safeguards to protect the relationship.
The fundamental transformation often occurs inadvertently over time as
the contracting parties find more efficient ways of collaborating during con-
tract duration. In this sense, the fundamental transformation is an emergent
(as opposed to deliberately designed) property of a contractual relationship.
The deliberate design decisions pertain to the kinds of safeguards that are
implemented to address the risks that arise from bilateral dependency.

Governance

We define governance as the totality of the deliberate choices the designer

makes about management, oversight, and risk. To this end, governance can
be circumscribed by three questions:

1. Who in the organization is trusted to make the most important

decisions about how activities are organized, how resources are
allocated, and how performance is evaluated (management)?
2. What general guidelines and principles govern decision-​making to
ensure that decisions are made in the best interest of the organization
(oversight)?
3. What safeguards are in place to ensure the cooperation of the
organization’s most important constituencies, particularly those who
have voluntarily put something at stake in the organization (risk)?

These three questions must be addressed as a going concern, not merely at

the founding of the organization. In fact, an essential aspect of governance
is how management, oversight, and risk evolve over time. For example, a
 Glossary of Terms  251

crucial governance question in a growing corporation heading toward an ini-

tial public offering is how and when to separate oversight from management.
Our approach to governance embraces a predominantly private-​ordering
perspective. However, we readily acknowledge that those who view govern-
ance from an external point of view would define it in a way that emphasizes
compliance, law, regulation, and policy. An illustrative example of an exter-
nally oriented approach can be found in OECD’s definition of corporate gov-
ernance, which emphasizes the role of the legal, regulatory, and institutional
environments. But this is not a sign of inconsistency, it merely means that
OECD’s definition of governance serves a different purpose than ours.

Hazard

Consider an undesirable event such as a car accident. To safeguard yourself

against the risk of a total economic loss of your vehicle, you contract with an
insurance company for comprehensive collision coverage. At the same time,
it should be obvious that the insurance policy in and of itself does nothing
to help you avoid the hazard of a collision. Acknowledging this, you choose
to mitigate the hazard by always obeying the speed limits, not operating
your smartphone while driving, never driving while intoxicated, and so on.
The measures we take to manage risk differ from those we take to mitigate
hazards.
As an example of a hazard in contracting, consider the oil refining value
chain (Klein, Crawford, and Alchian 1978). Suppose an oil company owns
and operates the oil fields and the refineries, both of which are subject to con-
siderable site specificity and physical asset specificity. Suppose further that
an oil pipeline is the only way of transporting the oil from the fields to the
refineries. The oil company is well advised to own and operate the pipelines
as well. If it did not, it would expose its (highly specific) assets to an economic
holdup problem (Goldberg 1976) by the company that owns and operates the
pipelines.
Just like texting while driving constitutes a hazard, so does exposing one’s
organization to the holdup problem. Klein et al. (1978, 310) elaborate:

[The] specialized producing and refining assets are therefore “hostage” to

the pipeline owner. At the “gathering end” of the pipeline, the monopsonist
pipeline could and would purchase all its oil at the same well-​head price
252  Glossary of Terms

regardless of the distance of the well from the refinery. This price could be
as low as the marginal cost of getting oil out of the ground (or its reser-
vation value for future use, if higher) and might not generate a return to
the oil-​well owner sufficient to recoup the initial investment of exploration
and drilling. At the delivery-​to-​refinery end of the pipeline, the pipeline
owner would be able to appropriate [a significant portion of the profits] of
the refineries. The pipeline owner could simply raise the price of crude oil
at least to the price of alternative sources of supply to each refinery that are
specialized to the pipeline.

A straightforward way to avoid the economic holdup problem is common

ownership of the oil fields, the pipelines, and the refineries; in other words,
vertical integration would effectively attenuate the contracting hazard. At the
same time, vertical integration would not eliminate various risks due to high
levels of asset specificity, which would call for additional safeguards.
The economic holdup problem is a special case of the more general no-
tion of contractual hazards, which arise from inadequate safeguarding
of the relationship. Inadequate safeguarding may pose a hazard to one
of the contracting parties, or both, but it always poses a hazard to the
relationship.

Incentive Intensity

When an employee is incentivized by an hourly rate or a fixed salary, the

amount of compensation does not depend on the amount of effort exerted or
the amount of output produced. A fixed salary is an example of low incentive
intensity (or low-​powered incentive). In piece rate, in contrast, the worker’s
hourly pay depends on the amount of output produced; piece rate is an ex-
ample of high incentive intensity (or a high-​powered incentive).
The appropriate level of incentive intensity in an exchange relationship
should be determined through a comparative analysis by asking, “What level
of incentive intensity makes the contractual relationship comparatively ef-
ficient?” As a general rule, high-​powered incentives are preferred over low-​
powered ones, because higher incentive intensity tends to foster efficiency
by mitigating free riding, among other positive effects. In situations in which
the individual’s work effort and performance are salient, high-​powered
incentives are preferable; employees can voluntarily exert as much effort as
 Glossary of Terms  253

they choose and be compensated accordingly. A self-​employed taxi driver

is a good example of a high-​powered incentive: The driver is incentivized to
take on as many customers as possible during a work shift. High-​powered
incentives tend to work well also when not only effort and performance but
also the way effort links to performance is salient.
However, there are contexts in which comparatively low-​ powered
incentives may be a better alternative. The eight-​hour shift of a police of-
ficer is a good example. If police officers were incentivized by the number of
arrests they make or the number of speeding tickets they write during a work
shift, the predictable outcome is that even the most insignificant and incon-
sequential infractions (such as driving just over the speed limit on a deserted
highway) would be penalized. In the case of law enforcement, a sufficiently
high base salary that increases with years of service is a better alternative.
Linking salary increases to years of service is beneficial, because the public
benefits from having more experienced law enforcement officers. However,
because of the low-​powered incentive, the police officer cannot make more
money simply by “working harder” (however defined).
Incentive intensity can also be applied at the level of entire organizations,
or organizational subunits. For example, one decision industrial firms must
make regarding their production units is whether to assign them profit-​
and-​loss (P&L) or simply cost responsibility—​the former constitutes a com-
paratively high-​powered incentive. But assigning a production unit P&L
responsibility makes sense only if both the costs and the revenues of the unit
are salient. Costs tend to be salient, but revenue is unambiguous only if a
market price can be credibly established. This is not the case in many indus-
trial firms where production units sell their outputs at some arbitrary transfer
price to an internal sales unit. Arbitrary transfer prices make the production
unit’s revenue, and consequently its profit, arbitrary as well. Arbitrary profit-
ability calculations may be useful for accounting and tax purposes, but they
likely lack credibility as a basis for compensation. It is generally ill advised to
hold organizational subunits or their managers responsible for something
over which they have little control. Williamson (1994, 102) elaborates: “High-​
powered incentives obtain if a party has a clear entitlement to and can estab-
lish the magnitude of its net receipts easily. Lower-​powered incentives obtain
if the net receipts are pooled and/​or if the magnitude is difficult to ascertain.”
That many industrial firms assign only cost responsibility to their production
units stems precisely from the fact that operations create value jointly with
other business functions.
254  Glossary of Terms

Note that incentive intensity is not about the level of compensation but,
rather, about the extent to which compensation is affected by the amount of
output produced. The stronger this link, the higher the incentive intensity,
and the stronger the effect of increased effort on compensation.

Institution

Colloquially, we think of institutions in terms of large, important organi-

zations such as the University of Illinois, the United States Department of
State, or the World Bank. In the context of organization design and gov-
ernance, we adopt a more abstract, broader definition offered by North
(1991, 97): “Institutions are the humanly devised constraints that structure
political, economic and social interaction. They consist of both informal
constraints (sanctions, taboos, customs, traditions, and codes of conduct),
and formal rules (constitutions, laws, property rights).” North (1991, 97) fur-
ther described the purpose of institutions in a useful way: “Throughout his-
tory, institutions have been devised by human beings to create order and
reduce uncertainty in exchange.” The enforcement of institutions can simi-
larly occur through both formal (e.g., law enforcement) and informal (e.g.,
codes of conduct) means (Granovetter 1985).
The University of Illinois, the United States Department of State, and the
World Bank continue to be institutions under North’s definition as well; how-
ever, under the broader definition we would also classify contracts, invoices,
and warranty clauses as institutions, as they “create order and reduce uncer-
tainty” in contractual relationships.
An institution can also assume the form of an abstract principle. For ex-
ample, the principle of contra proferentem in dispute resolution maintains
that if there is an ambiguous term in the contract that can be interpreted
in multiple ways, the courts should apply the principle of “construing am-
biguous language against the drafter [of the contract]” (Boardman 2006,
1108). The objective of contra proferentem is “to give drafters an incentive
to draft cleanly” (Boardman 2006, 1108). Contra proferentem is sometimes
described as a doctrine, which suggests it has indeed acquired the status of an
institution.
It is useful to think of institutions generally as “the rules of the game” that
are enforced by some (formal or informal) authority (Alston et al. 2018, 3).
In the context of organization design and governance, institutions pertain to
 Glossary of Terms  255

the rules that govern how the work is done, who exercises oversight, and how
risk is governed. The institutional environment of an organization is defined
as the collection of all the central “rules of the game” that affect the economic
actions within and across organizations.
Some institutions are a matter of law, policy, and regulation, but the ones
particularly relevant to the designer are the ones associated with private
ordering; that is where the designer can seek organizational efficiency by
making choices among feasible alternatives.

Main Problem

Should the CEO also chair the board of directors in a limited liability com-
pany? Before this question can be addressed, the designer must formulate the
problem that board composition is aimed to address. We call this formula-
tion the main problem. In all organization design and governance decisions,
the designer should always start by specifying the main problem. If there
are multiple problems, the designer should seek to prioritize them, and the
problem with top priority becomes the main problem.
In the board composition example, the main problem is often formulated
as one of agency: Does the CEO (the agent) act in the best interest of the or-
ganization (the principal)? If the main problem board composition is aimed
at addressing is the agency problem, then the separation of the CEO and the
chairperson roles is recommended.
In contrast, if the main problem is defined as the ability to make strategic
decisions quickly in a rapidly changing environment, then having the same
person be in charge of the top management team and the board of directors
may be comparatively efficient. However, if the CEO also chairs the board,
additional safeguards may be needed to ensure one person does not acquire
too much power in the organization and create a potential entrenchment
hazard.
The prescription of formulating the main problem does not imply that the
designer should formulate only one problem; the prescription is to have the
designer think about the design problems in an analytical, prioritizing, and
discriminating way. As the COVID-​19 vaccine example in c­ hapter 4 seeks
to establish, the designer’s task is impossible without clear definition and
prioritization of problems. If all problems and all stakeholders are impor-
tant, then nothing and no one is important. Overpermissive and all-​inclusive
256  Glossary of Terms

formulations of governance problems are in our view one of the most chal-
lenging obstacles to the designer, which is why specifying the main problem
is essential.

Management

For the purposes of organization design and governance, management refers

to all those aspects of the organization that are responsible for ensuring that
the work gets done: resource allocation, design of the central value-​creating
processes, control, coordination, and reward systems.
Management may or may not involve organizational members whose
title is that of a manager. A case in point, in self-​managing organizations
(SMOs), work planning and execution are allocated directly to employees.
Furthermore, one need not venture into SMOs to find situations in which
organizational members are empowered to take on managerial tasks, for ex-
ample, by being included in the planning of their own work; employee em-
powerment was one of the cornerstones of the Total Quality Management
movement in the 1990s (Cole and Scott 2000).
Value-​adding activities in organizations involve a combination of humans
and technology. We also include in management all those nonhuman as-
sets that are used to coordinate activities. In our view, whereas thinking of
employer-​employee and buyer-​supplier relationships in contracting terms
is salient, we suggest that similar contractual thinking be extended to tech-
nology. For example, whether an industrial firm finances production equip-
ment through debt or equity is a matter of governance, with important
contractual implications (Williamson 1988).

Myopia

In ophthalmology, myopia (or nearsightedness) refers to a condition

in which objects are seen distinctly only when they are near to the eye.
Analogously, myopia in governance decisions describes a situation in
which the contracting parties consider primarily the short-​term efficiency
implications. Note that being myopic is not about the inability to predict the
future; it is about the inability, or perhaps more accurately reluctance, to ad-
dress governance decisions as long-​term issues.
 Glossary of Terms  257

Myopia links directly to efficiency. Although not all attempts at being effi-
cient in the short term are myopic, focusing on the short-​term implications
in governance and contracting runs the risk of being myopic, particularly if
the quest for short-​term gains jeopardizes credibility.
Consider the example of a large buyer that purchases standard components
from a smaller component supplier. There are many alternative suppliers
from which the buyer can choose, and consequently, purchasing managers
at the buyer firm may feel tempted to “squeeze” the supplier to improve its
own productivity and profitability. This may be ultimately myopic due to the
fundamental transformation. A forward-​looking buyer would seek ways to
design the relationship such that the supplier has a high-​powered incentive
to learn and develop its skills.

Organization

Universities, firms, and legislatures are organizations. Indeed, an obvious

way of thinking about organization is to think about legal entities called
organizations.
In this book, we adopt a broader definition. Whereas entities called or-
ganizations are relevant to our exposition, they are not the essence of or-
ganization. From the point of view of organization design and governance,
the question is less about how entities called organizations are designed and
more about how value-​creating cooperative relationships are organized. In
this book, the word organization refers more broadly to the principles and
practices of organizing contractual relationships.
Consider the conventional buyer-​supplier contract. Although the buyer
and the supplier are obviously relevant contracting entities, we propose that
the essence of organizing resides not in the entities but in their relationship.
Therefore, it makes little sense to consider the characteristics of one entity
without simultaneously incorporating both the characteristics of the other
entity and the interactions the two entities have with one another. When the
pioneer organization economist John R. Commons (1934) noted that the
basic unit of analysis should be the transaction, he specifically instructed us
to look beyond the entities into the structure of the relationship in a way that
infuses order, mitigates conflict, and ultimately achieves mutual gain.
Emphasizing the relational aspects of organizing usefully turns attention
to the ways in which contracting parties are dependent on one another and
258  Glossary of Terms

complement one another. Particularly relevant for designers are situations

of bilateral dependency, that is, when both contracting parties have a
vested interested in maintaining the relationship. Understanding bilateral
dependency paves the way to understanding how stakeholder relationships
emerge.
As a final example of thinking of organization not in terms of entities
but sets of contractual relationships, consider antitakeover provisions.
Most importantly, it is not the entity (the firm) that requires protec-
tion. Instead, the designer’s attention turns to the specific stakeholder
relationships that may (or may not) require protection. In contemplating
antitakeover provisions, designers commonly direct attention to the rela-
tionship that shareholders have with the organization. Consequently, the
analysis of antitakeover provisions focuses on their potential impact on
shareholder interests, and protection is prescribed if it increases share-
holder bargaining power.

Oversight

All organizations must ensure that the decisions its members make serve
the best interest not of those who make the decisions but of the entire or-
ganization. One of the central tasks of oversight is to ensure such alignment;
ensuring that the organization satisfies the demands of the institutional envi-
ronment is another. We use the general label oversight to refer collectively to
all the individuals, groups, structures, and principles that work toward these
objectives.
An obvious entity contributing to oversight is a board of some kind: board
of directors, board of trustees, board of regents, and so on. However, there are
also many entities external to the organization that have central roles in over-
sight. For example, stock exchanges, government agencies, and legislatures
exercise either direct or indirect oversight over publicly traded corporations.
These entities are sources of various compliance requirements imposed on
the organization.
Various aspects of oversight can also be embedded in the organization’s
founding documents, such as the articles of incorporation, corporate bylaws,
and shareholders’ agreements. These documents tend to focus more on how
the organization is governed than on how it is managed.
 Glossary of Terms  259

Ownership

The owner of a house is the person whose name appears on the property
deed. More generally, we tend to think of ownership of an asset in terms of
who has title. However, in the context of governance, we propose that the
notion of ownership be viewed differently. We define ownership through
control rights, and in particular, residual rights of control. Ownership also
includes the right to transfer an asset (e.g., Barzel 1989), but transfer is less
relevant for our exposition.
In a contractual relationship, the owner of an asset has two prerogatives
with relevant implications. Specifically, the owner is entitled (1) to exercise
control over the decisions that are not specified in contracts (residual con-
trol), and (2) to the economic surplus the organization generates (residual
claimancy). Both residual control and residual claimancy are at the heart
of ownership. Since these two rights may not go hand in hand (Hart 1989,
1766), both require the designer’s attention.

Private Ordering (vs. Legal Centralism)

Consider the way in which two contracting parties settle disputes. Will they
try to work things out themselves or will a third party be involved? Private
ordering refers to contractual arrangements whereby the contracting parties
seek to settle disputes as private matters, without involving external third
parties. There is no clear-​cut definition for what constitutes a third party but,
to be sure, one of the parties suing the other would clearly involve a third
party (the courts).
Dispute resolution offers a salient example of private ordering. More
generally, Williamson (1985, xii) noted that “the governance of contractual
relationships is primarily effected through the institutions of private ordering
rather than through legal centralism.” Indeed, private ordering (relying on
the contractual pillar) becomes salient as we contrast it with jurisprudence
(relying on the institutional pillar). Consistent with Williamson, governance
questions in this book are approached primarily from the private-​ordering
perspective where the focus is on the “self-​created mechanisms” (Williamson
1996, 378) by which exchange parties structure and manage their
relationships.
260  Glossary of Terms

Probity (and Probity Hazard, Sovereign Transaction)

In some exchange relationships, it is imperative that all transactions be exe-

cuted with utmost integrity or probity. Of course, one might argue that pro-
bity is important in all contractual relationships that involve any degree of
risk. However, there are contexts in which its absence is particularly discon-
certing. For example, it is one thing for a supplier to act in a self-​serving and
opportunistic way in its relationship with a buyer; a government agency or a
branch of the military doing the same in its relationship with the president
is quite another. In some contexts, probity must be placed at the very core of
the relationship.
Because probity links directly to risk, it is often discussed in conjunc-
tion with the probity hazard. Probity is further relevant in contexts such
as foreign policy, which belongs to the general category of sovereign
transactions: “[T]‌here are certain tasks that we expect government to per-
form [ . . . ] because it alone embodies the public’s authority. Certain tasks are
sovereign tasks” (Wilson 1989, 359). But as Williamson argued, even sover-
eign transactions may benefit from an efficiency analysis, most notably, “a
governance structure that supports a presumption of or predisposition to-
ward cooperativeness will relieve the hazards of probity” (Williamson 1999,
324). Williamson (1999, 324) further maintained that “the potential cost sav-
ings that would accrue to high-​powered incentives in foreign affairs are not
great.”
The concept of probity is useful in establishing important boundary
conditions for efficiency. Unless the organization has not resolutely es-
tablished probity, conducting efficiency analyses is both premature
and misguided. This is in our view best described by political scientist
James Q. Wilson, whose position is aptly summarized by Williamson
(1999, 310):

While Wilson invites the application of transaction cost economics to pol-

itics, he also cautions that ‘Careful attention to transaction costs will not
alone determine where [the] line should be drawn’ (1989, 359). Not only is
the output of government ‘complex and often controversial’ (Wilson 1989,
348), but agencies often have ‘multiple objectives, government programs
have distributional effects, and considerations of equity and accountability
are often important’. (Wilson 1989, 348)
 Glossary of Terms  261

As an example of the interplay between probity and efficiency, we might

ask whether it makes sense to direct attention to inferior prison food quality
(indeed an efficiency-​related issue, see Hart 2003) if there are more funda-
mental concerns, such as violation of the inmates’ constitutional rights, other
legal rights, and even basic human rights. Should not all attention first be
directed at establishing the integrity of the prison organization, and only sub-
sequently engage the designer in a discussion of organizational efficiency?

Profit-​seeking vs. Nonprofit Organizations

The conventional definition of a for-​profit organization as one that seeks an

economic surplus is in our view not useful for governance. The reason is that
it is common for nonprofits to produce an economic surplus and have net
worth; universities and many cooperatives are good examples. What is rel-
evant is whether the organization has a stakeholder that can claim the sur-
plus, that is, whether there is a residual claimant. Accordingly, we make the
distinction by invoking residual claimancy: A for-​profit organization has a
residual claimant that is entitled to the surplus, but a nonprofit organization
has no other residual claimant than the organization itself (Fama and Jensen
1983a).
Whether the organization has a residual claimant or not is relevant to
governance because it has fundamental implications for how a possible sur-
plus is governed. Nonprofit organizations must ensure that the surplus is not
expropriated through private benefits. In fact, a tax-​exempt nonprofit organ-
ization may lose its tax-​exempt status for the fiscal year if tax authorities dis-
cover its surplus has been privately expropriated.

Public vs. Private Organizations

In contrast with the for-​profit/​nonprofit distinction where the two categories

describe actual organizations, the public/​private distinction tends to define
the ends of a continuum rather than describe actual organizations; many
if not most public organizations are more accurately described as public-​
private partnerships. Consider the university as an example. The university
may be considered a public organization but, at the same time, it has scores
of private actors embedded within it. For example, security services, cafeteria
262  Glossary of Terms

services, and custodial services are often provided by private organizations.

Similarly, many commercial airlines in Europe in particular are at least
partly state owned. For example, the Republic of Finland (represented by the
Finnish Prime Minister’s Office) owns 56 percent of the shares in Finnair, the
flag carrier and largest airline in Finland. Although some organizations are
fully private and others fully public, there are numerous hybrid forms.
Because there are many hybrid forms, it is more useful to examine the
public/​private distinction by reference to the internal workings of the organ-
ization: Which parts of management should be populated by civil servants
and which parts should be contracted out to private entities? How should
compensation be structured in employment contracts? If there is a public
interest in the organization, what kinds of oversight roles can be delegated to
private actors? Can representatives of private interests occupy fiduciary roles
in the first place? What kind of risk will the public (e.g., taxpayers) bear? How
does the designer implement safeguards to protect the public interest against
expropriation by private actors?
The designer’s task is to make informed choices regarding the role of
public and private actors in the organization. This task does not in our view
involve the categorization of the organization as either public or private,
which is why most discussions on the privatization of public organizations
remain elusive. We must start by specifying what exactly is being privatized,
why, and how.

Rationality (and Bounded Rationality, Self-​Interest vs.

Opportunism)

In the general sense, any goal-​oriented behavior is rational: If one wants to

attain x and y is a means to it, then it is rational to do y (this is known as the
practical syllogism, see von Wright 1963). Note that this definition implies
that one cannot speak about rationality without first specifying an objec-
tive. For the purposes of this book, we assume the designer seeks efficiency.
However, the general objective of efficiency is too general to be actionable.
For an objective to become actionable, the designer must be able to justify
a direct link from the objective to alternative courses of action (March and
Simon 1993, 177). This is why formulating a more specific main problem is
required. In the context of efficient governance, designers are rational when
their choices address the main problem in a comparatively efficient way.
 Glossary of Terms  263

Conventional economic theory assumes decision makers are rational in

the sense that they choose the best option from all possible options; indeed,
rationality implies optimization. In this book, we subscribe to the more real-
istic notion of being able to choose the comparatively efficient option from
the known alternatives. We also acknowledge that the designer’s rationality
is bounded. Simon (1997, 88) aptly described human behavior as “intendedly
rational, but only limitedly so.”
That behavior is intendedly rational is crucial because it embraces the
premise that the designer is acting in the best interest of the organization.
Consequently, we offer no encouragement to designers who engage in self-​
serving behaviors and pursue their own agendas at the expense of organi-
zational efficiency. In fact, curbing individual rationality that is aimed at
goals inconsistent with those of the organization is one of the central tasks
of oversight.
Seeking what is best for organizational efficiency is built on the general
idea of self-​interest. The assumption (and prescription) is that designers
behave in ways that are beneficial for their organizations. However, the as-
sumption (and prescription) is also that self-​interest is sought by adhering to
the constraints and expectations of the institutional environment.
There are also stronger forms of self-​ interest such as opportunism,
described by Williamson (1996, 378) as “[s]‌elf-​interest seeking with guile, to
include calculated efforts to mislead, deceive, obfuscate, and otherwise con-
fuse.” Just like we have little to say to those engaging in self-​serving behaviors,
our message to those inclined to act opportunistically is that our objective is
to help designers make the lives of the dishonest maximally uncomfortable.
To this end, we seek to help those who act in good faith safeguard their con-
tractual relationships.
A bounded-​rationality approach toward an uncertain future maintains
that although the future can never be fully or even adequately predicted,
designers are capable of making decisions in a forward-​looking manner that
incorporates conscious foresight. Importantly, therefore, bounded rationality
does not imply myopia. Throughout our exposition, we want to emphasize
the premise that no matter how boundedly rational behavior is, rationality
always remains the intention: “Parties to a contract who look ahead, recog-
nize potential hazards, work out the contractual ramifications, and fold these
into the ex ante contractual agreement obviously enjoy advantages over those
who are myopic or take their chances and knock on wood” (Williamson
2000, 601).
264  Glossary of Terms

Residual (and Residual Claimant, Residual Interest)

Let us start from the top line of a firm’s income statement and work our
way to the bottom line. Once all the requisite appropriations and contrac-
tual obligations—​salaries, invoices, interest, taxes, depreciation—​have
been subtracted from revenue, there may be something left over. This
economic surplus or net income is the residual. The rules that govern
the management and the distribution of the residual are central to
governance.
In most jurisdictions, shareholders are the only stakeholder whose rights
to the residual have been secured in law. In short, shareholders “have a legal
claim on the firm’s net receipts” (Klein et al. 2012, 311 [emphasis added]).
A relevant question in for-​profit organizations is whether constituencies
other than shareholders can reasonably claim residual interest.
The general position taken in this book is that shareholders may not
be the only stakeholders with a justifiable residual interest. For example,
suppose a group of employees in a high-​technology firm commits their
time and their effort to developing a set of skills that not only create con-
siderable value for the firm but are also firm-​specific in the sense that they
will be less useful if the employees leave the organization. Commitment to
such specificity is a form of investment risk that creates bilateral depend-
ency between the firm and the employee group in a way that is analogous
to the bilateral dependency between the firm and its shareholders. In fact,
one might argue that the dependency is even stronger in the case of the
employees, because they cannot sell their investments at market value the
same way shareholders of public corporations can. Consequently, it would
be reasonable to conclude that the employees committing to specificity
are a stakeholder with a legitimate (but not legal) residual interest (Klein
et al. 2012).
Economic surplus is a salient manifestation of the residual. There is, how-
ever, another type of residual that also merits attention in a contractual rela-
tionship. Since all complex contracts are unavoidably incomplete, there are
many issues that remain unspecified. Insofar as these issues involve the use
of assets and the allocation of resources, decisions regarding these unspec-
ified issues are the prerogative of the owner of the asset. In the economics
literature, these prerogatives are discussed under the rubric of residual rights
of control. In the case of economic surplus, we simply speak of rights to the
residual.
 Glossary of Terms  265

Risk (and Switching Cost)

An industrial firm that performs the final assembly of a product such as an

automobile or a smartphone does not have the capability to produce many
of the components needed in the final assembly—​the firm needs suppliers.
Similarly, the scale of many firms is not sufficiently high to justify in-​house
legal counsel; therefore, they must contract with external law firms. All or-
ganizations are, in one way or another, dependent on other organizations.
As long as a contracting party depends on other parties only in the ag-
gregate sense, the risks in contracting are both negligible and remediable;
a bad experience with a supplier can be remedied by switching to another.
Risk enters when switching costs are no longer negligible and an organiza-
tion becomes dependent on a specific organization. Single-​sourcing is a good
example. HP is dependent on Canon for laser printer cartridges and on Intel
and AMD for processors. The reason HP cannot buy processors from a fully
competitive market is because the market structure of processors is oligopo-
listic, that is, dominated by a small number of sellers.
Risk must be considered separately for unilateral and bilateral depend-
ency. When dependency is unilateral, as in the case of a large powerful
buyer and a smaller captive supplier, the exchange relationship is subject
to the holdup problem (Goldberg 1976). To be sure, as consumers we can
all think of situations in which we have become dependent on a specific
supplier, but the supplier is in no appreciable way dependent on us; instead,
they are only dependent on the customer base in the aggregate sense. Due
to this asymmetric dependency, the supplier may be able to “hold us up” as
customers. Being exposed to a holdup problem is a risk under unilateral
dependency.
Under bilateral dependency, there really is no holdup problem. Instead, the
relevant risk has to do with the fact that bilaterally dependent relationships
tend to be more efficient because they are characterized by various forms of
specificity that lead to higher switching costs. Investments in specificity are
often justified due to the efficiency gains they bestow upon the relationship.
In car manufacturing, for example, component suppliers sometimes physi-
cally collocate their component plants with the car manufacturer’s final as-
sembly plant. When the two plants are physically collocated, the supplier is
able to deliver its components to the final assembler both “just in time” (ex-
actly when they are needed) and “in sequence” (in the order in which they
are required in the final assembly). In-​sequence delivery is relevant when the
266  Glossary of Terms

components are parts of customizable configurations (e.g., seat materials and

colors).
It is easy to imagine the astronomical switching costs associated with a
car manufacturer terminating a contract with a seat supplier that operates a
massive seat plant right next to the car manufacturer’s final assembly plant.
In the case of unilateral dependency, unilateral risk stems from the potential
holdup problem on the side of the potential hostage; in the case of bilateral
dependency, switching costs give rise to risk on both sides.

Safeguard

When contracting parties become bilaterally dependent, they expose them-

selves to potential exchange hazards and risk. Premature and unexpected
termination of the contract would have severe economic consequences for
both parties. Consequently, the parties are well advised to think of both ex
ante and ex post ways to safeguard the contractual relationship. An obvious
safeguard is a carefully crafted long-​term contract that secures the rights
and the responsibilities of both transacting parties. Or, in the case of one
party supplying the other components that can only be used in the buyer’s
products, potential order cancellations due to unpredictable demand must
be safeguarded; the supplier having to bear the full cost of demand unpre-
dictability is unlikely to constitute a mutually credible alternative.
Here, we emphasize reciprocal acts and the mutual interest of safeguarding
the relationship. In short, the unit of analysis is the relationship, not the
contracting parties, let alone just one of them.

Specificity (and Temporality, Dedicated Assets)

Consider a situation in which a productive asset such as a piece of produc-

tion equipment or an employee skill is being applied in a way that generates
the highest amount of value. Then imagine that the asset can no longer be
applied in this best use. What is the difference in the value of the asset in its
best versus next-​best use? The higher the difference, the higher the (asset)
specificity. Highly specific assets have low redeployability.
Williamson (1996) presented five different types of specificity. First is
site specificity, which arises from the geographic location of the asset. For
 Glossary of Terms  267

example, a coal mining plant may be collocated with an electricity plant

(Joskow 1988). Once sited, the coal mining and the electricity-​producing as-
sets are for all practical purposes immobile. Furthermore, in the case one was
relocated, the other would likely have to relocate as well.
Second is physical asset specificity, in which one or several transacting
parties make investments in equipment that involves design characteristics
specific to the transaction. Because the technology acquires relation-​specific
characteristics, it may have lower economic value in alternative uses. For ex-
ample, a supplier in the automobile industry may invest in customer-​specific
dies to stamp components (Klein, Crawford, and Alchian 1978).
Third is human capital specificity, in which investments in relationship-​
specific human capital often arise through various learning-​ by-​
doing
processes (Harris and Helfat 1997). For example, software programmers may
be required to commit their time to developing a company-​specific proprie-
tary programming language.
Fourth is temporal specificity that arises from interdependencies over time
(Masten 1984). For example, because produce must be handled in a timely
manner in the supply chain, it may be inefficient to run a highly fragmented
supply chain that involves multiple organizations; instead, vertical inte-
gration of the different supply chain stages may be required as a safeguard
(Bucheli, Mahoney, and Vaaler 2010).
Finally, dedicated assets are found in contexts where a supplier makes
investments in production capacity with the aim of selling a substantial
volume of output to a particular customer (Williamson 1985, 194). Even if
these assets did not exhibit site or physical asset specificity, they would still
constitute a relation-​specific investment. If the contract in which the dedi-
cated assets are involved were terminated prematurely, the supplier would be
left with substantial excess capacity, which would be problematic even in the
absence of physical asset specificity.

Stakeholder (and Stakeholder Analysis)

There are many constituencies that are, in one way or another, involved in
the activities of an organization: employees, customers, suppliers, financiers,
local communities, and the society more generally. But there is a specific
category of constituencies who have, by virtue of becoming involved in the
organization, made a wager of some kind on the organization’s outcomes
268  Glossary of Terms

(Orts and Strudler 2002). We reserve the label stakeholder to refer to these
constituencies. We further define stakeholder in symmetrical terms: Two
parties are one another’s stakeholders if they have responsibilities toward one
another and are interested in one another’s success.
Providing equity financing to a firm by purchasing shares is a conspicuous
example of a wager; therefore, shareholders are the obvious stakeholder in
a limited liability company. However, there are also other, less conspicuous
wagers. In general, any commitments that constituencies make to the spe-
cific organization constitute wagers that may merit the designer’s attention.
Employees who commit to the development of the organization’s distinct core
competences are effectively making a wager on the outcome that these core
competences will be valuable in the future. If the core competences lose their
value, so do employee commitments to specificity. Therefore, employees who
have committed to specificity can be argued to have a legitimate residual in-
terest comparable to that of the shareholders. Consequently, those employees
who commit to specificity may require not only a salary in exchange for their
time and their efforts but also a return on their investments to specificity.
One governance option is to turn employees formally into residual claimants
by means of an employee stock ownership program. Some corporations may
go even further and offer stock options to all employees, not just executives
(Oyer and Schaefer 2005).
In general, any constituency that has a legitimate residual interest in
the organization should be considered a stakeholder. However, because
not all wagers are the same, stakeholder status subscribes to degrees. The
objective of a stakeholder analysis is to determine which constituencies
have more at stake than others, and subsequently, devise the requisite gov-
ernance structures that safeguard the cooperation of those with the most
at stake.
The ultimate objective of a stakeholder analysis is to ensure that the
wagers the organization needs are actually made. For example, if the
organization’s success hinges on its ability to attract equity financing, gov-
ernance structures must be designed in a way that ensures that investors
place their bets; if the organization depends heavily on organization-​
specific innovation and R&D, it must contract with its employees in a way
that ensures sufficient commitments to specificity; and so on. If stakeholder
governance fails and the requisite wagers are not made, the organization
faces an underinvestment problem, which immediately jeopardizes the
organization’s viability.
 Glossary of Terms  269

Transaction Cost

Transaction costs include all ex ante (searching, drafting, and negotiating)

and ex post (addressing unanticipated disturbances) costs associated with the
execution of a contract. In simple transactions, transaction costs are negli-
gible; buying a carton of milk from the grocery store is a representative ex-
ample. In more complex relationships, transaction costs can be so substantial
that they merit the designer’s attention; a parts manufacturer that produces
make-​and model-​specific components to the final assembler of automobiles
is a representative example.
Whereas both ex ante and ex post costs are relevant, the latter tend to be
more significant and merit more attention from the designer. Because com-
plex contracts are often unavoidably incomplete, long-​term relationships
tend to involve the need for adaptation due to unanticipated disturbances
that the ex ante contract did not, or simply could not, cover. Not being able to
address such disturbances can be the source of significant ex post transaction
costs. In general, the magnitude of transaction costs in exchange relationships
tends to be associated with three main factors: (1) specificity, (2) frequency
of transacting, and (3) uncertainty. Transaction costs tend to be higher at
higher degrees of specificity, frequency, and uncertainty (Williamson 1985).
The relationship between transaction costs and governance decisions
is, however, not straightforward. For example, even though increasing fre-
quency of transacting does increase transaction costs, it also justifies the de-
sign of specialized governance structures that may in fact result in net gains
even if the exchange relationship remained interorganizational.
The association between specificity and governance choice is more
straightforward in that increasing specificity tends to associate with the
choice of intraorganizational transacting. But even here, the designer must
resist the temptation of following simple rules. One reason is that depending
on the context, it may well be the case that specificity is not a given but a
decision variable. For example, an industrial firm may be able to choose
between using general-​purpose and special-​purpose technology in its pro-
duction. If the firm wants to avoid carrying production equipment on its
balance sheet, it may choose to contract with an external supplier that uses
general-​purpose equipment that can be readily modified by proper tooling
to serve the buying firm. This is an important reminder that transaction costs
being associated with specificity does not necessarily mean they are driven
by specificity. Specificity may constitute a decision variable, not something
270  Glossary of Terms

the designer must take as a given (Riordan and Williamson 1985). As always,
the designer’s task is to engage in an analysis of the alternative governance
options in the specific exchange context.

Value Creation vs. Value Capture (and Preappropriation vs.

Postappropriation)

An automobile without defects is more valuable than one with defects; a fully
recovered patient is more valuable than one who must be readmitted to the
hospital; a one-​time prison inmate is economically more valuable than a ca-
reer criminal.
Value manifests itself in numerous ways in organizations and in the society
more broadly. Some forms of value are unambiguous and can be assigned
a monetary value, others are more elusive. But the idea that hospitals and
prisons create value just like an automobile assembly plant does should not
be too controversial. Further, the notion of more valuable in the context of
prisons or hospitals does not require that we assess value in monetary terms.
That lowering the number of readmissions in hospitals and recidivism rates
in prisons are sources of value is salient without assigning monetary values
to readmission and recidivism. Since we address efficiency in comparative
terms, there is no need for absolute measurement of value. Instead, if there
are two known, feasible alternatives one of which creates comparatively less
waste than the other, then that option should be preferred no matter what ab-
solute level of waste it generates.
A fundamental realization about efficient organization is that one can
pursue efficiency without actually having to measure it in the conventional
economic sense.
Who benefits from defect-​ free automobiles, recovered patients,
and a released prisoner who never returns to prison? Who ultimately
appropriates the value that is created is obviously an important question,
but we propose it is in fact not central to efficient organization. The only
assumption required is that value is distributed in a way that maintains the
credibility of the organization in the eyes of those on whom it depends for
value creation. But beyond this assumption, nothing else needs to be said
or assumed about appropriation or value capture. Comparative efficiency
analysis operates largely on the preappropriation, value-​creation side of the
organization.
 Glossary of Terms  271

A focus on value creation means that in the context of for-​ profit

corporations, profit is less relevant than value added—​profit is a residual and
as such, a postappropriation measure of performance. As we proceed from
the top line to the bottom line of an income statement, profit is what remains
after all the appropriations have been made. Powerful stakeholders may be in
a position to appropriate disproportionate amounts of value. Consequently,
what remains at the end is more a reflection of the power dynamics of the
appropriation process than efficient organization. Therefore, a corpora-
tion need not be profitable to be efficient, and efficiency does not ensure
profitability.
Coff (1999, 120) offers two illustrative examples of value creation and ap-
propriation not going hand in hand. After World War II,

U.S. auto manufacturers clearly were generating [earnings] before in-

tense foreign competition changed the nature of the game. Yet unions and
management, as opposed to shareholders, appropriated a sizable portion
of the [earnings]. Similarly, IBM assembled the strategic capabilities that
built most of the modern personal computer industry. However, Intel
and Microsoft were ultimately able to appropriate much of the associated
[earnings].

These two examples aptly show that value appropriation occurs both within
and across organizations. This book is not about how appropriation occurs.

Viability

An organization is viable when it has secured all the inputs it needs as well as
the means by which the inputs are converted into outputs.
Some inputs are generic and can be purchased from input markets at negli-
gible risk. As far as viability is concerned, the organization must secure suffi-
cient financing to obtain these market inputs. Other inputs and, in particular,
means of conversion are not readily available in the market; instead, they in-
volve various degrees and kinds of specificity. In industrial production, for
example, some parts and components required for final assembly must be
engineered to customer specifications. Similarly, some parts of the produc-
tion process can use off-​the-​shelf, general-​purpose technologies, but others
require the design of special-​purpose equipment with low redeployability.
272  Glossary of Terms

Any input, be it a component, conversion technology, or an employee skill

designed for the specific organization, involves a wager of some kind. The
organization’s viability hinges on ensuring that all the requisite wagers
are made.
Wagers come in many forms. Providers of equity financing make their
wagers as they purchase shares in the company; inability to convince the
investors to place their bets leads to underinvestment in equity, which
jeopardizes viability in situations in which there are no alternatives to eq-
uity financing. Some employees make their wagers when they commit their
time and their effort to organization-​specific innovation and R&D; failure
to convince employees to commit to such specificity leads to underinvest-
ment in innovation and R&D, which may have devastating consequences
for a growing high-​technology firm. Finally, some suppliers make their
wagers when they become dependent on the buyer through relation-​specific
investments; failure to convince the supplier to make such investments may
jeopardize viability in contexts such as the automotive supply chain, where
specialized suppliers are crucial for efficiency. The common denominator in
all these commitments is that they are wagers on the outcome of the organiza-
tion (Orts and Strudler 2002).
We propose that the central objective of governance is to establish the vi-
ability of the organization by ensuring that all the requisite wagers are made.
To this end, the designer must analyze the type and the magnitude of all
wagers made by the organization’s constituencies, and subsequently, safe-
guard the central relationships by the appropriate governance structures.
A systematic stakeholder analysis is aimed at achieving the objective of un-
derstanding all the wagers and their governance implications.
 References

Adler, Paul S., and Bryan Borys. 1996. “Two Types of Bureaucracy: Enabling and
Coercive.” Administrative Science Quarterly 41, no. 1: 61–​89.
Albert, Stuart, and David A. Whetten. 1985. “Organizational Identity.” Research in
Organizational Behavior 7: 263–​95.
Alchian, Armen A., and Harold Demsetz. 1972. “Production, Information Costs, and
Economic Organization.” American Economic Review 62, no. 5: 777–​95.
Alchian, Armen A., and Susan Woodward. 1988. “The Firm Is Dead; Long Live the
Firm: A Review of Oliver E. Williamson’s The Economic Institutions of Capitalism.”
Journal of Economic Literature 26, no. 1: 65–​79.
Alonso, Ricardo, Wouter Dessein, and Niko Matouschek. 2008. “When Does
Coordination Require Centralization?” American Economic Review 98, no. 1: 145–​79.
Alston, Eric, Lee J. Alston, Bernardo Mueller, and Tomas Nonnenmacher.
2018. Institutional and Organizational Analysis: Concepts and Applications.
Cambridge: Cambridge University Press.
Argyres, Nicholas, and Kyle J. Mayer. 2007. “Contract Design as a Firm Capability: An
Integration of Learning and Transaction Cost Perspectives.” Academy of Management
Review 32, no. 4: 1060–​77.
Argyres, Nicholas S., and Brian S. Silverman. 2004. “R&D, Organization Structure, and
the Development of Corporate Technological Knowledge.” Strategic Management
Journal 25, no. 8–​9: 929–​58.
Balakrishnan, Srinivasan, and Birger Wernerfelt. 1986. “Technical Change, Competition
and Vertical Integration.” Strategic Management Journal 7, no. 4: 347–​59.
Barnard, Chester I. 1938. The Functions of the Executive. Cambridge, MA: Harvard
University Press.
Barney, Jay B. 1991. “Firm Resources and Sustained Competitive Advantage.” Journal of
Management 17, no. 1: 99–​120.
Barzel, Yoram 1989. Economic Analysis of Property Rights. Cambridge: Cambridge
University Press.
Bebchuk, Lucian A. 2003. “Why Firms Adopt Antitakeover Arrangements.” University of
Pennsylvania Law Review 152, no. 2: 713–​53.
Bebchuk, Lucian A., John C. Coates, and Guhan Subramanian. 2002. “The Powerful
Antitakeover Force of Staggered Boards: Theory, Evidence, and Policy.” Stanford Law
Review 54, no. 5: 887–​951.
Bebchuk, Lucian A., and Alma Cohen. 2003. “Firm’s Decisions Where to Incorporate.”
Journal of Law & Economics 46, no. 2: 383–​425.
Bebchuk, Lucian A., and Roberto Tallarita. 2020. “The Illusory Promise of Stakeholder
Governance.” Cornell Law Review 106: 91–​177.
Berle, Adolf A., and Gardiner C. Means. 1932. The Modern Corporation and Private
Property. New York: Macmillan.
274 References

Bernstein, Ethan, and Nitin Nohria. 2016. Note on Organizational Structure (9-​491-​083).
Cambridge, MA: Harvard Business School.
Bhardwaj, Akhil, and Mikko Ketokivi. 2021. “Bilateral Dependency and Supplier
Performance Ambiguity in Supply Chain Contracting: Evidence from the Railroad
Industry.” Journal of Operations Management 67, no. 1: 49–​70.
Bhardwaj, Akhil, and Anastasia Sergeeva. Forthcoming. “Values-​ Based Trust as a
Shift Parameter for Collective Organizing: The Case of Magnum Photos.” Journal of
Management Studies.
Blair, Margaret M., and Lynn A. Stout. 1999. “A Team Production Theory of Corporate
Law.” Virginia Law Review 85, no. 2: 247–​328.
Blair, Margaret M., and Lynn A. Stout. 2001. “Director Accountability and the Mediating
Role of the Corporate Board.” Washington University Law Review 79, no. 2: 403–​49.
Boardman, Michelle E. 2006. “Contra Proferentem: The Allure of Ambiguous Boilerplate.”
Michigan Law Review 104, no. 5: 1105–​28.
Broom, Leonard, and Philip Selznick. 1955. Sociology: A Text with Adapted Readings.
New York: Row, Peterson.
Brownlee, Hunter J. 1994. “The Shareholders’ Agreement: A Contractual Alternative to
Oppression as a Ground for Dissolution.” Stetson Law Review 24, no. 2: 267–​310.
Bucheli, Marcelo, Joseph T. Mahoney, and Paul M. Vaaler. 2010. “Chandler’s Living
History: The Visible Hand of Vertical Integration in Nineteenth Century America
Viewed under a Twenty-​First Century Transaction Costs Economics Lens.” Journal of
Management Studies 47, no. 5: 859–​83.
Burns, Tom, and George M. Stalker. 1961. The Management of Innovation.
London: Tavistock.
Caterpillar. 2021. Proxy Statement. Retrieved from investors.caterpillar.com/​financials/​
sec-​filings.
CDW Corporation. 2019. Form 10-​K Annual Report. Retrieved from sec.gov/​edgar.shtml.
Chandler, Alfred D. 1962. Strategy and Structure: Chapters in the History of the American
Industrial Enterprise. Cambridge, MA: MIT Press.
Chandler, Alfred D. 1977. The Visible Hand: The Managerial Revolution in American
Business. Cambridge, MA: Harvard University Press.
Chandler, Alfred D. 1990. Scale and Scope: The Dynamics of Industrial Capitalism.
Cambridge, MA: Belknap Press.
Clark, Robert C. 1985. “Agency Costs versus Fiduciary Duties.” In Principals and
Agents: The Structure of Business, edited by John W. Pratt and Richard J. Zeckhauser,
55–​80. Boston, MA: Harvard Business School Press.
Clegg, Stewart R., David Courpasson, and Nelson Phillips. 2006. Power in Organizations.
Thousand Oaks, CA: SAGE.
Coase, Ronald H. 1937. “The Nature of the Firm.” Economica 4, no. 16: 386–​405.
Coff, Russell W. 1999. “When Competitive Advantage Doesn’t Lead to Performance: The
Resource-​Based View and Stakeholder Bargaining Power.” Organization Science 10, no.
2: 119–​33.
Colburn, Kimberly. 2007. The Art of Artistic Direction. Eugene, OR: University of Oregon.
Cole, Robert E., and W. Richard Scott. 2000. The Quality Movement & Organization
Theory. Thousand Oaks, CA: SAGE.
Commons, John R. 1934. Institutional Economics. Madison, WI: University of
Wisconsin Press.
 References  275

Cremers, K. J. Martijn, Vinay B. Nair, and Urs Peyer. 2008. “Takeover Defenses and
Competition: The Role of Stakeholders.” Journal of Empirical Legal Studies 5, no.
4: 791–​818.
Cuypers, Ilya R. P., Jean-​François Hennart, Brian S. Silverman, and Gokhan Ertug. 2021.
“Transaction Cost Theory: Past Progress, Current Challenges, and Suggestions for the
Future.” Academy of Management Annals 15, no. 1: 111–​50.
Cyert, Richard M., and James G. March. [1963] 1992. A Behavioral Theory of the Firm. 2nd
ed. Englewood Cliffs, NJ: Prentice-​Hall.
David, Paul A. 1985. “Clio and the Economics of QWERTY.” American Economic Review
75, no. 2: 332–​37.
Davidson, Carl. 1988. “Multiunit Bargaining in Oligopolistic Industries.” Journal of Labor
Economics 6, no. 3: 397–​422.
DeAngelo, Harry, and Edward M. Rice. 1983. “Antitakeover Charter Amendments and
Stockholder Wealth.” Journal of Financial Economics 11, no. 1: 329–​59.
Dearborn, DeWitt C., and Herbert A. Simon. 1958. “Selective Perception: A Note on the
Departmental Identification of the Executive.” Sociometry 21, no. 2: 140–​44.
Dey, Aiyesha, and Joshua T. White. 2021. “Labor Mobility and Antitakeover Provisions.”
Journal of Accounting and Economics 71, no. 2: Article 101388.
DiIulio, John J. 1987. Governing Prisons: A Comparative Study of Correctional
Management. New York: Free Press.
Dodd Jr, E. Merrick. 1932. “For Whom Are Corporate Managers Trustees?” Harvard Law
Review 45, no. 7: 1145–​63.
Drucker, Peter F. 1973. “On Managing the Public Service Institution.” The Public Interest
33, no. 1: 43–​60.
Ebrahim, Alnoor, Julie Battilana, and Johanna Mair. 2014. “The Governance of Social
Enterprises: Mission Drift and Accountability Challenges in Hybrid Organizations.”
Research in Organizational Behavior 34, no. 1: 81–​100.
Eccles, Robert G., and Harrison C. White. 1988. “Price and Authority in Inter-​Profit
Center Transactions.” American Journal of Sociology 94, Suppl.: S17–​S51.
Fama, Eugene F., and Michael C. Jensen. 1983a. “Separation of Ownership and Control.”
Journal of Law and Economics 26, no. 2: 301–​25.
Fama, Eugene F., and Michael C. Jensen. 1983b. “Agency Problems and Residual Claims.”
Journal of Law and Economics 26, no. 2: 327–​49.
Feyerabend, Paul. 1995. Killing Time: The Autobiography of Paul Feyerabend. Chicago,
IL: University of Chicago Press.
Fisch, Jill E. 2018. “Governance by Contract: The Implications for Corporate Bylaws.”
California Law Review 106, no. 2: 373–​409.
Fiss, Owen M. 1984. “Against Settlement.” Yale Law Journal 93, no. 6: 1073–​90.
Fuqua, Don. 1986. Investigation of the Challenger Accident: Report of the Committee on
Science and Technology, House of Representatives, Ninety-​ Ninth Congress, Second
Session. Washington, DC: US Government Printing Office.
Galbraith, Jay R. 1971. “Matrix Organization Designs.” Business Horizons 14, no. 1: 29–​40.
Galbraith, Jay R. 2002a. Designing Organizations: An Executive Guide to Strategy,
Structure, and Process. Rev. ed. San Francisco, CA: Jossey-​Bass.
Galbraith, Jay R. 2002b. “Organizing to Deliver Solutions.” Organizational Dynamics 31,
no. 2: 194–​207.
Galbraith, Jay R. 2009. Designing Matrix Organizations That Really Work. San Francisco,
CA: Jossey-​Bass.
276 References

Galbraith, Jay R. 2014. Designing Organizations: Strategy, Structure, and Process at the
Business Unit and Enterprise Levels. 3rd ed. San Francisco, CA: Jossey-​Bass.
Galli, Jaime D. 2011. Organizational Management in the Non-​ Profit Performing
Arts: Exploring New Models of Structure, Management, and Leadership. Eugene,
OR: University of Oregon.
Gamble, John, Margaret Peteraf, and Arthur Thompson. 2021. Essentials of Strategic
Management: The Quest for Competitive Advantage. 7th ed. New York: McGraw Hill.
Ghinger, John J. 1974. “Shareholders’ Agreements for Closely Held Corporations: Special
Tools for Special Circumstances.” University of Baltimore Law Review 4, no. 2: 211–​44.
Gioia, Dennis A., Shubha D. Patvardhan, Aimee L. Hamilton, and Kevin G. Corley. 2013.
“Organizational Identity Formation and Change.” Academy of Management Annals 7,
no. 1: 123–​92.
Glaeser, Edward L., and Andrei Shleifer. 2001. “Not-​for-​Profit Entrepreneurs.” Journal of
Public Economics 81, no. 1: 99–​115.
Goldberg, Victor P. 1976. “Regulation and Administered Contracts.” Bell Journal of
Economics 7, no. 2: 426–​48.
Gomery, Douglas. 2005. The Hollywood Studio System: A History. London: British Film
Institute.
Goold, Michael, and Andrew Campbell. 2002a. Designing Effective Organizations: How to
Create Structured Networks. San Francisco, CA: Jossey-​Bass.
Goold, Michael, and Andrew Campbell. 2002b. “Do You Have a Well-​ Designed
Organization?” Harvard Business Review 80, no. 2: 117–​24.
Granovetter, Mark. 1985. “Economic Action and Social Structure: The Problem of
Embeddedness.” American Journal of Sociology 91, no. 3: 481–​510.
Greenwood, Royston, Christine Oliver, Kerstin Sahlin, and Roy Suddaby, eds. 2008. The
SAGE Handbook of Organizational Institutionalism. London: SAGE.
Gross, Malvern J., John H. McCarthy, and Nancy E. Shelmon. 2005. Financial and
Accounting Guide for Not-​for-​Profit Organizations. New York: Wiley.
Grossman, Sanford J., and Oliver D. Hart. 1980. “Takeover Bids, the Free-​Rider Problem,
and the Theory of the Corporation.” Bell Journal of Economics 11, no. 1: 42–​69.
Grossman, Sanford J., and Oliver D. Hart. 1986. “The Costs and Benefits of
Ownership: A Theory of Vertical and Lateral Integration.” Journal of Political Economy
94, no. 4: 691–​719.
Gulati, Ranjay. 2010. Cisco Business Councils (2007): Unifying a Functional Enterprise with
an Internal Governance System (Harvard Business School Case 9-​409-​062). Cambridge,
MA: Harvard Business School.
Gulati, Ranjay, and Harbir Singh. 1998. “The Architecture of Cooperation: Managing
Coordination Costs and Appropriation Concerns in Strategic Alliances.” Administrative
Science Quarterly 43, no. 4: 781–​814.
Hamel, Gary. 2011. “First, Let’s Fire All the Managers.” Harvard Business Review 89, no.
12: 48–​60.
Hansmann, Henry B. 1980. “The Role of Nonprofit Enterprise.” Yale Law Journal 89, no.
5: 835–​901.
Hansmann, Henry B. 1987. “The Effect of Tax Exemption and Other Factors on the Market
Share of Nonprofit versus For-​Profit Firms.” National Tax Journal 40, no. 1: 71–​82.
Harris, Dawn, and Constance Helfat. 1997. “Specificity in CEO Human Capital and
Compensation.” Strategic Management Journal 18, no. 11: 895–​920.
 References  277

Hart, Oliver. 1989. “An Economist’s Perspective on the Theory of the Firm.” Columbia Law
Review 89, no. 7: 1757–​74.
Hart, Oliver. 2003. “Incomplete Contracts and Public Ownership: Remarks, and
an Application to Public-​ Private Partnerships.” Economic Journal 113, March
issue: C69–​C76.
Hart, Oliver, Andrei Shleifer, and Robert W. Vishny. 1997. “The Proper Scope of
Government: Theory and an Application to Prisons.” Quarterly Journal of Economics
112, no. 4: 1127–​61.
Hatch, Mary Jo, and Majken Schultz, eds. 2004. Organizational Identity: A Reader.
Oxford: Oxford University Press.
Hayek, Friedrich A. 1945. “The Use of Knowledge in Society.” American Economic Review
35, no. 4: 519–​30.
Hirshleifer, David, and Sheridan Titman. 1992. “Share Tendering Strategies and the
Success of Hostile Takeover Bids.” Journal of Political Economy 98, no. 2: 295–​324.
Holmström, Bengt R., and Jean Tirole. 1989. “The Theory of the Firm.” In Handbook
of Industrial Organization, edited by R. Schmalensee and R. D. Willig, 61–​133.
Amsterdam: North Holland.
HP. 2019. Sustainable Impact Report. Retrieved from investor.hp.com/​governance/​sus-
tainability/​.
HP. 2020. Form 10-​K Annual Report. Retrieved from investor.hp.com/​financials/​sec-​
filings/​.
Hughes, Everett C. 1939. “Institutions.” In An Outline of the Principles of Sociology, edited
by Robert E. Park, 283–​346. New York: Barnes and Noble.
Illinois Department of Public Health. 2021. COVID-​19 Vaccination Plan. Retrieved from
https://​dph.illin​ois.gov/​covi​d19/​vacc​inat​ion-​plan.html.
Jenoff, Pam. 2012. “Going Native: Incentive, Identity, and the Inherent Ethical Problem of
In-​House Counsel.” West Virginia Law Review 114, no. 2: 725–​57.
Jensen, Michael C. 1986. “Agency Costs of Free Cash Flow, Corporate Finance, and
Takeovers.” American Economic Review 76, no. 2: 323–​29.
Jensen, Michael C. 2001. “Value Maximization, Stakeholder Theory, and the Corporate
Objective Function.” Journal of Applied Corporate Finance 14, no. 3: 8–​21.
Jensen, Michael C. 2002. “Value Maximization, Stakeholder Theory, and the Corporate
Objective Function.” Business Ethics Quarterly 12, no. 2: 235–​56.
Jensen, Michael C., and William H. Meckling. 1976. “Theory of the Firm: Managerial
Behavior, Agency Costs and Ownership Structure.” Journal of Financial Economics 3,
no. 4: 305–​60.
Jensen, Michael C., and Richard S. Ruback. 1983. “The Market for Corporate Control: The
Scientific Evidence.” Journal of Financial Economics 11, no. 1: 5–​50.
Johnson, William C., Jonathan M. Karpoff, and Sangho Yi. 2015. “The Bonding
Hypothesis of Takeover Defenses: Evidence from IPO Firms.” Journal of Financial
Economics 117, no. 2: 307–​32.
Joskow, Paul L. 1988. “Asset Specificity and the Structure of Vertical Relationships.”
Journal of Law, Economics & Organization 4, no. 1: 95–​117.
Kahneman, Daniel. 2011. Thinking, Fast and Slow. New York: Farrar, Straus and Giroux.
Kahneman, Daniel, Paul Slovic, and Amos Tversky. 1982. Judgment under
Uncertainty: Heuristics and Biases. Cambridge: Cambridge University Press.
278 References

Kaplan, Robert S., and David P. Norton. 2008. The Execution Premium: Linking Strategy
to Operations for Competitive Advantage. Cambridge, MA: Harvard Business
School Press.
Keski-​Valkama, Alice, Eila Sailas, Markku Eronen, Anna-​ Maija Koivisto, Jouko
Lönnqvist, and Riittakerttu Kaltiala-​ Heino. 2007. “A 15-​ Year National Follow-​
Up: Legislation Is Not Enough to Reduce the Use of Seclusion and Restraint.” Social
Psychiatry and Psychiatric Epidemiology 42, no. 9: 747–​52.
Kesner, Idalene F., and Roy B. Johnson. 1990. “An Investigation of the Relationship be-
tween Board Composition and Stockholder Suits.” Strategic Management Journal 11,
no. 4: 327–​36.
Ketokivi, Mikko, and Xavier Castañer. 2004. “Strategic Planning as an Integrative Device.”
Administrative Science Quarterly 49, no. 3: 337–​65.
Ketokivi, Mikko, and Joseph T. Mahoney. 2017. “Transaction Cost Economics as a Theory
of the Firm, Management, and Governance.” Oxford Research Encyclopedia of Business
and Management, edited by Ramon J. Aldag. Oxford: Oxford University Press.
Ketokivi, Mikko, and Joseph T. Mahoney. 2020. “Transaction Cost Economics as a Theory
of Supply Chain Efficiency.” Production and Operations Management 29, no. 4: 1011–​31.
Kirkpatrick, David, and Tyler Maroney. 1998. “The Second Coming of Apple.” Fortune
138, no. 9: 86–​92.
Klein, Benjamin, Robert G. Crawford, and Armen A. Alchian. 1978. “Vertical Integration,
Appropriable Rents, and the Competitive Contracting Process.” Journal of Law and
Economics 21, no. 2: 297–​326.
Klein, Peter G., Joseph T. Mahoney, Anita M. McGahan, and Christos N. Pitelis. 2012.
“Who Is in Charge? A Property Rights Perspective on Stakeholder Governance.”
Strategic Organization 10, no. 3: 304–​15.
Knight, Frank H. 1941. “Anthropology and Economics.” Journal of Political Economy 49,
no. 2: 247–​68.
Larcker, David, and Brian Tayan. 2015. Corporate Governance Matters: A Closer Look
at Organizational Choices and Their Consequences. 2nd ed. Old Tappan, NJ: Pearson
Education.
Leavitt, Harold J. 2005. Top Down: Why Hierarchies Are Here to Stay and How to Manage
Them Effectively. Boston, MA: Harvard Business School Press.
Lee, Michael Y., and Amy C. Edmondson. 2017. “Self-​Managing Organizations: Exploring
the Limits of Less-​Hierarchical Organizing.” Research in Organizational Behavior 37,
no. 1: 35–​58.
Leslie, Christopher R. 2014. “Antitrust Made (Too) Simple.” Antitrust Law Journal 79, no.
3: 917–​40.
Levitt, Barbara, and James G. March. 1988. “Organizational Learning.” Annual Review of
Sociology 14, no. 1: 319–​40.
Liebowitz, Stan J., and Stephen E. Margolis. 1990. “The Fable of the Keys.” Journal of Law
& Economics 33, no. 1: 1–​25.
Liu, Tingting, and J. Harold Mulherin. 2019. “How Has Takeover Competition Changed
over Time?” Journal of Applied Corporate Finance 31, no. 1: 81–​94.
Lowe, Katherine E., Joe Zein, Umur Hatipoğlu, and Amy Attaway. 2021. “Association
of Smoking and Cumulative Pack-​Year Exposure with COVID-​19 Outcomes in the
Cleveland Clinic COVID-​19 Registry.” JAMA Internal Medicine 181, no. 5: 709–​11.
Mace, Myles. 1971. Directors: Myth and Reality. Cambridge, MA: Harvard University Press.
 References  279

Macher, Jeffrey T., and Barak D. Richman. 2008. “Transaction Cost Economics: An
Assessment of Empirical Research in the Social Sciences.” Business and Politics 10, no.
1: 1–​63.
Mahoney, Joseph T. 2002. “The Relevance of Chester I. Barnard’s Teachings to
Contemporary Management Education: Communicating the Aesthetics of
Management.” International Journal of Organization Theory and Behavior 5, no.
1: 159–​72.
Manne, Henry G. 1965. “Mergers and the Market for Corporate Control.” Journal of
Political Economy 73, no. 2: 110–​20.
March, James G., and Herbert A. Simon. 1993. Organizations. 2nd ed. New York: Wiley.
Masten, Scott E. 1984. “The Organization of Production: Evidence from the Aerospace
Industry.” Journal of Law and Economics 27, no. 2: 403–​17.
Masten, Scott E., and Edward A. Snyder. 1993. “United States versus United Shoe Machinery
Corporation: On the Merits.” Journal of Law & Economics 36, no. 1: 33–​70.
McNamara, Gerry, Paul M. Vaaler, and Cynthia Devers. 2003. “Same as It Ever Was: The
Search for Evidence of Increasing Hypercompetition.” Strategic Management Journal
24, no. 3: 261–​78.
Meyer, John W., and Brian Rowan. 1977. “Institutionalized Organizations: Formal
Structure as Myth and Ceremony.” American Journal of Sociology 83, no. 2: 340–​63.
Milgrom, Paul, and John Roberts. 1990. “The Economics of Modern
Manufacturing: Technology, Strategy, and Organization.” American Economic Review
80, no. 3: 511–​28.
Milgrom, Paul, and John Roberts. 1992. Economics, Organization & Management. Upper
Saddle River, NJ: Prentice-​Hall.
Mulherin, J. Harold, Jeffry M. Netter, and Annette B. Poulsen. 2017. “The Evidence on
Mergers and Acquisitions: A Historical and Modern Report.” In The Handbook of the
Economics of Corporate Governance, edited by Benjamin E. Hermalin and Michael S.
Weisbach, 235–​90. North-​Holland: Elsevier.
Neste. 2020. Annual Report. Retrieved from neste.com/​for-​media/​material/​
annual-​reports.
North, Douglass C. 1991. “Institutions.” Journal of Economic Perspectives 5, no. 1: 97–​112.
Organisation for Co-​operation and Economic Development. 2015. G20/​OECD Principles
of Corporate Governance. 3rd ed. Paris: OECD.
Orts, Eric W. 1992. “Beyond Shareholders: Interpreting Corporate Constituency Statutes.”
George Washington Law Review 61, no. 1: 14–​135.
Orts, Eric W., and Alan Strudler. 2002. “The Ethical and Environmental Limits of
Stakeholder Theory.” Business Ethics Quarterly 12, no. 2: 215–​33.
Oyer, Paul, and Scott Schaefer. 2005. “Why Do Some Firms Give Stock Options to All
Employees? An Empirical Examination of Alternative Theories.” Journal of Financial
Economics 76, no. 1: 99–​133.
Parmar, Bidhan, R. Edward Freeman, Jeffrey S. Harrison, Andrew C. Wicks, Lauren
Purnell, and Simone de Colle. 2010. “Stakeholder Theory: The State of the Art.”
Academy of Management Annals 4, no. 1: 403–​45.
Penrose, Edith T. 1959. The Theory of the Growth of the Firm. Oxford: Oxford
University Press.
Penrose, Edith T. 1995. The Theory of the Growth of the Firm. 3rd ed. Oxford: Oxford
University Press.
280 References

Pfeffer, Jeffrey. 1987. “A Resource Dependence Perspective on Intercorporate Relations.”

In Intercorporate Relations: The Structural Analysis of Business, edited by Mark Mizruchi
and Michael Schwartz, 25–​55. Cambridge: Cambridge University Press.
Pfeffer, Jeffrey, and Gerald R. Salancik. 1978. The External Control of Organizations: A
Resource Dependence Perspective. New York: Harper & Row.
Phillips, Robert A., and Joel Reichart. 2000. “The Environment as a Stakeholder?
A Fairness-​Based Approach.” Journal of Business Ethics 23, no. 2: 185–​97.
Pindur, Wolfgang, Sandra E. Rogers, and Pan Suk Kim. 1995. “The History of
Management: A Global Perspective.” Journal of Management History 1, no. 1: 59–​77.
Pollman, Elizabeth. 2019. “Startup Governance.” University of Pennsylvania Law Review
168, no. 1: 155–​221.
Porter, Michael E. 1985. Competitive Advantage: Creating and Sustaining Superior
Performance. New York: Free Press.
Powell, W. W., and P. J. DiMaggio, eds. 1991. The New Institutionalism in Organizational
Analysis. Chicago, IL: University of Chicago Press.
Prahalad, C. K., and G. Hamel. 1990. “The Core Competence of the Corporation.”
Harvard Business Review 68, no. 3: 79–​91.
Pratt, Katherine. 2000. “The Debt-​Equity Distinction in a Second-​Best World.” Vanderbilt
Law Review 53, no. 4: 1055–​158.
Pucker, Kenneth P. 2022. “The Myth of Sustainable Fashion.” Harvard Business Review.
An online essay retrieved from hbr.org/​2022/​01/​the-​myth-​of-​sustainable-​fashion.
Rawls, John. 1999. A Theory of Justice. Rev. ed. Cambridge, MA: Belknap Press.
Real Madrid. 2019. Corporate Social Responsibility and Sustainability Report. Retrieved
from realmadrid.com/​media/​document/​corporate-​social-​responsibility-​2019-​20.pdf.
Rechner, Paula L., and Dan R. Dalton. 1991. “CEO Duality and Organizational
Performance: A Longitudinal Analysis.” Strategic Management Journal 12, no.
2: 155–​60.
Rindfleisch, Aric, and Jan B. Heide. 1997. “Transaction Cost Analysis: Past, Present, and
Future Applications.” Journal of Marketing 61, no. 4: 30–​54.
Riordan, Michael H., and Oliver E. Williamson. 1985. “Asset Specificity and Economic
Organization.” International Journal of Industrial Organization 3, no. 4: 365–​78.
Rogers, William P., Neil A. Armstrong, David C. Acheson, Eugene E. Covert, Richard
P. Feynman, Robert B. Hotz, Donald J. Kutyna, Sally K. Ride, Robert W. Rummel,
Joseph F. Sutter, Arthur B. C. Walker Jr., Albert D. Wheelon, and Charles E. Yeager.
1986. Report of the Presidential Commission on the Space Shuttle Challenger Accident.
Washington, DC: US Government Printing Office.
Ross, Stephen A. 1973. “The Economic Theory of Agency: The Principal’s Problem.”
American Economic Review 63, no. 2: 134–​39.
Rousseau, Denise M., and Zipi Shperling. 2003. “Pieces of the Action: Ownership and
the Changing Employment Relationship.” Academy of Management Review 28, no.
4: 553–​70.
Rousseau, Denise M., and Zipi Shperling. 2004. “Ownership and the Changing
Employment Relationship: Why Stylized Notions of Labor No Longer Generally
Apply—​A Reply to Zardkoohi and Paetzold.” Academy of Management Review 29, no.
4: 562–​69.
Rubin, Paul H. 1978. “The Theory of the Firm and the Structure of the Franchise Contract.”
Journal of Law and Economics 21, no. 1: 223–​33.
 References  281

Ryan, Harley E., and Roy A. Wiggins. 2004. “Who Is in Whose Pocket? Director
Compensation, Board Independence, and Barriers to Effective Monitoring.” Journal of
Financial Economics 73, no. 3: 497–​524.
Santos, Filipe M., and Kathleen M. Eisenhardt. 2005. “Organizational Boundaries and
Theories of Organization.” Organization Science 16, no. 5: 491–​508.
Sawhney, Mohanbir, Robert C. Wolcott, and Inigo Arroniz. 2006. “The 12 Different Ways
for Companies to Innovate.” MIT Sloan Management Review 47, no. 3: 75–​81.
Selznick, Philip. 1957. Leadership in Administration. New York: Harper & Row.
Selznick, Philip. 1996. “Institutionalism ‘Old’ and ‘New’.” Administrative Science Quarterly
41, no. 2: 270–​77.
Shelanski, Howard A., and Peter G. Klein. 1995. “Empirical Research in Transaction Cost
Economics: A Review and Assessment.” Journal of Law, Economics, and Organization
11, no. 2: 335–​61.
Shleifer, Andrei, and Lawrence H. Summers. 1988. “Breach of Trust in Hostile Takeovers.”
In Corporate Takeovers: Causes and Consequences, edited by Alan J. Auerbach, 33–​56.
Chicago, IL: University of Chicago Press.
Silverman, Brian S. 2002. “Organizational economics.” In The Blackwell Companion to
Organizations, edited by Joel A. C. Baum, 467–​93. Oxford: Blackwell.
Simon, Herbert A. 1997. Administrative Behavior. 4th ed. New York: Macmillan.
Starik, Mark. 1995. “Should Trees Have Managerial Standing? Toward Stakeholder Status
for Non-​Human Nature.” Journal of Business Ethics 14, no. 3: 207–​17.
Stinchcombe, Arthur. 1965. “Social Structure and Organization.” In Handbook of
Organizations, edited by James March, 142–​93. Chicago, IL: Rand McNally.
Stipanowich, Thomas J. 2010. “Arbitration: The ‘new litigation’.” Illinois Law Review 10,
no. 1: 1–​60.
Sumner, D. A., and J. V. Balagtas. 2002. “An Overview of U.S. Dairy Policy.” In
Encyclopedia of Dairy Sciences, edited by H. Roginski, J. W. Fuquay and P. F. Fox.
London: Academic Press.
Sundaramurthy, Chamu, James M. Mahoney, and Joseph T. Mahoney. 1997. “Board
Structure, Antitakeover Provisions, and Stockholder Wealth.” Strategic Management
Journal 18, no. 3: 231–​45.
Swedberg, Richard. 1990. Economics and Sociology: Redefining Their Boundaries.
Princeton, NJ: Princeton University Press.
Thompson, James D. 1967. Organizations in Action: Social Science Bases of Administrative
Theory. New York: McGraw-​Hill.
Turbide, Johanne, and Claude Laurin. 2014. “Governance in the Arts and Culture
Nonprofit Sector: Vigilance or Indifference?” Administrative Sciences 4, no. 4: 413–​31.
UAW-​GM. 2015. Agreement between the UAW and GENERAL MOTORS LLC. Retrieved
from uaw.org/​uaw-​auto-​bargaining/​generalmotors/​.
University of Illinois. 2020. Annual Financial Report. Retrieved from obfs.uillinois.edu/​
about-​obfs/​annual-​reports/​.
Volkswagen Group. 2020. Annual Report. Retrieved from annualreport2020.
volkswagenag.com.
von Wright, Georg H. 1963. “Practical Inference.” Philosophical Review 72, no. 2: 159–​79.
Wasserman, Noam. 2012. The Founder’s Dilemmas. Princeton, NJ: Princeton
University Press.
Weick, Karl E. 1995. Sensemaking in Organizations. Thousand Oaks, CA: SAGE.
282 References

Weingast, Barry R., and William J. Marshall. 1988. “The Industrial Organization of
Congress; or, Why Legislatures, Like Firms, Are Not Organized as Markets.” Journal of
Political Economy 96, no. 1: 132–​63.
Wilkins, David B. 2012. “Is the In-​House Counsel Movement Going Global? A Preliminary
Assessment of the Role of Internal Counsel in Emerging Economies.” Wisconsin Law
Review 2012, no. 2: 251–​304.
Williamson, Oliver E. 1975. Markets and Hierarchies: Analysis and Antitrust Implications.
New York: Free Press.
Williamson, Oliver E. 1979. “Transaction-​ Cost Economics: The Governance of
Contractual Relations.” Journal of Law and Economics 22, no. 2: 233–​61.
Williamson, Oliver E. 1983. “Credible Commitments: Using Hostages to Support
Exchange.” American Economic Review 73, no. 4: 519–​40.
Williamson, Oliver E. 1985. The Economic Institutions of Capitalism: Firms, Markets,
Relational Contracting. New York: Free Press.
Williamson, Oliver E. 1988. “Corporate Finance and Corporate Governance.” Journal of
Finance 43, no. 3: 567–​91.
Williamson, Oliver E. 1991. “Comparative Economic Organization: The Analysis of
Discrete Structural Alternatives.” Administrative Science Quarterly 36, no. 2: 269–​96.
Williamson, Oliver E. 1994. “Transaction Cost Economics and Organization Theory.” In
The Handbook of Economic Sociology, edited by Neil J. Smelser and Richard Swedberg,
77–​107. Princeton, NJ: Princeton University Press.
Williamson, Oliver E. 1996. The Mechanisms of Governance. Oxford: Oxford
University Press.
Williamson, Oliver E. 1999. “Public and Private Bureaucracies: A Transaction Cost
Economics Perspective.” Journal of Law, Economics & Organization 15, no. 1: 306–​42.
Williamson, Oliver E. 2000. “The New Institutional Economics: Taking Stock, Looking
Ahead.” Journal of Economic Literature 38, no. 3: 595–​613.
Williamson, Oliver E. 2002a. “The Merger Guidelines of the U.S. Department of Justice: In
Perspective. Talk given at the 20th anniversary of the adoption of the 1982 Merger
Guidelines, June 20, 2002, The Antitrust Division of the US Department of Justice.
Retrieved from http://​www.just​ice.gov/​atr/​hmer​ger/​11257.htm.
Williamson, Oliver E. 2002b. “The Theory of the Firm as Governance Structure: From
Choice to Contract.” Journal of Economic Perspectives 16, no. 3: 171–​95.
Williamson, Oliver E. 2008. “Corporate Boards of Directors: In Principle and in Practice.”
Journal of Law, Economics, & Organization 24, no. 2: 247–​72.
Wilson, James Q. 1989. Bureaucracy. New York: Basic Books.
Wolf, W. B. 1973. Conversations with Chester I. Barnard (ILR Paperback Series No 12).
Ithaca, NY: Cornell University.
Worley, Christopher G., and Edward E. Lawler, III. 2006. “Designing Organizations That
Are Built to Change.” MIT Sloan Management Review 48, no. 1: 19–​23.
Wynn, Rolf. 2004. “Psychiatric Inpatients’ Experiences with Restraint.” Journal of Forensic
Psychiatry & Psychology 15, no. 1: 124–​44.
 Index

For the benefit of digital users, indexed terms that span two pages (e.g., 52–​53) may, on
occasion, appear on only one of those pages.
Tables and figures are indicated by t and f following the page number

adaptation, 12, 20–​21, 86, 141, 179–​80, asymmetric information, 4, 53–​54, 135,
187, 214, 224, 228, 230–​31 192, 195, 198, 228, 230–​31
agency
agent vs. principal, 14, 39–​40, 187–​90, bargaining, 96, 120
195–​96, 198, 208 codetermination, 120–​121
problem of, 14, 15, 17–​18, 39, 60, 95, collective, 121
144, 185, 195 multiunit, 120–​121, 126–​27
Alchian, Armen A., 40, 43–​44, 122–​23 plea, 184
alignment power, 156–​57, 199, 201, 201t, 203
adjustment, 161, 163–​64, 179–​ Barnard, Chester I., 33, 101–​2
81, 184–​86 inducements and contributions, 101–​3
and adjustment cost, 75 Bebchuk, Lucian A., 20–​21, 58, 101, 128,
discriminating alignment, 118–​20, 122, 200n.7, 203–​4, 203n.8
130, 166 Bhardwaj, Akhil, 12, 135–​36, 142–​43
maladaptation problems, 87 Blair, Margaret M., 22, 30, 59, 186, 196–​97,
misalignment, 39–​41 197n.6, 235
alliance, 89, 94–​95, 179–​80 board
collaborative contracting, 90t, 91t, 94–​ CEO duality vs. CEO
95, 179–​80 independence, 13–​14
joint equity, 90t, 91t, 94–​95, 142, 179–​80 chairperson of the, 13–​14, 38–​39, 48–​
analogy, 196–​97, 207, 233 49, 62–​63, 99, 127–​28, 165, 169, 194,
metaphor, 30, 31 204, 215
Apple, 61, 112n.5, 138, 190, 194n.5, 218 composition, 11, 14, 64, 91t, 128, 148,
appropriation, 133, 144–​45 169–​71, 170t, 190, 195
vs. expropriation, 144–​45 of directors, 11, 13–​17, 20, 28, 38–​39,
preappropriation, 17 41–​42, 58–​59, 62, 64, 90t, 91t, 94–​95,
postappropriation, 17, 208 99, 103, 109, 118, 120, 123–​28, 134,
value capture, 16 137, 139–​44, 148–​49, 164–​66, 169–​
Argyres, Nicholas S., 22, 32–​33 71, 170t, 173, 175–​78, 180, 183–​88,
arts organizations 190–​91, 193–​98, 199–​202, 201t, 203–​
artistic director, 142–​43, 146–​49 5, 208, 210, 213, 215
donors in, 134–​36, 142–​49 fiduciary and fiduciary duty, 41–​42,
managing (or executive) director, 142–​ 95, 125–​26, 128, 140, 148–​49, 170t,
43, 147–​48 173–​74, 183, 185–​86, 189, 193–​94,
theater, 38, 131, 142–​49 196–​97, 197n.6, 199, 202, 204, 210
284 Index

board (cont.) contracting, 6, 25, 37, 47, 57, 67, 70–​71,

of governors, 110 70t, 73, 75–​76, 78–​79, 84–​86, 88–​92,
independence, 14, 16–​17, 28, 58, 64, 90t, 91t, 93–​95, 98, 119, 120, 122–​23,
125–​28, 149, 186, 190, 190n.2, 213 126–​27, 130, 145–​46, 164–​65, 169,
of regents, 110, 258 179–​80, 182–​83, 186, 196, 208, 219–​
staggered, 203–​5 20, 226
of trustees, 110, 146, 147–​48, 258 duration, 44–​45, 119, 130
employment, 41–​42, 45–​49, 55, 72, 99,
Canon, 111, 115 102–​3, 114, 119–​20, 129, 139, 147–​
Caterpillar, 99, 141 48, 157, 179, 196, 202–​3, 213, 225–​26
Challenger space shuttle, 83–​84 enforceability, 96, 163–​64, 179, 180,
chief executive officer (CEO), 13–​15, 17–​ 182–​84, 186
18, 38–​39, 52, 59, 61–​63, 82–​83, 149, termination, 45, 49–​50, 54–​55, 79, 225
194, 194n.5, 204, 215–​16 Cook, Tim, 61, 194n.5
Cisco Systems, 34, 112n.5 core competence, 115–​16
Coase, Ronald H., 67, 70 cost center, 47, 222
codetermination. See bargaining cost of capital, 23–​24, 91t, 166
commitment COVID-​19, 9, 19, 49–​50, 104–​6, 154
mutual credible commitment, 112n.5, credibility. See commitment
118–​19, 205–​6, 207 Cuypers, Ilya, 77n.2, 141
compensation, 8–​9, 26, 38, 53, 58, 76, 114, Cyert, Richard M., 42–​43
119, 206
of board members, 126–​28, 141 deliberate. See also rationality
committee of the board of directors, governance as a set of deliberate
190, 190n.2 choices, 1, 11–​13, 64, 134, 135, 156,
equity-​based, 58, 126–​28, 210 177, 213–​14, 229
of executives, 145n.5, 229n.8 vs. emergent, 11–​12
fixed vs. residual payments, 53, 99, 102–​ Demsetz, Harold, 43–​44
4, 116, 123–​24, 130, 145n.5, 196, 233 dependency. See also interdependence
non-​equity-​based, 59, 126 asymmetric, 114
stock options, 41, 268 bilateral, 88, 110, 112, 119–​20, 145–​
composition fallacy, 130 46, 187–​88
constituency, 15–​16, 55n.6, 59, 65, 67, 95, unilateral, 18, 110
99–​100, 109–​18, 129–​30, 134, 136, dispute resolution, 21–​24, 40, 44, 69, 79–​
142–​43, 147–​49, 153–​57, 179, 187–​ 82, 90–​92, 164, 170t, 175, 181, 183–​
88, 196–​98, 200, 200n.7, 203, 206–​9, 84, 216, 228, 230–​31
215, 233 alternative dispute resolution (ADR),
vs. stakeholder, 17–​19, 25–​26, 28, 51–​ 80–​81, 183, 184, 216
58, 67, 100–​7, 118–​29, 143–​47 arbitration, 22, 79–​81, 170t,
Constitution of the United States, 127, 183 175, 183–​84
contract litigation, 21–​22, 60–​61, 79, 80–​82, 90–​
arm’s length, 26, 110, 122, 208–​9 92, 102, 175, 204, 216
buyer-​supplier, 11–​13, 18, 45, 72, 77, 84, mediation, 22, 79
87, 90t, 91t, 102, 110, 111n.4, 112n.5, quasijudicial function of the internal
114, 119, 121–​23, 196 organization, 22, 82, 228, 230–​31
complete vs. incomplete, 44–​45, 184 diversification, 55, 89, 90t, 91t, 93–​94,
contractibility, 27, 45, 124, 161, 163–​64, 187, 205
178–​81, 184–​86 donor, 134–​36, 142–​45, 147–​49
 Index  285

Eccles, Robert G., 25, 228 and net gains, 13, 27, 93–​94, 113, 118,
economies of scale, 46–​47, 84–​85, 131, 148, 151–​53, 151t, 174, 218–​20,
92, 166–​67 224, 234
economies of scope, 46, 91t, 92, 166–​67, and remediableness, 12, 78, 118, 218–​
199, 221t, 223 20, 231, 235
economies of specialization, 84–​85, fiduciary, 41–​42, 95, 125–​26, 128, 140,
221t, 221–​22 148–​49, 185–​86, 189, 193–​94, 196–​
Edmondson, Amy C., 35, 35n.1 97, 197n.6, 199, 202, 204, 210
efficiency the best interest of the organization,
as avoidance of waste, 1, 4, 6–​7, 8, 10, 14, 38–​42, 63, 65, 84, 125, 144, 172,
16, 17, 23–​24, 34–​35, 58, 74–​75, 122, 173, 199
128, 129, 233 duty of loyalty, 125, 170t, 173–​74, 183
comparative, 5, 6, 8, 11, 14, 22, 29, 33–​ loyalty vs. care, 173–​74
34, 36, 46–​47, 57, 67, 69, 70, 75, 79, Finnair, 131, 138–​41
84–​87, 87n.5, 93–​96, 125, 135–​36, Finnish Limited Liability Companies Act,
142, 148–​59, 207, 209n.10, 212, 214, 169, 170t, 170–​72, 174–​76, 178
218–​219, 219n.7, 221, 223–​24, 226, founding, 11, 20, 27, 158–​59, 161, 163–​78,
230–​31, 234–​35 180, 185–​86, 190–​91, 202, 216
efficiency distortion, 28, 47, 211, 219–​20 frequency
myopic vs. sustainable, 1, 15, 49, 150–​51 of transacting, 85–​86, 98
productivity, 10, 13, 84–​85, 91t, 92 fundamental transformation, 11–​12,
and slack, 42–​43 115n.6, 145, 187–​88
Efficiency Lens, 1–​2, 22–​25, 28, 30–​37, 38–​
51, 58–​64, 65, 70, 75–​77, 109, 139–​40, Galbraith, Jay R., 188n.1, 221–​23, 226–​27
143, 155–​56, 164–​65, 165f, 169, 185, General Motors, 81–​84, 121, 213–​14
187, 191, 191f, 204 Ghinger, John J., 180, 182–​83
Eisenhardt, Kathleen M., 71, 71n.1 Goldberg, Victor P., 114, 122–​23
emergence. See deliberate governance
Enron, 62–​63, 194 definition of, 64–​66
entrenchment. See takeover Grossman, Sanford J., 184, 189, 203n.8
ex ante vs. ex post, 5–​7, 27, 55n.6, 69, 78–​ Gulati, Ranjay, 34, 95
84, 89, 161, 163–​65, 176–​77, 179–​81,
185–​86, 196, 206–​207, 209–​10 Hamel, Gary, 35n.1, 115–​16, 216
expansion of the organization Hansmann, Henry B., 132–​33, 135–​
growth dynamic, 187, 188–​89, 36, 138
189f, 208 Hart, Oliver D., 51, 155n.8, 184,
separation dynamic, 161, 188–​89, 189f, 189, 203n.8
190, 196, 207–​8 hazard, 32, 69, 85, 114, 122–​23, 134–​35,
externality, 5, 7, 56–​58 152–​53, 171, 173, 182, 185–​86, 189,
194, 199, 202, 204, 206–​8, 230
failure contractual, 15–​16, 78, 87–​88, 93–​94,
market, 78, 164, 182–​83 98, 130, 182
organizational, 83, 106, 207 expropriation, 144
Fama, Eugene F., 16–​18, 43, 53, 61, 143–​ probity, 155–​57
44, 148–​49, 168 separation, 195–​98
feasibility, 12–​13, 22, 34, 90, 92n.7, 96, 98, transactional, 4
118, 131, 135–​36, 147, 151, 164, 168, holdup problem, 122–​23, 134–​35, 172–​
194, 214, 224, 234 73, 195
286 Index

Hollywood studio system, 145–​47 Larcker, David F., 14–​15, 125, 203n.8
HP, 107–​18 legislature, 89, 90t, 91t, 95–​96, 156
hybrid governance, 5, 7 bicameral, 156
financing, 116 limited liability
franchising, 89, 89n.6, 90t, 91t company, 11, 16–​17, 19–​20, 26, 39n.3,
front/​back hybrid, 226–​27, 229, 229n.8 41–​43, 53–​54, 61–​62, 64, 94, 99, 102–​
leasing, 92 4, 120, 127–​28, 137–​39, 142, 163,
organization, 147 178, 185, 187–​88, 192, 195–​96, 199,
public-​private partnership, 138 213, 215
teaching, 152t, 154–​55 partnership, 168
principle of, 164, 170–​71, 173–​74, 177–​78
incentive, 41, 47, 49, 60, 84–​86, 90–​92, 91t,
93, 95, 122, 126, 134, 189, 192, 211–​ Mace, Myles, 149, 194, 198
12, 230–​31 Mahoney, Joseph T., 33, 89, 90t, 150–​51,
high-​powered, 5, 7–​10, 63, 119, 135, 203n.8, 204–​5
221t, 222–​23 main problem, 13–​15, 32–​33, 41n.4, 76,
intensity, 9–​10, 122, 134–​36, 211–​12, 94–​95, 128–​30, 136, 140, 158, 165–​
222, 225, 229, 231 66, 176–​78, 202–​5, 208–​9
low-​powered, 8–​9, 119, 134–​36, 221t, management
222, 229 definition of, 34–​37
initial public offering (IPO), 24–​25, 38, March, James G., 42–​43, 158
63–​64, 81, 163–​64, 187–​88, 190–​91, Masten, Scott E., 92–​93, 122
193, 194, 203–​4, 207 Meckling, William H., 14, 187, 196
innovation, 32–​33, 36, 47, 58, 106–​7, Microsoft, 111, 112n.5
179, 206–​8 Milgrom, Paul R., 6–​7, 43–​44
Gross Domestic Expenditure in R&D minority dividend, 170t, 174, 178
(GERD), 217 myopia. See efficiency
R&D, 106–​7, 179, 206, 217 definition of, 256–​57
R&D intensity, 218
R&D spending, 217–​18 Nasdaq (National Association of Securities
underinvestment in, 23–​24, 46–​47, Dealers Automated Quotations),
55n.6, 208, 229 190n.2, 193, 193n.4
institution (and institutionalization) Neste, 127–​28, 128n.11, 140–​41
definition of, 213–​20 net gains. See feasibility
pillar, institutional vs. contractual, 20–​
22, 78, 168–​76, 190, 193 oil industry, 37, 122–​23, 127–​28, 140–​41
Intel, 111 organization design
interdependence. See also dependency designer (of the organization), 1–​2, 11–​
organizational, 98 14, 18–​22, 25, 27–​30, 32–​34, 35n.1,
pooled, 97, 97f 39–​40, 41–​42, 45, 51–​57, 62–​63, 66–​
reciprocal, 97, 97f 70, 76–​80, 85–​88, 92, 94–​95, 98–​101,
sequential, 97, 97f 104, 105–​7, 110, 117–​21, 128–​30,
workflow, 98 134–​38, 141, 143, 145–​46, 148–​50,
154–​59, 161–​65, 168–​69, 173, 174,
Jensen, Michael C., 10–​11, 14, 16–​18, 43, 178–​80, 182–​86, 189, 196–​97, 197n.6,
53, 61, 101, 143–​44, 148–​49, 168, 187, 198–​200, 202–​3, 205–​10, 209n.10,
196, 199, 203n.8, 209n.10 211, 212, 214–​15, 214n.2, 216–​26,
Jobs, Steve, 190, 218 229–​31, 233–​35
Organisation for Economic Co-​operation
Ketokivi, Mikko, 12, 40–​41, 89, and Development (OECD), 65–​66,
90t, 150–​51 217, 217n.5
 Index  287

organizational form, 16–​17, 24–​25, 67–​68, vs. public ordering (or legal centralism),
136–​38, 141, 157–​58, 213 22, 57, 133–​34, 164–​65, 186
closed (or close), 137, 137n.1, 168, 180, 184 probity, 155–​57, 192–​93
cooperative, 19–​20, 133, 135–​38 profit, 8
corporation (see limited liability) as economic surplus, 10, 43
for-​profit as a postappropriation measure,
nonprofit, 136 17, 17n.3
open, 137n.1 for-​profit vs. nonprofit, 24–​27,
private, 137n.1 43, 132–​38
public, 137n.1 profit-​and-​loss (P&L) responsibility, 13,
public-​private partnership, 27, 67–​68, 47, 163, 222
131, 137–​39 profit center vs. cost center, 47,
organizational structure, 29, 30, 32, 35n.1, 222, 229n.8
93–​94, 137, 163, 187, 215, 215n.3, as residual, 17
220, 225–​26 seeking of, 10, 16
ad hoc, 220–​21, 221t vs. survival, 16
divisional, 214, 220, 221t, 222–​23 psychiatric care, 27, 152t
formalized vs. ad hoc, 221–​22 physical restraint of patients,
front/​back hybrid, 226–​27, 229, 229n.8 152t, 153–​54
functional, 34, 214, 220–​24, 221t
matrix, 163, 187, 188n.1, 220, 221t, 223 Qarnain, Salma, 147–​48
Orts, Eric W., 100, 104, 128, 200n.7
other-​regarding behavior, 100, 105–​6, rationality, 1, 8
196–​97, 235 bounded yet intended, 40–​41
oversight conscious foresight, 15, 163, 168, 186
definition of, 38–​51 impossibility of maximization and
ownership optimization, 209n.10
as rights to residual, 17, 44–​51 Real Madrid, 131–​34, 137
separation of ownership and remediableness. See feasibility
control, 188–​89 research and development (R&D). See
shareholders as owners of the limited innovation
liability company, 17, 43–​44 residual
as title, 45, 51 alienability of residual claims, 53–​54,
144, 163, 168, 170t, 176
Parmar, Bidhan, 106–​7, 129 claim and claimant, 24–​26, 42–​44,
participation 53–​54, 99, 103, 117, 119–​20, 124,
involuntary (as opposed to voluntary), 129–​30, 132–​38, 140–​45, 147–​49,
27, 56–​58, 172, 181 157–​58, 163, 176, 183, 186, 195–​96,
voluntary (as opposed to involuntary), 201–​2, 208–​10
6, 11, 33, 66, 129, 151t, 199 interest, 17–​18, 26, 103, 107, 119–​20,
Penrose, Edith T., 71n.1, 188n.1 136, 144, 147, 177–​78, 187–​88, 195,
Prahalad, C. K., 115–​16 197n.6, 201t, 202–​3, 205–​7, 208–​10
preferences rights of control, 38, 42, 43–​51, 73, 138,
commensurate vs. incommensurate, 140, 140n.3, 152–​53, 164–​65, 167,
131, 151–​52, 151t, 152t, 153–​55 176, 180, 184–​86, 199
prison, 153, 155n.8 risk, 26, 99, 104, 108, 118, 122, 136, 139,
inmate discretion, 152–​53 142–​49, 195
privatization, 155n.8, 158 risk, 51–​58
private ordering, 19–​22, 25, 27, 42, 57, 58, arising from participating in the
65–​66, 78, 120–​21, 128, 168–​69, 178–​ organization, 26, 51–​52, 118–​19,
79, 182–​84, 196, 200n.7, 230–​31 129, 143
288 Index

risk (cont.) dedicated assets, 121

and specificity, 26, 51–​58, 65, 99, 119–​ employees and other stakeholders
23, 139 committing to, 50, 55, 99, 107, 113,
and switching cost, 79, 88, 110–​14, 122, 118–​20, 123, 129, 134–​35, 139, 140,
145–​47, 206 145, 187–​88, 197n.6, 205–​10
and vulnerability, 28, 51–​55, 99–​100, general-​purpose vs. special-​purpose
103, 105, 114, 118–​19, 123, 129–​30, technology, 37, 56, 123–​24,
143, 161, 202 167, 167n.1
Roberts, John, 6–​7, 43–​44 human capital, 88, 114–​16, 121, 123–​27,
130, 146, 157, 206
safeguard, 15–​16, 25–​27, 40, 60–​61, 65, physical asset, 56, 88, 115–​16, 121–​23,
67, 69, 80, 82, 86–​88, 92–​93, 95, 100, 168, 169
102–​3, 112, 112n.5, 115–​24, 117n.7, redeployability, 54–​55, 119–​20, 168, 219
130, 143–​47, 163–​64, 166, 172–​73, and risk, 55, 55n.6
179, 181, 182–​85, 192, 197–​98, 202–​ site, 88, 99, 122–​23
3, 206 vs. specialization, 115–​16
self-​interest, 1, 4, 99–​100, 134–​35 stakeholder, 17–​19, 161, 187–​88, 206–​7
individual vs. organizational goals, 201t analysis, 25–​26, 67, 99–​100, 107–​18
informed vs. hardball advocacy, 100 board member as, 196–​97, 204
and morality, 58 vs. constituency, 18, 53–​54, 101–​7
vs. opportunism, 1 customer as a, 112–​14, 146–​47
Selznick, Philip, 28, 213–​14, 214n.2, definition of, 19, 106–​7
215n.3, 230 dilemma, 101
Sergeeva, Anastasia, 135–​36, 142–​43 employee as a, 114–​16
shareholders’ agreement, 20, 27, 53–​54, environment as a, 56–​57
168–​69, 171, 173–​75, 181, 192 financier as a, 116–​18
and binding arbitration, 175, 183–​ in nonprofit organizations, 134–​
84, 216 35, 143–​47
and contractibility, 180 participation vs. representation, 59,
contributions clauses in, 170t, 124–​26, 195
174, 177–​78 prioritization, 19, 104–​6
drag-​along clauses in, 170t, 172, 195 and residual risk, 26, 51–​52, 65
and the duty of loyalty, 173–​74 shareholder as a, 102–​4, 169, 205–​6
and limited liability, 173–​74 supplier as a, 141
non-​compete clauses in, 170t stock exchange, 19–​20, 64, 126, 137, 190,
non-​disclosure clauses in, 170t 192–​93, 213–​14
tag-​along clauses in, 170t, 172, 195 Stout, Lynn A., 22, 59, 100, 186, 196–​97,
Silverman, Brian S., 32–​33, 77n.2 197n.6, 235
size Strudler, Alan, 100, 104, 128
scale of the organization, 28, 165–​ Sundaramurthy, Chamu, 203n.8, 204, 205
67, 179, 187–​91, 193, 198, 220, Supreme Court of the United States, 127,
222, 226–​27 127n.10, 128, 156, 192–​93
scope of the organization, 28, 97f, 98, switching cost. See risk
98n.8, 166–​67, 179, 187–​88, 190–​91,
193, 198–​99, 220, 222, 226–​27 takeover, 28, 161
slack, 42–​43 antitakeover provision, 184, 185,
Snyder, Edward A., 92–​93, 122 203n.8, 204–​7, 210
specificity, 4, 7, 55, 85–​88, 119, 121–​23, bargaining power in the event of, 199,
145–​46, 167 201, 201t, 203
 Index  289

entrenchment, 181, 185, 201t, 201, uncertainty, 54–​56, 78, 82, 85, 87–​88, 91t,
204, 224 95, 96, 98, 124, 179–​80, 184, 206
friendly vs. hostile, 199–​200, behavioral, 86–​87
200n.7, 201 demand, 87
market for corporate control, 198–​99 technological, 86, 87n.5
poison pill, 205 University of Illinois, 131–​33
protection against, 202
staggered board of directors, 203–​4 value. See also appropriation
wanted vs. unwanted, 171, 200–​1, creation, 1, 15–​19, 33–​35, 37, 42, 47, 51–​
201t, 202 52, 91t, 100, 108, 114–​15, 128–​129,
Tallarita, Roberto, 58, 101, 128, 200n.7 199, 206, 208, 222–​28
Tayan, Brian, 14–​15, 125, 203n.8 instrumental vs. intrinsic, 151, 152t,
Tesla Motors, 14, 38–​39, 190–​91, 191f, 194, 153, 156–​57, 214
194n.5, 200, 207–​8 viability, 15–​17, 19, 62f, 64, 67, 101–​2,
Musk, Elon, 14, 38–​39, 190, 194, 194n.5 106–​7, 148, 173, 206, 208, 229–​30
Thompson, James D., 71n.1, 97 Volkswagen Group, 93–​94, 223, 227–​28
transaction cost, 5–​6, 69, 76–​77, 91t, 136,
150, 219–​20 wager
determinants of, 84–​88 on organizational outcomes, 100, 103–​
ex ante vs. ex post, 5–​6, 78–​84, 182 4, 106–​8, 113, 116–​18, 147–​48, 176
instrumentality of, 151–​52 waste, elimination of. See efficiency
in inter-​vs. intraorganizational Williamson, Oliver E., 3, 11–​12, 13, 16n.2,
transactions, 92 20–​22, 77n.2, 82, 85, 87–​89, 89n.6,
Transaction Cost Economics 93–​94, 118, 120, 123, 131, 135, 148,
(TCE), 77n.2 155–​56, 166, 177–​78, 211, 219, 226,
transfer price, 47, 222, 229–​230 228–​29, 233, 235